postgresql/src/backend/utils/adt/ruleutils.c

1878 lines
45 KiB
C
Raw Normal View History

This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/**********************************************************************
* get_ruledef.c - Function to get a rules definition text
* out of it's tuple
*
* IDENTIFICATION
* $Header: /cvsroot/pgsql/src/backend/utils/adt/ruleutils.c,v 1.30 1999/11/07 23:08:24 momjian Exp $
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
*
* This software is copyrighted by Jan Wieck - Hamburg.
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
*
* The author hereby grants permission to use, copy, modify,
* distribute, and license this software and its documentation
* for any purpose, provided that existing copyright notices are
* retained in all copies and that this notice is included
* verbatim in any distributions. No written agreement, license,
* or royalty fee is required for any of the authorized uses.
* Modifications to this software may be copyrighted by their
* author and need not follow the licensing terms described
* here, provided that the new terms are clearly indicated on
* the first page of each file where they apply.
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
*
* IN NO EVENT SHALL THE AUTHOR OR DISTRIBUTORS BE LIABLE TO ANY
* PARTY FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR
* CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OF THIS
* SOFTWARE, ITS DOCUMENTATION, OR ANY DERIVATIVES THEREOF, EVEN
* IF THE AUTHOR HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH
* DAMAGE.
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
*
* THE AUTHOR AND DISTRIBUTORS SPECIFICALLY DISCLAIM ANY
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
* PURPOSE, AND NON-INFRINGEMENT. THIS SOFTWARE IS PROVIDED ON
* AN "AS IS" BASIS, AND THE AUTHOR AND DISTRIBUTORS HAVE NO
* OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES,
* ENHANCEMENTS, OR MODIFICATIONS.
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
*
**********************************************************************/
#include <unistd.h>
#include <fcntl.h>
#include "postgres.h"
#include "catalog/pg_index.h"
#include "catalog/pg_operator.h"
#include "catalog/pg_shadow.h"
#include "catalog/pg_type.h"
#include "executor/spi.h"
#include "lib/stringinfo.h"
#include "optimizer/clauses.h"
#include "optimizer/tlist.h"
#include "parser/parsetree.h"
#include "utils/builtins.h"
#include "utils/lsyscache.h"
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* Local data types
* ----------
*/
typedef struct
1999-05-25 18:15:34 +02:00
{
StringInfo buf; /* output buffer to append to */
List *rangetables; /* List of List of RangeTblEntry */
bool varprefix; /* TRUE to print prefixes on Vars */
} deparse_context;
typedef struct {
Index rt_index;
int levelsup;
} check_if_rte_used_context;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* Global data
* ----------
*/
static char *rulename = NULL;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
static void *plan_getrule = NULL;
static char *query_getrule = "SELECT * FROM pg_rewrite WHERE rulename = $1";
static void *plan_getview = NULL;
static char *query_getview = "SELECT * FROM pg_rewrite WHERE rulename = $1 or rulename = $2";
static void *plan_getam = NULL;
static char *query_getam = "SELECT * FROM pg_am WHERE oid = $1";
static void *plan_getopclass = NULL;
static char *query_getopclass = "SELECT * FROM pg_opclass WHERE oid = $1";
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* Local functions
*
* Most of these functions used to use fixed-size buffers to build their
* results. Now, they take an (already initialized) StringInfo object
* as a parameter, and append their text output to its contents.
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
* ----------
*/
static void make_ruledef(StringInfo buf, HeapTuple ruletup, TupleDesc rulettc);
static void make_viewdef(StringInfo buf, HeapTuple ruletup, TupleDesc rulettc);
static void get_query_def(Query *query, StringInfo buf, List *parentrtables);
static void get_select_query_def(Query *query, deparse_context *context);
static void get_insert_query_def(Query *query, deparse_context *context);
static void get_update_query_def(Query *query, deparse_context *context);
static void get_delete_query_def(Query *query, deparse_context *context);
static RangeTblEntry *get_rte_for_var(Var *var, deparse_context *context);
static void get_rule_expr(Node *node, deparse_context *context);
static void get_func_expr(Expr *expr, deparse_context *context);
static void get_tle_expr(TargetEntry *tle, deparse_context *context);
static void get_const_expr(Const *constval, deparse_context *context);
static void get_sublink_expr(Node *node, deparse_context *context);
static char *quote_identifier(char *ident);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
static char *get_relation_name(Oid relid);
static char *get_attribute_name(Oid relid, int2 attnum);
static bool check_if_rte_used(Node *node, Index rt_index, int levelsup);
static bool check_if_rte_used_walker(Node *node,
check_if_rte_used_context *context);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
#define inherit_marker(rte) ((rte)->inh ? "*" : "")
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* get_ruledef - Do it all and return a text
* that could be used as a statement
* to recreate the rule
* ----------
*/
text *
pg_get_ruledef(NameData *rname)
{
text *ruledef;
Datum args[1];
char nulls[2];
int spirc;
HeapTuple ruletup;
TupleDesc rulettc;
StringInfoData buf;
int len;
/* ----------
* We need the rules name somewhere deep down
* ----------
*/
rulename = pstrdup(NameStr(*rname));
/* ----------
* Connect to SPI manager
* ----------
*/
if (SPI_connect() != SPI_OK_CONNECT)
elog(ERROR, "get_ruledef: cannot connect to SPI manager");
/* ----------
* On the first call prepare the plan to lookup pg_proc.
* We read pg_proc over the SPI manager instead of using
* the syscache to be checked for read access on pg_proc.
* ----------
*/
if (plan_getrule == NULL)
{
Oid argtypes[1];
void *plan;
argtypes[0] = NAMEOID;
plan = SPI_prepare(query_getrule, 1, argtypes);
if (plan == NULL)
elog(ERROR, "SPI_prepare() failed for \"%s\"", query_getrule);
plan_getrule = SPI_saveplan(plan);
}
/* ----------
* Get the pg_rewrite tuple for this rule
* ----------
*/
args[0] = PointerGetDatum(rulename);
nulls[0] = (rulename == NULL) ? 'n' : ' ';
nulls[1] = '\0';
spirc = SPI_execp(plan_getrule, args, nulls, 1);
if (spirc != SPI_OK_SELECT)
elog(ERROR, "failed to get pg_rewrite tuple for %s", rulename);
if (SPI_processed != 1)
{
if (SPI_finish() != SPI_OK_FINISH)
elog(ERROR, "get_ruledef: SPI_finish() failed");
ruledef = SPI_palloc(VARHDRSZ + 1);
VARSIZE(ruledef) = VARHDRSZ + 1;
VARDATA(ruledef)[0] = '-';
return ruledef;
}
ruletup = SPI_tuptable->vals[0];
rulettc = SPI_tuptable->tupdesc;
/* ----------
* Get the rules definition and put it into executors memory
* ----------
*/
initStringInfo(&buf);
make_ruledef(&buf, ruletup, rulettc);
len = buf.len + VARHDRSZ;
ruledef = SPI_palloc(len);
VARSIZE(ruledef) = len;
memcpy(VARDATA(ruledef), buf.data, buf.len);
pfree(buf.data);
/* ----------
* Disconnect from SPI manager
* ----------
*/
if (SPI_finish() != SPI_OK_FINISH)
elog(ERROR, "get_ruledef: SPI_finish() failed");
/* ----------
* Easy - isn't it?
* ----------
*/
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
return ruledef;
}
/* ----------
* get_viewdef - Mainly the same thing, but we
* only return the SELECT part of a view
* ----------
*/
text *
pg_get_viewdef(NameData *rname)
{
text *ruledef;
Datum args[2];
char nulls[3];
int spirc;
HeapTuple ruletup;
TupleDesc rulettc;
StringInfoData buf;
int len;
char name1[NAMEDATALEN + 5];
char name2[NAMEDATALEN + 5];
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* We need the rules name somewhere deep down
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
* ----------
*/
rulename = pstrdup(NameStr(*rname));
/* ----------
* Connect to SPI manager
* ----------
*/
if (SPI_connect() != SPI_OK_CONNECT)
elog(ERROR, "get_viewdef: cannot connect to SPI manager");
/* ----------
* On the first call prepare the plan to lookup pg_proc.
* We read pg_proc over the SPI manager instead of using
* the syscache to be checked for read access on pg_proc.
* ----------
*/
if (plan_getview == NULL)
{
Oid argtypes[2];
void *plan;
argtypes[0] = NAMEOID;
argtypes[1] = NAMEOID;
plan = SPI_prepare(query_getview, 2, argtypes);
if (plan == NULL)
elog(ERROR, "SPI_prepare() failed for \"%s\"", query_getview);
plan_getview = SPI_saveplan(plan);
}
/* ----------
* Get the pg_rewrite tuple for this rule
* ----------
*/
sprintf(name1, "_RET%s", rulename);
sprintf(name2, "_ret%s", rulename);
args[0] = PointerGetDatum(name1);
args[1] = PointerGetDatum(name2);
nulls[0] = ' ';
nulls[1] = ' ';
nulls[2] = '\0';
spirc = SPI_execp(plan_getview, args, nulls, 1);
if (spirc != SPI_OK_SELECT)
elog(ERROR, "failed to get pg_rewrite tuple for view %s", rulename);
initStringInfo(&buf);
if (SPI_processed != 1)
appendStringInfo(&buf, "Not a view");
else
{
/* ----------
* Get the rules definition and put it into executors memory
* ----------
*/
ruletup = SPI_tuptable->vals[0];
rulettc = SPI_tuptable->tupdesc;
make_viewdef(&buf, ruletup, rulettc);
}
len = buf.len + VARHDRSZ;
ruledef = SPI_palloc(len);
VARSIZE(ruledef) = len;
memcpy(VARDATA(ruledef), buf.data, buf.len);
pfree(buf.data);
/* ----------
* Disconnect from SPI manager
* ----------
*/
if (SPI_finish() != SPI_OK_FINISH)
elog(ERROR, "get_viewdef: SPI_finish() failed");
/* ----------
* Easy - isn't it?
* ----------
*/
return ruledef;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_indexdef - Get the definition of an index
* ----------
*/
1999-05-25 18:15:34 +02:00
text *
pg_get_indexdef(Oid indexrelid)
{
1999-05-25 18:15:34 +02:00
text *indexdef;
HeapTuple ht_idx;
HeapTuple ht_idxrel;
HeapTuple ht_indrel;
HeapTuple spi_tup;
TupleDesc spi_ttc;
1999-05-25 18:15:34 +02:00
int spi_fno;
Form_pg_index idxrec;
Form_pg_class idxrelrec;
Form_pg_class indrelrec;
Datum spi_args[1];
char spi_nulls[2];
1999-05-25 18:15:34 +02:00
int spirc;
int len;
int keyno;
StringInfoData buf;
StringInfoData keybuf;
1999-05-25 18:15:34 +02:00
char *sep;
/* ----------
* Connect to SPI manager
* ----------
*/
if (SPI_connect() != SPI_OK_CONNECT)
elog(ERROR, "get_indexdef: cannot connect to SPI manager");
/* ----------
* On the first call prepare the plans to lookup pg_am
* and pg_opclass.
* ----------
*/
if (plan_getam == NULL)
{
Oid argtypes[1];
void *plan;
argtypes[0] = OIDOID;
plan = SPI_prepare(query_getam, 1, argtypes);
if (plan == NULL)
elog(ERROR, "SPI_prepare() failed for \"%s\"", query_getam);
plan_getam = SPI_saveplan(plan);
argtypes[0] = OIDOID;
plan = SPI_prepare(query_getopclass, 1, argtypes);
if (plan == NULL)
elog(ERROR, "SPI_prepare() failed for \"%s\"", query_getopclass);
plan_getopclass = SPI_saveplan(plan);
}
/* ----------
* Fetch the pg_index tuple by the Oid of the index
* ----------
*/
ht_idx = SearchSysCacheTuple(INDEXRELID,
1999-05-25 18:15:34 +02:00
ObjectIdGetDatum(indexrelid), 0, 0, 0);
if (!HeapTupleIsValid(ht_idx))
elog(ERROR, "syscache lookup for index %u failed", indexrelid);
1999-05-25 18:15:34 +02:00
idxrec = (Form_pg_index) GETSTRUCT(ht_idx);
/* ----------
* Fetch the pg_class tuple of the index relation
* ----------
*/
ht_idxrel = SearchSysCacheTuple(RELOID,
1999-05-25 18:15:34 +02:00
ObjectIdGetDatum(idxrec->indexrelid), 0, 0, 0);
if (!HeapTupleIsValid(ht_idxrel))
elog(ERROR, "syscache lookup for relid %u failed", idxrec->indexrelid);
1999-05-25 18:15:34 +02:00
idxrelrec = (Form_pg_class) GETSTRUCT(ht_idxrel);
/* ----------
* Fetch the pg_class tuple of the indexed relation
* ----------
*/
ht_indrel = SearchSysCacheTuple(RELOID,
1999-05-25 18:15:34 +02:00
ObjectIdGetDatum(idxrec->indrelid), 0, 0, 0);
if (!HeapTupleIsValid(ht_indrel))
elog(ERROR, "syscache lookup for relid %u failed", idxrec->indrelid);
1999-05-25 18:15:34 +02:00
indrelrec = (Form_pg_class) GETSTRUCT(ht_indrel);
/* ----------
* Get the am name for the index relation
* ----------
*/
spi_args[0] = ObjectIdGetDatum(idxrelrec->relam);
spi_nulls[0] = ' ';
spi_nulls[1] = '\0';
spirc = SPI_execp(plan_getam, spi_args, spi_nulls, 1);
if (spirc != SPI_OK_SELECT)
elog(ERROR, "failed to get pg_am tuple for index %s",
idxrelrec->relname);
if (SPI_processed != 1)
elog(ERROR, "failed to get pg_am tuple for index %s",
idxrelrec->relname);
spi_tup = SPI_tuptable->vals[0];
spi_ttc = SPI_tuptable->tupdesc;
spi_fno = SPI_fnumber(spi_ttc, "amname");
/* ----------
* Start the index definition
* ----------
*/
initStringInfo(&buf);
appendStringInfo(&buf, "CREATE %sINDEX %s ON %s USING %s (",
idxrec->indisunique ? "UNIQUE " : "",
quote_identifier(pstrdup(NameStr(idxrelrec->relname))),
quote_identifier(pstrdup(NameStr(indrelrec->relname))),
quote_identifier(SPI_getvalue(spi_tup, spi_ttc,
spi_fno)));
1999-05-25 18:15:34 +02:00
/* ----------
* Collect the indexed attributes
* ----------
*/
initStringInfo(&keybuf);
sep = "";
for (keyno = 0; keyno < INDEX_MAX_KEYS; keyno++)
{
if (idxrec->indkey[keyno] == InvalidAttrNumber)
break;
appendStringInfo(&keybuf, sep);
sep = ", ";
/* ----------
* Add the indexed field name
* ----------
*/
appendStringInfo(&keybuf, "%s",
quote_identifier(get_attribute_name(idxrec->indrelid,
idxrec->indkey[keyno])));
/* ----------
* If not a functional index, add the operator class name
* ----------
*/
if (idxrec->indproc == InvalidOid)
{
spi_args[0] = ObjectIdGetDatum(idxrec->indclass[keyno]);
spi_nulls[0] = ' ';
spi_nulls[1] = '\0';
spirc = SPI_execp(plan_getopclass, spi_args, spi_nulls, 1);
if (spirc != SPI_OK_SELECT)
elog(ERROR, "failed to get pg_opclass tuple %u", idxrec->indclass[keyno]);
if (SPI_processed != 1)
elog(ERROR, "failed to get pg_opclass tuple %u", idxrec->indclass[keyno]);
spi_tup = SPI_tuptable->vals[0];
spi_ttc = SPI_tuptable->tupdesc;
spi_fno = SPI_fnumber(spi_ttc, "opcname");
appendStringInfo(&keybuf, " %s",
quote_identifier(SPI_getvalue(spi_tup, spi_ttc,
spi_fno)));
}
}
/* ----------
* For functional index say 'func (attrs) opclass'
* ----------
*/
if (idxrec->indproc != InvalidOid)
{
HeapTuple proctup;
1999-05-25 18:15:34 +02:00
Form_pg_proc procStruct;
proctup = SearchSysCacheTuple(PROOID,
1999-05-25 18:15:34 +02:00
ObjectIdGetDatum(idxrec->indproc), 0, 0, 0);
if (!HeapTupleIsValid(proctup))
elog(ERROR, "cache lookup for proc %u failed", idxrec->indproc);
procStruct = (Form_pg_proc) GETSTRUCT(proctup);
appendStringInfo(&buf, "%s(%s) ",
quote_identifier(pstrdup(NameStr(procStruct->proname))),
keybuf.data);
spi_args[0] = ObjectIdGetDatum(idxrec->indclass[0]);
spi_nulls[0] = ' ';
spi_nulls[1] = '\0';
spirc = SPI_execp(plan_getopclass, spi_args, spi_nulls, 1);
if (spirc != SPI_OK_SELECT)
elog(ERROR, "failed to get pg_opclass tuple %u", idxrec->indclass[0]);
if (SPI_processed != 1)
elog(ERROR, "failed to get pg_opclass tuple %u", idxrec->indclass[0]);
spi_tup = SPI_tuptable->vals[0];
spi_ttc = SPI_tuptable->tupdesc;
spi_fno = SPI_fnumber(spi_ttc, "opcname");
appendStringInfo(&buf, "%s",
quote_identifier(SPI_getvalue(spi_tup, spi_ttc,
spi_fno)));
}
else
1999-05-25 18:15:34 +02:00
/* ----------
* For the others say 'attr opclass [, ...]'
* ----------
*/
appendStringInfo(&buf, "%s", keybuf.data);
/* ----------
* Finish
* ----------
*/
appendStringInfo(&buf, ")");
/* ----------
* Create the result in upper executor memory
* ----------
*/
len = buf.len + VARHDRSZ;
indexdef = SPI_palloc(len);
VARSIZE(indexdef) = len;
memcpy(VARDATA(indexdef), buf.data, buf.len);
pfree(buf.data);
pfree(keybuf.data);
/* ----------
* Disconnect from SPI manager
* ----------
*/
if (SPI_finish() != SPI_OK_FINISH)
elog(ERROR, "get_viewdef: SPI_finish() failed");
return indexdef;
}
/* ----------
* get_userbyid - Get a user name by usesysid and
* fallback to 'unknown (UID=n)'
* ----------
*/
NameData *
pg_get_userbyid(int32 uid)
{
HeapTuple usertup;
1999-05-25 18:15:34 +02:00
Form_pg_shadow user_rec;
NameData *result;
/* ----------
* Allocate space for the result
* ----------
*/
result = (NameData *) palloc(NAMEDATALEN);
memset(NameStr(*result), 0, NAMEDATALEN);
/* ----------
* Get the pg_shadow entry and print the result
* ----------
*/
usertup = SearchSysCacheTuple(USESYSID,
1999-05-25 18:15:34 +02:00
ObjectIdGetDatum(uid), 0, 0, 0);
if (HeapTupleIsValid(usertup))
{
1999-05-25 18:15:34 +02:00
user_rec = (Form_pg_shadow) GETSTRUCT(usertup);
StrNCpy(NameStr(*result), NameStr(user_rec->usename), NAMEDATALEN);
}
else
1999-05-25 18:15:34 +02:00
sprintf((char *) result, "unknown (UID=%d)", uid);
return result;
}
/* ----------
* deparse_expression - General utility for deparsing expressions
*
* expr is the node tree to be deparsed. It must be a transformed expression
* tree (ie, not the raw output of gram.y).
*
* rangetables is a List of Lists of RangeTblEntry nodes: first sublist is for
* varlevelsup = 0, next for varlevelsup = 1, etc. In each sublist the first
* item is for varno = 1, next varno = 2, etc. (Each sublist has the same
* format as the rtable list of a parsetree or query.)
*
* forceprefix is TRUE to force all Vars to be prefixed with their table names.
* Otherwise, a prefix is printed only if there's more than one table involved
* (and someday the code might try to print one only if there's ambiguity).
*
* The result is a palloc'd string.
* ----------
*/
char *
deparse_expression(Node *expr, List *rangetables, bool forceprefix)
{
StringInfoData buf;
deparse_context context;
initStringInfo(&buf);
context.buf = &buf;
context.rangetables = rangetables;
context.varprefix = (forceprefix ||
length(rangetables) != 1 ||
length((List *) lfirst(rangetables)) != 1);
rulename = ""; /* in case of errors */
get_rule_expr(expr, &context);
return buf.data;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* make_ruledef - reconstruct the CREATE RULE command
* for a given pg_rewrite tuple
* ----------
*/
static void
make_ruledef(StringInfo buf, HeapTuple ruletup, TupleDesc rulettc)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
char ev_type;
Oid ev_class;
int2 ev_attr;
bool is_instead;
char *ev_qual;
char *ev_action;
List *actions = NIL;
int fno;
bool isnull;
/* ----------
* Get the attribute values from the rules tuple
* ----------
*/
fno = SPI_fnumber(rulettc, "ev_type");
ev_type = (char) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "ev_class");
ev_class = (Oid) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "ev_attr");
ev_attr = (int2) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "is_instead");
is_instead = (bool) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "ev_qual");
ev_qual = SPI_getvalue(ruletup, rulettc, fno);
fno = SPI_fnumber(rulettc, "ev_action");
ev_action = SPI_getvalue(ruletup, rulettc, fno);
if (ev_action != NULL)
actions = (List *) stringToNode(ev_action);
/* ----------
* Build the rules definition text
* ----------
*/
appendStringInfo(buf, "CREATE RULE %s AS ON ",
quote_identifier(rulename));
/* The event the rule is fired for */
switch (ev_type)
{
case '1':
appendStringInfo(buf, "SELECT");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
break;
case '2':
appendStringInfo(buf, "UPDATE");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
break;
case '3':
appendStringInfo(buf, "INSERT");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
break;
case '4':
appendStringInfo(buf, "DELETE");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
break;
default:
elog(ERROR, "get_ruledef: rule %s has unsupported event type %d",
rulename, ev_type);
break;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* The relation the rule is fired on */
appendStringInfo(buf, " TO %s",
quote_identifier(get_relation_name(ev_class)));
if (ev_attr > 0)
appendStringInfo(buf, ".%s",
quote_identifier(get_attribute_name(ev_class,
ev_attr)));
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* If the rule has an event qualification, add it */
if (ev_qual == NULL)
ev_qual = "";
if (strlen(ev_qual) > 0 && strcmp(ev_qual, "<>") != 0)
{
Node *qual;
Query *query;
deparse_context context;
appendStringInfo(buf, " WHERE ");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
qual = stringToNode(ev_qual);
query = (Query *) lfirst(actions);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
context.buf = buf;
context.rangetables = lcons(query->rtable, NIL);
context.varprefix = (length(query->rtable) != 1);
get_rule_expr(qual, &context);
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
appendStringInfo(buf, " DO ");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* The INSTEAD keyword (if so) */
if (is_instead)
appendStringInfo(buf, "INSTEAD ");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* Finally the rules actions */
if (length(actions) > 1)
{
List *action;
Query *query;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
appendStringInfo(buf, "(");
foreach(action, actions)
{
query = (Query *) lfirst(action);
get_query_def(query, buf, NIL);
appendStringInfo(buf, "; ");
}
appendStringInfo(buf, ");");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
else
{
if (length(actions) == 0)
{
appendStringInfo(buf, "NOTHING;");
}
else
{
Query *query;
query = (Query *) lfirst(actions);
get_query_def(query, buf, NIL);
appendStringInfo(buf, ";");
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
}
/* ----------
* make_viewdef - reconstruct the SELECT part of a
* view rewrite rule
* ----------
*/
static void
make_viewdef(StringInfo buf, HeapTuple ruletup, TupleDesc rulettc)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
Query *query;
char ev_type;
Oid ev_class;
int2 ev_attr;
bool is_instead;
char *ev_qual;
char *ev_action;
List *actions = NIL;
int fno;
bool isnull;
/* ----------
* Get the attribute values from the rules tuple
* ----------
*/
fno = SPI_fnumber(rulettc, "ev_type");
ev_type = (char) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "ev_class");
ev_class = (Oid) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "ev_attr");
ev_attr = (int2) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "is_instead");
is_instead = (bool) SPI_getbinval(ruletup, rulettc, fno, &isnull);
fno = SPI_fnumber(rulettc, "ev_qual");
ev_qual = SPI_getvalue(ruletup, rulettc, fno);
fno = SPI_fnumber(rulettc, "ev_action");
ev_action = SPI_getvalue(ruletup, rulettc, fno);
if (ev_action != NULL)
actions = (List *) stringToNode(ev_action);
if (length(actions) != 1)
{
appendStringInfo(buf, "Not a view");
return;
}
query = (Query *) lfirst(actions);
if (ev_type != '1' || ev_attr >= 0 || !is_instead || strcmp(ev_qual, "<>"))
{
appendStringInfo(buf, "Not a view");
return;
}
get_query_def(query, buf, NIL);
appendStringInfo(buf, ";");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_query_def - Parse back one action from
* the parsetree in the actions
* list
* ----------
*/
static void
get_query_def(Query *query, StringInfo buf, List *parentrtables)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
deparse_context context;
context.buf = buf;
context.rangetables = lcons(query->rtable, parentrtables);
context.varprefix = (parentrtables != NIL ||
length(query->rtable) != 1);
switch (query->commandType)
{
1999-05-25 18:15:34 +02:00
case CMD_SELECT:
get_select_query_def(query, &context);
break;
case CMD_UPDATE:
get_update_query_def(query, &context);
break;
case CMD_INSERT:
get_insert_query_def(query, &context);
break;
case CMD_DELETE:
get_delete_query_def(query, &context);
break;
case CMD_NOTHING:
appendStringInfo(buf, "NOTHING");
break;
default:
elog(ERROR, "get_ruledef of %s: query command type %d not implemented yet",
rulename, query->commandType);
break;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_select_query_def - Parse back a SELECT parsetree
* ----------
*/
static void
get_select_query_def(Query *query, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
char *sep;
TargetEntry *tle;
RangeTblEntry *rte;
bool *rt_used;
int rt_length;
int rt_numused = 0;
bool rt_constonly = TRUE;
int i;
List *l;
/* ----------
* First we need to know which and how many of the
* range table entries in the query are used in the target list
* or queries qualification
* ----------
*/
rt_length = length(query->rtable);
rt_used = palloc(sizeof(bool) * rt_length);
for (i = 0; i < rt_length; i++)
{
if (check_if_rte_used((Node *) (query->targetList), i + 1, 0) ||
check_if_rte_used(query->qual, i + 1, 0) ||
check_if_rte_used(query->havingQual, i + 1, 0))
{
rt_used[i] = TRUE;
rt_numused++;
}
else
rt_used[i] = FALSE;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* Now check if any of the used rangetable entries is different
* from *NEW* and *CURRENT*. If so we must provide the FROM clause
* later.
* ----------
*/
i = 0;
foreach(l, query->rtable)
{
if (!rt_used[i++])
continue;
rte = (RangeTblEntry *) lfirst(l);
if (!strcmp(rte->refname, "*NEW*"))
continue;
if (!strcmp(rte->refname, "*CURRENT*"))
continue;
rt_constonly = FALSE;
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* Build up the query string - first we say SELECT
* ----------
*/
appendStringInfo(buf, "SELECT");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* Then we tell what to select (the targetlist) */
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
sep = " ";
foreach(l, query->targetList)
{
bool tell_as = FALSE;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
tle = (TargetEntry *) lfirst(l);
appendStringInfo(buf, sep);
sep = ", ";
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
get_tle_expr(tle, context);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* Check if we must say AS ... */
if (! IsA(tle->expr, Var))
tell_as = strcmp(tle->resdom->resname, "?column?");
else
{
Var *var = (Var *) (tle->expr);
char *attname;
rte = get_rte_for_var(var, context);
attname = get_attribute_name(rte->relid, var->varattno);
if (strcmp(attname, tle->resdom->resname))
tell_as = TRUE;
}
/* and do if so */
if (tell_as)
appendStringInfo(buf, " AS %s",
quote_identifier(tle->resdom->resname));
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* If we need other tables than *NEW* or *CURRENT* add the FROM clause */
if (!rt_constonly && rt_numused > 0)
{
sep = " FROM ";
i = 0;
foreach(l, query->rtable)
{
if (rt_used[i++])
{
rte = (RangeTblEntry *) lfirst(l);
if (!strcmp(rte->refname, "*NEW*"))
continue;
if (!strcmp(rte->refname, "*CURRENT*"))
continue;
appendStringInfo(buf, sep);
sep = ", ";
appendStringInfo(buf, "%s%s",
quote_identifier(rte->relname),
inherit_marker(rte));
if (strcmp(rte->relname, rte->refname) != 0)
appendStringInfo(buf, " %s",
quote_identifier(rte->refname));
}
}
}
/* Add the WHERE clause if given */
if (query->qual != NULL)
{
appendStringInfo(buf, " WHERE ");
get_rule_expr(query->qual, context);
}
/* Add the GROUP BY CLAUSE */
if (query->groupClause != NULL)
{
appendStringInfo(buf, " GROUP BY ");
sep = "";
foreach(l, query->groupClause)
{
GroupClause *grp = (GroupClause *) lfirst(l);
Node *groupexpr;
groupexpr = get_sortgroupclause_expr(grp,
query->targetList);
appendStringInfo(buf, sep);
get_rule_expr(groupexpr, context);
sep = ", ";
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
}
/* ----------
* get_insert_query_def - Parse back an INSERT parsetree
* ----------
*/
static void
get_insert_query_def(Query *query, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
char *sep;
TargetEntry *tle;
RangeTblEntry *rte;
bool *rt_used;
int rt_length;
int rt_numused = 0;
bool rt_constonly = TRUE;
int i;
List *l;
/* ----------
* We need to know if other tables than *NEW* or *CURRENT*
* are used in the query. If not, it's an INSERT ... VALUES,
* otherwise an INSERT ... SELECT.
* ----------
*/
rt_length = length(query->rtable);
rt_used = palloc(sizeof(bool) * rt_length);
for (i = 0; i < rt_length; i++)
{
if (check_if_rte_used((Node *) (query->targetList), i + 1, 0) ||
check_if_rte_used(query->qual, i + 1, 0) ||
check_if_rte_used(query->havingQual, i + 1, 0))
{
rt_used[i] = TRUE;
rt_numused++;
}
else
rt_used[i] = FALSE;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
i = 0;
foreach(l, query->rtable)
{
if (!rt_used[i++])
continue;
rte = (RangeTblEntry *) lfirst(l);
if (!strcmp(rte->refname, "*NEW*"))
continue;
if (!strcmp(rte->refname, "*CURRENT*"))
continue;
rt_constonly = FALSE;
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* Start the query with INSERT INTO relname
* ----------
*/
rte = rt_fetch(query->resultRelation, query->rtable);
appendStringInfo(buf, "INSERT INTO %s",
quote_identifier(rte->relname));
/* Add the target list */
sep = " (";
foreach(l, query->targetList)
{
tle = (TargetEntry *) lfirst(l);
appendStringInfo(buf, sep);
sep = ", ";
appendStringInfo(buf, "%s", quote_identifier(tle->resdom->resname));
}
appendStringInfo(buf, ") ");
/* Add the VALUES or the SELECT */
if (rt_constonly && query->qual == NULL)
{
appendStringInfo(buf, "VALUES (");
sep = "";
foreach(l, query->targetList)
{
tle = (TargetEntry *) lfirst(l);
appendStringInfo(buf, sep);
sep = ", ";
get_tle_expr(tle, context);
}
appendStringInfo(buf, ")");
}
else
get_select_query_def(query, context);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_update_query_def - Parse back an UPDATE parsetree
* ----------
*/
static void
get_update_query_def(Query *query, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
char *sep;
TargetEntry *tle;
RangeTblEntry *rte;
List *l;
/* ----------
* Start the query with UPDATE relname SET
* ----------
*/
rte = rt_fetch(query->resultRelation, query->rtable);
appendStringInfo(buf, "UPDATE %s%s SET ",
quote_identifier(rte->relname),
inherit_marker(rte));
/* Add the comma separated list of 'attname = value' */
sep = "";
foreach(l, query->targetList)
{
tle = (TargetEntry *) lfirst(l);
appendStringInfo(buf, sep);
sep = ", ";
appendStringInfo(buf, "%s = ",
quote_identifier(tle->resdom->resname));
get_tle_expr(tle, context);
}
/* Finally add a WHERE clause if given */
if (query->qual != NULL)
{
appendStringInfo(buf, " WHERE ");
get_rule_expr(query->qual, context);
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_delete_query_def - Parse back a DELETE parsetree
* ----------
*/
static void
get_delete_query_def(Query *query, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
RangeTblEntry *rte;
/* ----------
* Start the query with DELETE FROM relname
* ----------
*/
rte = rt_fetch(query->resultRelation, query->rtable);
appendStringInfo(buf, "DELETE FROM %s%s",
quote_identifier(rte->relname),
inherit_marker(rte));
/* Add a WHERE clause if given */
if (query->qual != NULL)
{
appendStringInfo(buf, " WHERE ");
get_rule_expr(query->qual, context);
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/*
* Find the RTE referenced by a (possibly nonlocal) Var.
*/
static RangeTblEntry *
get_rte_for_var(Var *var, deparse_context *context)
{
List *rtlist = context->rangetables;
int sup = var->varlevelsup;
while (sup-- > 0)
rtlist = lnext(rtlist);
return rt_fetch(var->varno, (List *) lfirst(rtlist));
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* get_rule_expr - Parse back an expression
* ----------
*/
static void
get_rule_expr(Node *node, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
if (node == NULL)
return;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* Each level of get_rule_expr must emit an indivisible term
* (parenthesized if necessary) to ensure result is reparsed into
* the same expression tree.
*
* There might be some work left here to support additional node types.
* Can we ever see Param nodes here?
* ----------
*/
switch (nodeTag(node))
{
case T_Const:
get_const_expr((Const *) node, context);
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
case T_Var:
{
Var *var = (Var *) node;
RangeTblEntry *rte = get_rte_for_var(var, context);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
if (context->varprefix)
{
if (!strcmp(rte->refname, "*NEW*"))
appendStringInfo(buf, "new.");
else if (!strcmp(rte->refname, "*CURRENT*"))
appendStringInfo(buf, "old.");
else
appendStringInfo(buf, "%s.",
quote_identifier(rte->refname));
}
appendStringInfo(buf, "%s",
quote_identifier(get_attribute_name(rte->relid,
var->varattno)));
}
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
case T_Expr:
{
Expr *expr = (Expr *) node;
List *args = expr->args;
/* ----------
* Expr nodes have to be handled a bit detailed
* ----------
*/
switch (expr->opType)
{
case OP_EXPR:
appendStringInfo(buf, "(");
if (length(args) == 2)
{
/* binary operator */
get_rule_expr((Node *) lfirst(args), context);
appendStringInfo(buf, " %s ",
get_opname(((Oper *) expr->oper)->opno));
get_rule_expr((Node *) lsecond(args), context);
}
else
{
/* unary operator --- but which side? */
Oid opno = ((Oper *) expr->oper)->opno;
HeapTuple tp;
Form_pg_operator optup;
tp = SearchSysCacheTuple(OPROID,
ObjectIdGetDatum(opno),
0, 0, 0);
Assert(HeapTupleIsValid(tp));
optup = (Form_pg_operator) GETSTRUCT(tp);
switch (optup->oprkind)
{
case 'l':
appendStringInfo(buf, "%s ",
get_opname(opno));
get_rule_expr((Node *) lfirst(args),
context);
break;
case 'r':
get_rule_expr((Node *) lfirst(args),
context);
appendStringInfo(buf, " %s",
get_opname(opno));
break;
default:
elog(ERROR, "get_rule_expr: bogus oprkind");
}
}
appendStringInfo(buf, ")");
break;
case OR_EXPR:
appendStringInfo(buf, "(");
get_rule_expr((Node *) lfirst(args), context);
while ((args = lnext(args)) != NIL)
{
appendStringInfo(buf, " OR ");
get_rule_expr((Node *) lfirst(args), context);
}
appendStringInfo(buf, ")");
break;
case AND_EXPR:
appendStringInfo(buf, "(");
get_rule_expr((Node *) lfirst(args), context);
while ((args = lnext(args)) != NIL)
{
appendStringInfo(buf, " AND ");
get_rule_expr((Node *) lfirst(args), context);
}
appendStringInfo(buf, ")");
break;
case NOT_EXPR:
appendStringInfo(buf, "(NOT ");
get_rule_expr((Node *) lfirst(args), context);
appendStringInfo(buf, ")");
break;
case FUNC_EXPR:
get_func_expr((Expr *) node, context);
break;
default:
elog(ERROR, "get_rule_expr: expr opType %d not supported",
expr->opType);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
}
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
case T_Aggref:
{
Aggref *aggref = (Aggref *) node;
appendStringInfo(buf, "%s(",
quote_identifier(aggref->aggname));
get_rule_expr(aggref->target, context);
appendStringInfo(buf, ")");
}
break;
case T_Iter:
get_rule_expr(((Iter *) node)->iterexpr, context);
break;
case T_ArrayRef:
{
ArrayRef *aref = (ArrayRef *) node;
List *lowlist;
List *uplist;
get_rule_expr(aref->refexpr, context);
lowlist = aref->reflowerindexpr;
foreach(uplist, aref->refupperindexpr)
{
appendStringInfo(buf, "[");
if (lowlist)
{
get_rule_expr((Node *) lfirst(lowlist), context);
appendStringInfo(buf, ":");
lowlist = lnext(lowlist);
}
get_rule_expr((Node *) lfirst(uplist), context);
appendStringInfo(buf, "]");
}
/* XXX need to do anything with refassgnexpr? */
}
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
case T_CaseExpr:
{
CaseExpr *caseexpr = (CaseExpr *) node;
List *temp;
appendStringInfo(buf, "CASE");
foreach(temp, caseexpr->args)
{
CaseWhen *when = (CaseWhen *) lfirst(temp);
appendStringInfo(buf, " WHEN ");
get_rule_expr(when->expr, context);
appendStringInfo(buf, " THEN ");
get_rule_expr(when->result, context);
}
appendStringInfo(buf, " ELSE ");
get_rule_expr(caseexpr->defresult, context);
appendStringInfo(buf, " END");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
case T_SubLink:
get_sublink_expr(node, context);
break;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
default:
printf("\n%s\n", nodeToString(node));
elog(ERROR, "get_ruledef of %s: unknown node type %d in get_rule_expr()",
rulename, nodeTag(node));
break;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_func_expr - Parse back a Func node
* ----------
*/
static void
get_func_expr(Expr *expr, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
HeapTuple proctup;
Form_pg_proc procStruct;
List *l;
char *sep;
Func *func = (Func *) (expr->oper);
char *proname;
/* ----------
* Get the functions pg_proc tuple
* ----------
*/
proctup = SearchSysCacheTuple(PROOID,
ObjectIdGetDatum(func->funcid),
0, 0, 0);
if (!HeapTupleIsValid(proctup))
elog(ERROR, "cache lookup for proc %u failed", func->funcid);
procStruct = (Form_pg_proc) GETSTRUCT(proctup);
proname = pstrdup(NameStr(procStruct->proname));
/*
* nullvalue() and nonnullvalue() should get turned into special syntax
*/
if (procStruct->pronargs == 1 && procStruct->proargtypes[0] == InvalidOid)
{
if (!strcmp(proname, "nullvalue"))
{
appendStringInfo(buf, "(");
get_rule_expr((Node *) lfirst(expr->args), context);
appendStringInfo(buf, " ISNULL)");
return;
}
if (!strcmp(proname, "nonnullvalue"))
{
appendStringInfo(buf, "(");
get_rule_expr((Node *) lfirst(expr->args), context);
appendStringInfo(buf, " NOTNULL)");
return;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* Build a string of proname(args)
* ----------
*/
appendStringInfo(buf, "%s(", quote_identifier(proname));
sep = "";
foreach(l, expr->args)
{
appendStringInfo(buf, sep);
sep = ", ";
get_rule_expr((Node *) lfirst(l), context);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
appendStringInfo(buf, ")");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_tle_expr
*
* A target list expression is a bit different from a normal expression.
* If the target column has an atttypmod, the parser usually puts a
* padding-/cut-function call around the expression itself.
* We must get rid of it, otherwise dump/reload/dump... would blow up
* the expressions.
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
* ----------
*/
static void
get_tle_expr(TargetEntry *tle, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
Expr *expr = (Expr *) (tle->expr);
Func *func;
HeapTuple tup;
Form_pg_proc procStruct;
Form_pg_type typeStruct;
Const *second_arg;
/* ----------
* Check if the result has an atttypmod and if the
* expression in the targetlist entry is a function call
* ----------
*/
if (tle->resdom->restypmod < 0 ||
! IsA(expr, Expr) ||
expr->opType != FUNC_EXPR)
{
get_rule_expr(tle->expr, context);
return;
}
func = (Func *) (expr->oper);
/* ----------
* Get the functions pg_proc tuple
* ----------
*/
tup = SearchSysCacheTuple(PROOID,
ObjectIdGetDatum(func->funcid), 0, 0, 0);
if (!HeapTupleIsValid(tup))
elog(ERROR, "cache lookup for proc %u failed", func->funcid);
procStruct = (Form_pg_proc) GETSTRUCT(tup);
/* ----------
* It must be a function with two arguments where the first
* is of the same type as the return value and the second is
* an int4.
* ----------
*/
if (procStruct->pronargs != 2 ||
procStruct->prorettype != procStruct->proargtypes[0] ||
procStruct->proargtypes[1] != INT4OID)
{
get_rule_expr(tle->expr, context);
return;
}
/*
* Furthermore, the name of the function must be the same
* as the argument/result type name.
*/
tup = SearchSysCacheTuple(TYPOID,
ObjectIdGetDatum(procStruct->prorettype),
0, 0, 0);
if (!HeapTupleIsValid(tup))
elog(ERROR, "cache lookup for type %u failed",
procStruct->prorettype);
typeStruct = (Form_pg_type) GETSTRUCT(tup);
if (strncmp(NameStr(procStruct->proname),
NameStr(typeStruct->typname),
NAMEDATALEN) != 0)
{
get_rule_expr(tle->expr, context);
return;
}
/* ----------
* Finally (to be totally safe) the second argument must be a
* const and match the value in the results atttypmod.
* ----------
*/
second_arg = (Const *) lsecond(expr->args);
if (! IsA(second_arg, Const) ||
DatumGetInt32(second_arg->constvalue) != tle->resdom->restypmod)
{
get_rule_expr(tle->expr, context);
return;
}
/* ----------
* Whow - got it. Now get rid of the padding function
* ----------
*/
get_rule_expr((Node *) lfirst(expr->args), context);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_const_expr
*
* Make a string representation of a Const
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
* ----------
*/
static void
get_const_expr(Const *constval, deparse_context *context)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
StringInfo buf = context->buf;
HeapTuple typetup;
Form_pg_type typeStruct;
FmgrInfo finfo_output;
char *extval;
char *valptr;
bool isnull = FALSE;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
if (constval->constisnull)
{
appendStringInfo(buf, "NULL");
return;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
typetup = SearchSysCacheTuple(TYPOID,
ObjectIdGetDatum(constval->consttype),
0, 0, 0);
if (!HeapTupleIsValid(typetup))
elog(ERROR, "cache lookup of type %u failed", constval->consttype);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
typeStruct = (Form_pg_type) GETSTRUCT(typetup);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
fmgr_info(typeStruct->typoutput, &finfo_output);
extval = (char *) (*fmgr_faddr(&finfo_output)) (constval->constvalue,
&isnull, -1);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
switch (constval->consttype)
{
case INT2OID:
case INT4OID:
case OIDOID: /* int types */
case FLOAT4OID:
case FLOAT8OID: /* float types */
/* These types are printed without quotes */
appendStringInfo(buf, extval);
break;
default:
/*
* We must quote any funny characters in the constant's
* representation.
* XXX Any MULTIBYTE considerations here?
*/
appendStringInfoChar(buf, '\'');
for (valptr = extval; *valptr; valptr++)
{
char ch = *valptr;
if (ch == '\'' || ch == '\\')
{
appendStringInfoChar(buf, '\\');
appendStringInfoChar(buf, ch);
}
else if (ch >= 0 && ch < ' ')
appendStringInfo(buf, "\\%03o", (int) ch);
else
appendStringInfoChar(buf, ch);
}
appendStringInfoChar(buf, '\'');
break;
}
pfree(extval);
switch (constval->consttype)
{
case INT4OID:
case FLOAT8OID:
case UNKNOWNOID:
/* These types can be left unlabeled */
break;
default:
extval = pstrdup(NameStr(typeStruct->typname));
appendStringInfo(buf, "::%s", quote_identifier(extval));
pfree(extval);
break;
}
}
/* ----------
* get_sublink_expr - Parse back a sublink
* ----------
*/
static void
get_sublink_expr(Node *node, deparse_context *context)
{
StringInfo buf = context->buf;
SubLink *sublink = (SubLink *) node;
Query *query = (Query *) (sublink->subselect);
Oper *oper;
List *l;
char *sep;
appendStringInfo(buf, "(");
if (sublink->lefthand != NULL)
{
if (length(sublink->lefthand) > 1)
appendStringInfo(buf, "(");
sep = "";
foreach(l, sublink->lefthand)
{
appendStringInfo(buf, sep);
sep = ", ";
get_rule_expr((Node *) lfirst(l), context);
}
if (length(sublink->lefthand) > 1)
appendStringInfo(buf, ") ");
else
appendStringInfo(buf, " ");
}
1999-05-25 18:15:34 +02:00
switch (sublink->subLinkType)
{
case EXISTS_SUBLINK:
appendStringInfo(buf, "EXISTS ");
break;
case ANY_SUBLINK:
oper = (Oper *) lfirst(sublink->oper);
appendStringInfo(buf, "%s ANY ", get_opname(oper->opno));
break;
case ALL_SUBLINK:
oper = (Oper *) lfirst(sublink->oper);
appendStringInfo(buf, "%s ALL ", get_opname(oper->opno));
break;
case EXPR_SUBLINK:
oper = (Oper *) lfirst(sublink->oper);
appendStringInfo(buf, "%s ", get_opname(oper->opno));
break;
default:
elog(ERROR, "get_sublink_expr: unsupported sublink type %d",
1999-05-25 18:15:34 +02:00
sublink->subLinkType);
break;
}
appendStringInfo(buf, "(");
get_query_def(query, buf, context->rangetables);
appendStringInfo(buf, "))");
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* quote_identifier - Quote an identifier only if needed
*
* When quotes are needed, we palloc the required space; slightly
* space-wasteful but well worth it for notational simplicity.
* ----------
*/
static char *
quote_identifier(char *ident)
{
/*
* Can avoid quoting if ident starts with a lowercase letter and
* contains only lowercase letters, digits, and underscores.
* Otherwise, supply quotes.
*/
bool safe;
char *result;
/*
* would like to use <ctype.h> macros here, but they might yield
* unwanted locale-specific results...
*/
safe = (ident[0] >= 'a' && ident[0] <= 'z');
if (safe)
{
char *ptr;
for (ptr = ident+1; *ptr; ptr++)
{
char ch = *ptr;
safe = ((ch >= 'a' && ch <= 'z') ||
(ch >= '0' && ch <= '9') ||
(ch == '_'));
if (! safe)
break;
}
}
if (safe)
return ident; /* no change needed */
result = (char *) palloc(strlen(ident) + 2 + 1);
sprintf(result, "\"%s\"", ident);
return result;
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
/* ----------
* get_relation_name - Get a relation name by Oid
* ----------
*/
static char *
get_relation_name(Oid relid)
{
HeapTuple classtup;
Form_pg_class classStruct;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
classtup = SearchSysCacheTuple(RELOID,
ObjectIdGetDatum(relid), 0, 0, 0);
if (!HeapTupleIsValid(classtup))
elog(ERROR, "cache lookup of relation %u failed", relid);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
classStruct = (Form_pg_class) GETSTRUCT(classtup);
return pstrdup(NameStr(classStruct->relname));
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* get_attribute_name - Get an attribute name by it's
* relations Oid and it's attnum
* ----------
*/
static char *
get_attribute_name(Oid relid, int2 attnum)
{
HeapTuple atttup;
Form_pg_attribute attStruct;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
atttup = SearchSysCacheTuple(ATTNUM,
ObjectIdGetDatum(relid), (Datum) attnum,
0, 0);
if (!HeapTupleIsValid(atttup))
elog(ERROR, "cache lookup of attribute %d in relation %u failed",
attnum, relid);
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
attStruct = (Form_pg_attribute) GETSTRUCT(atttup);
return pstrdup(NameStr(attStruct->attname));
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
}
/* ----------
* check_if_rte_used
* Check a targetlist or qual to see if a given rangetable entry
* is used in it
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
* ----------
*/
static bool
check_if_rte_used(Node *node, Index rt_index, int levelsup)
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
{
check_if_rte_used_context context;
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
context.rt_index = rt_index;
context.levelsup = levelsup;
return check_if_rte_used_walker(node, &context);
}
This is the final state of the rule system for 6.4 after the patch is applied: Rewrite rules on relation level work fine now. Event qualifications on insert/update/delete rules work fine now. I added the new keyword OLD to reference the CURRENT tuple. CURRENT will be removed in 6.5. Update rules can reference NEW and OLD in the rule qualification and the actions. Insert/update/delete rules on views can be established to let them behave like real tables. For insert/update/delete rules multiple actions are supported now. The actions can also be surrounded by parantheses to make psql happy. Multiple actions are required if update to a view requires updates to multiple tables. Regular users are permitted to create/drop rules on tables they have RULE permissions for (DefineQueryRewrite() is now able to get around the access restrictions on pg_rewrite). This enables view creation for regular users too. This required an extra boolean parameter to pg_parse_and_plan() that tells to set skipAcl on all rangetable entries of the resulting queries. There is a new function pg_exec_query_acl_override() that could be used by backend utilities to use this facility. All rule actions (not only views) inherit the permissions of the event relations owner. Sample: User A creates tables T1 and T2, creates rules that log INSERT/UPDATE/DELETE on T1 in T2 (like in the regression tests for rules I created) and grants ALL but RULE on T1 to user B. User B can now fully access T1 and the logging happens in T2. But user B cannot access T2 at all, only the rule actions can. And due to missing RULE permissions on T1, user B cannot disable logging. Rules on the attribute level are disabled (they don't work properly and since regular users are now permitted to create rules I decided to disable them). Rules on select must have exactly one action that is a select (so select rules must be a view definition). UPDATE NEW/OLD rules are disabled (still broken, but triggers can do it). There are two new system views (pg_rule and pg_view) that show the definition of the rules or views so the db admin can see what the users do. They use two new functions pg_get_ruledef() and pg_get_viewdef() that are builtins. The functions pg_get_ruledef() and pg_get_viewdef() could be used to implement rule and view support in pg_dump. PostgreSQL is now the only database system I know, that has rewrite rules on the query level. All others (where I found a rule statement at all) use stored database procedures or the like (triggers as we call them) for active rules (as some call them). Future of the rule system: The now disabled parts of the rule system (attribute level, multiple actions on select and update new stuff) require a complete new rewrite handler from scratch. The old one is too badly wired up. After 6.4 I'll start to work on a new rewrite handler, that fully supports the attribute level rules, multiple actions on select and update new. This will be available for 6.5 so we get full rewrite rule capabilities. Jan
1998-08-24 03:38:11 +02:00
static bool
check_if_rte_used_walker(Node *node,
check_if_rte_used_context *context)
{
if (node == NULL)
return false;
if (IsA(node, Var))
{
Var *var = (Var *) node;
return var->varno == context->rt_index &&
var->varlevelsup == context->levelsup;
}
if (IsA(node, SubLink))
{
SubLink *sublink = (SubLink *) node;
Query *query = (Query *) sublink->subselect;
/* Recurse into subquery; expression_tree_walker will not */
if (check_if_rte_used((Node *) (query->targetList),
context->rt_index, context->levelsup + 1) ||
check_if_rte_used(query->qual,
context->rt_index, context->levelsup + 1) ||
check_if_rte_used(query->havingQual,
context->rt_index, context->levelsup + 1))
return true;
/* fall through to let expression_tree_walker examine lefthand args */
}
return expression_tree_walker(node, check_if_rte_used_walker,
(void *) context);
}