postgresql/src/backend/rewrite/rewriteHandler.c

2696 lines
51 KiB
C
Raw Normal View History

/*-------------------------------------------------------------------------
*
* rewriteHandler.c--
*
* Copyright (c) 1994, Regents of the University of California
*
*
* IDENTIFICATION
* $Header: /cvsroot/pgsql/src/backend/rewrite/rewriteHandler.c,v 1.23 1998/10/02 21:53:39 momjian Exp $
*
*-------------------------------------------------------------------------
*/
#include <string.h>
#include "postgres.h"
#include "miscadmin.h"
#include "utils/palloc.h"
#include "utils/elog.h"
#include "utils/rel.h"
#include "nodes/pg_list.h"
#include "nodes/primnodes.h"
#include "nodes/relation.h"
#include "parser/parsetree.h" /* for parsetree manipulation */
#include "parser/parse_relation.h"
#include "nodes/parsenodes.h"
#include "rewrite/rewriteSupport.h"
#include "rewrite/rewriteHandler.h"
#include "rewrite/rewriteManip.h"
#include "rewrite/locks.h"
#include "commands/creatinh.h"
#include "access/heapam.h"
#include "utils/lsyscache.h"
#include "utils/syscache.h"
#include "utils/acl.h"
#include "catalog/pg_shadow.h"
#include "catalog/pg_type.h"
static RewriteInfo *gatherRewriteMeta(Query *parsetree,
Query *rule_action,
Node *rule_qual,
int rt_index,
CmdType event,
bool *instead_flag);
static bool rangeTableEntry_used(Node *node, int rt_index, int sublevels_up);
static bool attribute_used(Node *node, int rt_index, int attno, int sublevels_up);
static void offset_varnodes(Node *node, int offset, int sublevels_up);
static void change_varnodes(Node *node, int rt_index, int new_index, int sublevels_up);
static void modifyAggregUplevel(Node *node);
static void modifyAggregChangeVarnodes(Node **nodePtr, int rt_index, int new_index, int sublevels_up);
static void modifyAggregDropQual(Node **nodePtr, Node *orignode, Expr *expr);
static SubLink *modifyAggregMakeSublink(Expr *origexp, Query *parsetree);
static void modifyAggregQual(Node **nodePtr, Query *parsetree);
static Query *fireRIRrules(Query *parsetree);
/*
* gatherRewriteMeta -
* Gather meta information about parsetree, and rule. Fix rule body
* and qualifier so that they can be mixed with the parsetree and
* maintain semantic validity
*/
static RewriteInfo *
gatherRewriteMeta(Query *parsetree,
Query *rule_action,
Node *rule_qual,
int rt_index,
CmdType event,
bool *instead_flag)
{
RewriteInfo *info;
int rt_length;
int result_reln;
info = (RewriteInfo *) palloc(sizeof(RewriteInfo));
info->rt_index = rt_index;
info->event = event;
info->instead_flag = *instead_flag;
info->rule_action = (Query *) copyObject(rule_action);
info->rule_qual = (Node *) copyObject(rule_qual);
if (info->rule_action == NULL)
info->nothing = TRUE;
else
{
info->nothing = FALSE;
info->action = info->rule_action->commandType;
info->current_varno = rt_index;
info->rt = parsetree->rtable;
rt_length = length(info->rt);
info->rt = append(info->rt, info->rule_action->rtable);
info->new_varno = PRS2_NEW_VARNO + rt_length;
OffsetVarNodes(info->rule_action->qual, rt_length);
OffsetVarNodes((Node *) info->rule_action->targetList, rt_length);
OffsetVarNodes(info->rule_qual, rt_length);
ChangeVarNodes((Node *) info->rule_action->qual,
1998-01-21 05:24:46 +01:00
PRS2_CURRENT_VARNO + rt_length, rt_index, 0);
ChangeVarNodes((Node *) info->rule_action->targetList,
1998-01-21 05:24:46 +01:00
PRS2_CURRENT_VARNO + rt_length, rt_index, 0);
ChangeVarNodes(info->rule_qual,
1998-01-21 05:24:46 +01:00
PRS2_CURRENT_VARNO + rt_length, rt_index, 0);
/*
* bug here about replace CURRENT -- sort of replace current is
* deprecated now so this code shouldn't really need to be so
* clutzy but.....
*/
if (info->action != CMD_SELECT)
{ /* i.e update XXXXX */
int new_result_reln = 0;
result_reln = info->rule_action->resultRelation;
switch (result_reln)
{
case PRS2_CURRENT_VARNO:
new_result_reln = rt_index;
break;
case PRS2_NEW_VARNO: /* XXX */
default:
new_result_reln = result_reln + rt_length;
break;
}
info->rule_action->resultRelation = new_result_reln;
}
}
return info;
}
/*
* rangeTableEntry_used -
* we need to process a RTE for RIR rules only if it is
* referenced somewhere in var nodes of the query.
*/
static bool
rangeTableEntry_used(Node *node, int rt_index, int sublevels_up)
{
if (node == NULL)
return FALSE;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
return rangeTableEntry_used(
(Node *)(tle->expr),
rt_index,
sublevels_up);
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
return rangeTableEntry_used(
(Node *)(agg->target),
rt_index,
sublevels_up);
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
return rangeTableEntry_used(
(Node *)(grp->entry),
rt_index,
sublevels_up);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
return rangeTableEntry_used(
(Node *)(exp->args),
rt_index,
sublevels_up);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
return rangeTableEntry_used(
(Node *)(iter->iterexpr),
rt_index,
sublevels_up);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
if (rangeTableEntry_used(
(Node *)(ref->refupperindexpr),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(ref->reflowerindexpr),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(ref->refexpr),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(ref->refassgnexpr),
rt_index,
sublevels_up))
return TRUE;
return FALSE;
}
break;
case T_Var:
{
Var *var = (Var *)node;
if (var->varlevelsup == sublevels_up)
return var->varno == rt_index;
else
return FALSE;
}
break;
case T_Param:
return FALSE;
case T_Const:
return FALSE;
case T_List:
{
List *l;
foreach (l, (List *)node) {
if (rangeTableEntry_used(
(Node *)lfirst(l),
rt_index,
sublevels_up))
return TRUE;
}
return FALSE;
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
if (rangeTableEntry_used(
(Node *)(sub->lefthand),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(sub->subselect),
rt_index,
sublevels_up + 1))
return TRUE;
return FALSE;
}
break;
case T_Query:
{
Query *qry = (Query *)node;
if (rangeTableEntry_used(
(Node *)(qry->targetList),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(qry->qual),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(qry->havingQual),
rt_index,
sublevels_up))
return TRUE;
if (rangeTableEntry_used(
(Node *)(qry->groupClause),
rt_index,
sublevels_up))
return TRUE;
return FALSE;
}
break;
default:
elog(NOTICE, "unknown node tag %d in rangeTableEntry_used()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
return FALSE;
}
/*
* attribute_used -
* Check if a specific attribute number of a RTE is used
* somewhere in the query
*/
static bool
attribute_used(Node *node, int rt_index, int attno, int sublevels_up)
{
if (node == NULL)
return FALSE;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
return attribute_used(
(Node *)(tle->expr),
rt_index,
attno,
sublevels_up);
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
return attribute_used(
(Node *)(agg->target),
rt_index,
attno,
sublevels_up);
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
return attribute_used(
(Node *)(grp->entry),
rt_index,
attno,
sublevels_up);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
return attribute_used(
(Node *)(exp->args),
rt_index,
attno,
sublevels_up);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
return attribute_used(
(Node *)(iter->iterexpr),
rt_index,
attno,
sublevels_up);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
if (attribute_used(
(Node *)(ref->refupperindexpr),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(ref->reflowerindexpr),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(ref->refexpr),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(ref->refassgnexpr),
rt_index,
attno,
sublevels_up))
return TRUE;
return FALSE;
}
break;
case T_Var:
{
Var *var = (Var *)node;
if (var->varlevelsup == sublevels_up)
return var->varno == rt_index;
else
return FALSE;
}
break;
case T_Param:
return FALSE;
case T_Const:
return FALSE;
case T_List:
{
List *l;
foreach (l, (List *)node) {
if (attribute_used(
(Node *)lfirst(l),
rt_index,
attno,
sublevels_up))
return TRUE;
}
return FALSE;
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
if (attribute_used(
(Node *)(sub->lefthand),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(sub->subselect),
rt_index,
attno,
sublevels_up + 1))
return TRUE;
return FALSE;
}
break;
case T_Query:
{
Query *qry = (Query *)node;
if (attribute_used(
(Node *)(qry->targetList),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(qry->qual),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(qry->havingQual),
rt_index,
attno,
sublevels_up))
return TRUE;
if (attribute_used(
(Node *)(qry->groupClause),
rt_index,
attno,
sublevels_up))
return TRUE;
return FALSE;
}
break;
default:
elog(NOTICE, "unknown node tag %d in attribute_used()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
return 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
/*
* offset_varnodes -
* We need another version of OffsetVarNodes() when processing
* RIR rules
*/
static void
offset_varnodes(Node *node, int offset, int sublevels_up)
{
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
offset_varnodes(
(Node *)(tle->expr),
offset,
sublevels_up);
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
offset_varnodes(
(Node *)(agg->target),
offset,
sublevels_up);
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
offset_varnodes(
(Node *)(grp->entry),
offset,
sublevels_up);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
offset_varnodes(
(Node *)(exp->args),
offset,
sublevels_up);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
offset_varnodes(
(Node *)(iter->iterexpr),
offset,
sublevels_up);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
offset_varnodes(
(Node *)(ref->refupperindexpr),
offset,
sublevels_up);
offset_varnodes(
(Node *)(ref->reflowerindexpr),
offset,
sublevels_up);
offset_varnodes(
(Node *)(ref->refexpr),
offset,
sublevels_up);
offset_varnodes(
(Node *)(ref->refassgnexpr),
offset,
sublevels_up);
}
break;
case T_Var:
{
Var *var = (Var *)node;
if (var->varlevelsup == sublevels_up) {
var->varno += offset;
var->varnoold += offset;
}
}
break;
case T_Param:
break;
case T_Const:
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
offset_varnodes(
(Node *)lfirst(l),
offset,
sublevels_up);
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
offset_varnodes(
(Node *)(sub->lefthand),
offset,
sublevels_up);
offset_varnodes(
(Node *)(sub->subselect),
offset,
sublevels_up + 1);
}
break;
case T_Query:
{
Query *qry = (Query *)node;
offset_varnodes(
(Node *)(qry->targetList),
offset,
sublevels_up);
offset_varnodes(
(Node *)(qry->qual),
offset,
sublevels_up);
offset_varnodes(
(Node *)(qry->havingQual),
offset,
sublevels_up);
offset_varnodes(
(Node *)(qry->groupClause),
offset,
sublevels_up);
}
break;
default:
elog(NOTICE, "unknown node tag %d in offset_varnodes()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
/*
* change_varnodes -
* and another ChangeVarNodes() too
*/
static void
change_varnodes(Node *node, int rt_index, int new_index, int sublevels_up)
{
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
change_varnodes(
(Node *)(tle->expr),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
change_varnodes(
(Node *)(agg->target),
rt_index,
new_index,
sublevels_up);
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
change_varnodes(
(Node *)(grp->entry),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
change_varnodes(
(Node *)(exp->args),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
change_varnodes(
(Node *)(iter->iterexpr),
rt_index,
new_index,
sublevels_up);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
change_varnodes(
(Node *)(ref->refupperindexpr),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(ref->reflowerindexpr),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(ref->refexpr),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(ref->refassgnexpr),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Var:
{
Var *var = (Var *)node;
if (var->varlevelsup == sublevels_up &&
var->varno == rt_index) {
var->varno = new_index;
var->varnoold = new_index;
}
}
break;
case T_Param:
break;
case T_Const:
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
change_varnodes(
(Node *)lfirst(l),
rt_index,
new_index,
sublevels_up);
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
change_varnodes(
(Node *)(sub->lefthand),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(sub->subselect),
rt_index,
new_index,
sublevels_up + 1);
}
break;
case T_Query:
{
Query *qry = (Query *)node;
change_varnodes(
(Node *)(qry->targetList),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(qry->qual),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(qry->havingQual),
rt_index,
new_index,
sublevels_up);
change_varnodes(
(Node *)(qry->groupClause),
rt_index,
new_index,
sublevels_up);
}
break;
default:
elog(NOTICE, "unknown node tag %d in change_varnodes()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
/*
* modifyAggregUplevel -
* In the newly created sublink for an aggregate column used in
* the qualification, we must adjust the varlevelsup in all the
* var nodes.
*/
static void
modifyAggregUplevel(Node *node)
{
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
modifyAggregUplevel(
(Node *)(tle->expr));
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
modifyAggregUplevel(
(Node *)(agg->target));
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
modifyAggregUplevel(
(Node *)(exp->args));
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
modifyAggregUplevel(
(Node *)(iter->iterexpr));
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
modifyAggregUplevel(
(Node *)(ref->refupperindexpr));
modifyAggregUplevel(
(Node *)(ref->reflowerindexpr));
modifyAggregUplevel(
(Node *)(ref->refexpr));
modifyAggregUplevel(
(Node *)(ref->refassgnexpr));
}
break;
case T_Var:
{
Var *var = (Var *)node;
var->varlevelsup++;
}
break;
case T_Param:
break;
case T_Const:
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
modifyAggregUplevel(
(Node *)lfirst(l));
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
modifyAggregUplevel(
(Node *)(sub->lefthand));
modifyAggregUplevel(
(Node *)(sub->oper));
modifyAggregUplevel(
(Node *)(sub->subselect));
}
break;
case T_Query:
{
Query *qry = (Query *)node;
modifyAggregUplevel(
(Node *)(qry->targetList));
modifyAggregUplevel(
(Node *)(qry->qual));
modifyAggregUplevel(
(Node *)(qry->havingQual));
modifyAggregUplevel(
(Node *)(qry->groupClause));
}
break;
default:
elog(NOTICE, "unknown node tag %d in modifyAggregUplevel()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
/*
* modifyAggregChangeVarnodes -
* Change the var nodes in a sublink created for an aggregate column
* used in the qualification that is subject of the aggregate
* function to point to the correct local RTE.
*/
static void
modifyAggregChangeVarnodes(Node **nodePtr, int rt_index, int new_index, int sublevels_up)
{
Node *node = *nodePtr;
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
modifyAggregChangeVarnodes(
(Node **)(&(tle->expr)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
modifyAggregChangeVarnodes(
(Node **)(&(agg->target)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
modifyAggregChangeVarnodes(
(Node **)(&(grp->entry)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
modifyAggregChangeVarnodes(
(Node **)(&(exp->args)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
modifyAggregChangeVarnodes(
(Node **)(&(iter->iterexpr)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
modifyAggregChangeVarnodes(
(Node **)(&(ref->refupperindexpr)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(ref->reflowerindexpr)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(ref->refexpr)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(ref->refassgnexpr)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_Var:
{
Var *var = (Var *)node;
if (var->varlevelsup == sublevels_up &&
var->varno == rt_index) {
var = copyObject(var);
var->varno = new_index;
var->varnoold = new_index;
var->varlevelsup = 0;
*nodePtr = (Node *)var;
}
}
break;
case T_Param:
break;
case T_Const:
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
modifyAggregChangeVarnodes(
(Node **)(&lfirst(l)),
rt_index,
new_index,
sublevels_up);
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
modifyAggregChangeVarnodes(
(Node **)(&(sub->lefthand)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(sub->oper)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(sub->subselect)),
rt_index,
new_index,
sublevels_up + 1);
}
break;
case T_Query:
{
Query *qry = (Query *)node;
modifyAggregChangeVarnodes(
(Node **)(&(qry->targetList)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(qry->qual)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(qry->havingQual)),
rt_index,
new_index,
sublevels_up);
modifyAggregChangeVarnodes(
(Node **)(&(qry->groupClause)),
rt_index,
new_index,
sublevels_up);
}
break;
default:
elog(NOTICE, "unknown node tag %d in modifyAggregChangeVarnodes()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
/*
* modifyAggregDropQual -
* remove the pure aggreg clase from a qualification
*/
static void
modifyAggregDropQual(Node **nodePtr, Node *orignode, Expr *expr)
{
Node *node = *nodePtr;
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_Var:
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
Aggreg *oagg = (Aggreg *)orignode;
modifyAggregDropQual(
(Node **)(&(agg->target)),
(Node *)(oagg->target),
expr);
}
break;
case T_Param:
break;
case T_Const:
break;
case T_GroupClause:
break;
case T_Expr:
{
Expr *this_expr = (Expr *)node;
Expr *orig_expr = (Expr *)orignode;
if (orig_expr == expr) {
Const *ctrue;
if (expr->typeOid != BOOLOID)
elog(ERROR,
"aggregate expression in qualification isn't of type bool");
ctrue = makeNode(Const);
ctrue->consttype = BOOLOID;
ctrue->constlen = 1;
ctrue->constisnull = FALSE;
ctrue->constvalue = (Datum)TRUE;
ctrue->constbyval = TRUE;
*nodePtr = (Node *)ctrue;
}
else
modifyAggregDropQual(
(Node **)(&(this_expr->args)),
(Node *)(orig_expr->args),
expr);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
Iter *oiter = (Iter *)orignode;
modifyAggregDropQual(
(Node **)(&(iter->iterexpr)),
(Node *)(oiter->iterexpr),
expr);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
ArrayRef *oref = (ArrayRef *)orignode;
modifyAggregDropQual(
(Node **)(&(ref->refupperindexpr)),
(Node *)(oref->refupperindexpr),
expr);
modifyAggregDropQual(
(Node **)(&(ref->reflowerindexpr)),
(Node *)(oref->reflowerindexpr),
expr);
modifyAggregDropQual(
(Node **)(&(ref->refexpr)),
(Node *)(oref->refexpr),
expr);
modifyAggregDropQual(
(Node **)(&(ref->refassgnexpr)),
(Node *)(oref->refassgnexpr),
expr);
}
break;
case T_List:
{
List *l;
List *ol = (List *)orignode;
int li = 0;
foreach (l, (List *)node) {
modifyAggregDropQual(
(Node **)(&(lfirst(l))),
(Node *)nth(li, ol),
expr);
li++;
}
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
SubLink *osub = (SubLink *)orignode;
modifyAggregDropQual(
(Node **)(&(sub->subselect)),
(Node *)(osub->subselect),
expr);
}
break;
case T_Query:
{
Query *qry = (Query *)node;
Query *oqry = (Query *)orignode;
modifyAggregDropQual(
(Node **)(&(qry->qual)),
(Node *)(oqry->qual),
expr);
modifyAggregDropQual(
(Node **)(&(qry->havingQual)),
(Node *)(oqry->havingQual),
expr);
}
break;
default:
elog(NOTICE, "unknown node tag %d in modifyAggregDropQual()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
/*
* modifyAggregMakeSublink -
* Create a sublink node for a qualification expression that
* uses an aggregate column of a view
*/
static SubLink *
modifyAggregMakeSublink(Expr *origexp, Query *parsetree)
{
SubLink *sublink;
Query *subquery;
Node *subqual;
RangeTblEntry *rte;
Aggreg *aggreg;
Var *target;
TargetEntry *tle;
Resdom *resdom;
Expr *exp = copyObject(origexp);
if (nodeTag(nth(0, exp->args)) == T_Aggreg)
if (nodeTag(nth(1, exp->args)) == T_Aggreg)
elog(ERROR, "rewrite: comparision of 2 aggregate columns not supported");
else
elog(ERROR, "rewrite: aggregate column of view must be at rigth side in qual");
aggreg = (Aggreg *)nth(1, exp->args);
target = (Var *)(aggreg->target);
rte = (RangeTblEntry *)nth(target->varno - 1, parsetree->rtable);
tle = makeNode(TargetEntry);
resdom = makeNode(Resdom);
aggreg->usenulls = TRUE;
resdom->resno = 1;
resdom->restype = ((Oper *)(exp->oper))->opresulttype;
resdom->restypmod = -1;
resdom->resname = pstrdup("<noname>");
resdom->reskey = 0;
resdom->reskeyop = 0;
resdom->resjunk = 0;
tle->resdom = resdom;
tle->expr = (Node *)aggreg;
subqual = copyObject(parsetree->qual);
modifyAggregDropQual((Node **)&subqual, (Node *)parsetree->qual, origexp);
sublink = makeNode(SubLink);
sublink->subLinkType = EXPR_SUBLINK;
sublink->useor = FALSE;
sublink->lefthand = lappend(NIL, copyObject(lfirst(exp->args)));
sublink->oper = lappend(NIL, copyObject(exp));
sublink->subselect = NULL;
subquery = makeNode(Query);
sublink->subselect = (Node *)subquery;
subquery->commandType = CMD_SELECT;
subquery->utilityStmt = NULL;
subquery->resultRelation = 0;
subquery->into = NULL;
subquery->isPortal = FALSE;
subquery->isBinary = FALSE;
subquery->unionall = FALSE;
subquery->uniqueFlag = NULL;
subquery->sortClause = NULL;
subquery->rtable = lappend(NIL, rte);
subquery->targetList = lappend(NIL, tle);
subquery->qual = subqual;
subquery->groupClause = NIL;
subquery->havingQual = NULL;
subquery->hasAggs = TRUE;
subquery->hasSubLinks = FALSE;
subquery->unionClause = NULL;
modifyAggregUplevel((Node *)sublink);
modifyAggregChangeVarnodes((Node **)&(sublink->lefthand), target->varno,
1, target->varlevelsup);
modifyAggregChangeVarnodes((Node **)&(sublink->oper), target->varno,
1, target->varlevelsup);
modifyAggregChangeVarnodes((Node **)&(sublink->subselect), target->varno,
1, target->varlevelsup);
return sublink;
}
/*
* modifyAggregQual -
* Search for qualification expressions that contain aggregate
* functions and substiture them by sublinks. These expressions
* originally come from qualifications that use aggregate columns
* of a view.
*/
static void
modifyAggregQual(Node **nodePtr, Query *parsetree)
{
Node *node = *nodePtr;
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_Var:
break;
case T_Param:
break;
case T_Const:
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
modifyAggregQual(
(Node **)(&(grp->entry)),
parsetree);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
SubLink *sub;
if (length(exp->args) != 2) {
modifyAggregQual(
(Node **)(&(exp->args)),
parsetree);
break;
}
if (nodeTag(nth(0, exp->args)) != T_Aggreg &&
nodeTag(nth(1, exp->args)) != T_Aggreg) {
modifyAggregQual(
(Node **)(&(exp->args)),
parsetree);
break;
}
sub = modifyAggregMakeSublink(exp,
parsetree);
*nodePtr = (Node *)sub;
parsetree->hasSubLinks = TRUE;
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
modifyAggregQual(
(Node **)(&(iter->iterexpr)),
parsetree);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
modifyAggregQual(
(Node **)(&(ref->refupperindexpr)),
parsetree);
modifyAggregQual(
(Node **)(&(ref->reflowerindexpr)),
parsetree);
modifyAggregQual(
(Node **)(&(ref->refexpr)),
parsetree);
modifyAggregQual(
(Node **)(&(ref->refassgnexpr)),
parsetree);
}
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
modifyAggregQual(
(Node **)(&(lfirst(l))),
parsetree);
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
modifyAggregQual(
(Node **)(&(sub->subselect)),
(Query *)(sub->subselect));
}
break;
case T_Query:
{
Query *qry = (Query *)node;
modifyAggregQual(
(Node **)(&(qry->qual)),
parsetree);
modifyAggregQual(
(Node **)(&(qry->havingQual)),
parsetree);
}
break;
default:
elog(NOTICE, "unknown node tag %d in modifyAggregQual()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
static Node *
FindMatchingTLEntry(List *tlist, char *e_attname)
{
List *i;
foreach(i, tlist)
{
TargetEntry *tle = lfirst(i);
char *resname;
resname = tle->resdom->resname;
if (!strcmp(e_attname, resname))
return (tle->expr);
}
return NULL;
}
static Node *
make_null(Oid type)
{
Const *c = makeNode(Const);
c->consttype = type;
c->constlen = get_typlen(type);
c->constvalue = PointerGetDatum(NULL);
c->constisnull = true;
c->constbyval = get_typbyval(type);
return (Node *) c;
}
static void
apply_RIR_view(Node **nodePtr, int rt_index, RangeTblEntry *rte, List *tlist, int *modified, int sublevels_up)
{
Node *node = *nodePtr;
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
apply_RIR_view(
(Node **)(&(tle->expr)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
apply_RIR_view(
(Node **)(&(agg->target)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
apply_RIR_view(
(Node **)(&(grp->entry)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
apply_RIR_view(
(Node **)(&(exp->args)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
apply_RIR_view(
(Node **)(&(iter->iterexpr)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
apply_RIR_view(
(Node **)(&(ref->refupperindexpr)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(ref->reflowerindexpr)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(ref->refexpr)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(ref->refassgnexpr)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_Var:
{
Var *var = (Var *)node;
if (var->varlevelsup == sublevels_up &&
var->varno == rt_index) {
Node *exp;
if (var->varattno < 0)
elog(ERROR, "system column %s not available - %s is a view", get_attname(rte->relid, var->varattno), rte->relname);
exp = FindMatchingTLEntry(
tlist,
get_attname(rte->relid,
var->varattno));
if (exp == NULL) {
*nodePtr = make_null(var->vartype);
return;
}
if (var->varlevelsup > 0 &&
nodeTag(exp) == T_Var) {
exp = copyObject(exp);
((Var *)exp)->varlevelsup = var->varlevelsup;
}
*nodePtr = exp;
*modified = TRUE;
}
}
break;
case T_Param:
break;
case T_Const:
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
apply_RIR_view(
(Node **)(&(lfirst(l))),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
apply_RIR_view(
(Node **)(&(sub->lefthand)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(sub->subselect)),
rt_index,
rte,
tlist,
modified,
sublevels_up + 1);
}
break;
case T_Query:
{
Query *qry = (Query *)node;
apply_RIR_view(
(Node **)(&(qry->targetList)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(qry->qual)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(qry->havingQual)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
apply_RIR_view(
(Node **)(&(qry->groupClause)),
rt_index,
rte,
tlist,
modified,
sublevels_up);
}
break;
default:
elog(NOTICE, "unknown node tag %d in apply_RIR_view()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
static void
ApplyRetrieveRule(Query *parsetree,
RewriteRule *rule,
int rt_index,
int relation_level,
Relation relation,
int *modified)
{
Query *rule_action = NULL;
Node *rule_qual;
List *rtable,
*rt;
int nothing,
rt_length;
int badsql = FALSE;
rule_qual = rule->qual;
if (rule->actions)
{
if (length(rule->actions) > 1) /* ??? because we don't handle
* rules with more than one
* action? -ay */
return;
rule_action = copyObject(lfirst(rule->actions));
nothing = FALSE;
}
else
nothing = TRUE;
rtable = copyObject(parsetree->rtable);
foreach(rt, rtable)
{
RangeTblEntry *rte = lfirst(rt);
/*
* this is to prevent add_missing_vars_to_base_rels() from adding
* a bogus entry to the new target list.
*/
rte->inFromCl = false;
}
rt_length = length(rtable);
rtable = nconc(rtable, copyObject(rule_action->rtable));
parsetree->rtable = rtable;
rule_action->rtable = rtable;
offset_varnodes((Node *) rule_qual, rt_length, 0);
offset_varnodes((Node *) rule_action, rt_length, 0);
change_varnodes((Node *) rule_qual,
PRS2_CURRENT_VARNO + rt_length, rt_index, 0);
change_varnodes((Node *) rule_action,
PRS2_CURRENT_VARNO + rt_length, rt_index, 0);
if (relation_level)
{
apply_RIR_view((Node **) &parsetree, rt_index,
(RangeTblEntry *)nth(rt_index - 1, rtable),
rule_action->targetList, modified, 0);
apply_RIR_view((Node **) &rule_action, rt_index,
(RangeTblEntry *)nth(rt_index - 1, rtable),
rule_action->targetList, modified, 0);
}
else
{
HandleRIRAttributeRule(parsetree, rtable, rule_action->targetList,
rt_index, rule->attrno, modified, &badsql);
}
if (*modified && !badsql) {
AddQual(parsetree, rule_action->qual);
/* This will only work if the query made to the view defined by the following
* groupClause groups by the same attributes or does not use group at all! */
if (parsetree->groupClause == NULL)
parsetree->groupClause=rule_action->groupClause;
AddHavingQual(parsetree, rule_action->havingQual);
parsetree->hasAggs = (rule_action->hasAggs || parsetree->hasAggs);
parsetree->hasSubLinks = (rule_action->hasSubLinks || parsetree->hasSubLinks);
}
}
static void
fireRIRonSubselect(Node *node)
{
if (node == NULL)
return;
switch(nodeTag(node)) {
case T_TargetEntry:
{
TargetEntry *tle = (TargetEntry *)node;
fireRIRonSubselect(
(Node *)(tle->expr));
}
break;
case T_Aggreg:
{
Aggreg *agg = (Aggreg *)node;
fireRIRonSubselect(
(Node *)(agg->target));
}
break;
case T_GroupClause:
{
GroupClause *grp = (GroupClause *)node;
fireRIRonSubselect(
(Node *)(grp->entry));
}
break;
case T_Expr:
{
Expr *exp = (Expr *)node;
fireRIRonSubselect(
(Node *)(exp->args));
}
break;
case T_Iter:
{
Iter *iter = (Iter *)node;
fireRIRonSubselect(
(Node *)(iter->iterexpr));
}
break;
case T_ArrayRef:
{
ArrayRef *ref = (ArrayRef *)node;
fireRIRonSubselect(
(Node *)(ref->refupperindexpr));
fireRIRonSubselect(
(Node *)(ref->reflowerindexpr));
fireRIRonSubselect(
(Node *)(ref->refexpr));
fireRIRonSubselect(
(Node *)(ref->refassgnexpr));
}
break;
case T_Var:
break;
case T_Param:
break;
case T_Const:
break;
case T_List:
{
List *l;
foreach (l, (List *)node)
fireRIRonSubselect(
(Node *)(lfirst(l)));
}
break;
case T_SubLink:
{
SubLink *sub = (SubLink *)node;
Query *qry;
fireRIRonSubselect(
(Node *)(sub->lefthand));
qry = fireRIRrules((Query *)(sub->subselect));
fireRIRonSubselect(
(Node *)qry);
sub->subselect = (Node *) qry;
}
break;
case T_Query:
{
Query *qry = (Query *)node;
fireRIRonSubselect(
(Node *)(qry->targetList));
fireRIRonSubselect(
(Node *)(qry->qual));
fireRIRonSubselect(
(Node *)(qry->havingQual));
fireRIRonSubselect(
(Node *)(qry->groupClause));
}
break;
default:
elog(NOTICE, "unknown node tag %d in fireRIRonSubselect()", nodeTag(node));
elog(NOTICE, "Node is: %s", nodeToString(node));
break;
}
}
/*
* fireRIRrules -
* Apply all RIR rules on each rangetable entry in a query
*/
static Query *
fireRIRrules(Query *parsetree)
{
int rt_index;
RangeTblEntry *rte;
Relation rel;
List *locks;
RuleLock *rules;
RewriteRule *rule;
RewriteRule RIRonly;
int modified;
int i;
List *l;
rt_index = 0;
while(rt_index < length(parsetree->rtable)) {
++rt_index;
if (!rangeTableEntry_used((Node *)parsetree, rt_index, 0))
continue;
rte = nth(rt_index - 1, parsetree->rtable);
rel = heap_openr(rte->relname);
if (rel->rd_rules == NULL) {
heap_close(rel);
continue;
}
rules = rel->rd_rules;
locks = NIL;
/*
* Collect the RIR rules that we must apply
*/
for (i = 0; i < rules->numLocks; i++) {
rule = rules->rules[i];
if (rule->event != CMD_SELECT)
continue;
if (rule->attrno > 0 &&
!attribute_used((Node *)parsetree,
rt_index,
rule->attrno, 0))
continue;
locks = lappend(locks, rule);
}
/*
* Check permissions
*/
checkLockPerms(locks, parsetree, rt_index);
/*
* Now apply them
*/
foreach (l, locks) {
rule = lfirst(l);
RIRonly.event = rule->event;
RIRonly.attrno = rule->attrno;
RIRonly.qual = rule->qual;
RIRonly.actions = rule->actions;
ApplyRetrieveRule(parsetree,
&RIRonly,
rt_index,
RIRonly.attrno == -1,
rel,
&modified);
}
heap_close(rel);
}
fireRIRonSubselect((Node *) parsetree);
modifyAggregQual((Node **) &(parsetree->qual), parsetree);
return parsetree;
}
/*
* idea is to fire regular rules first, then qualified instead
* rules and unqualified instead rules last. Any lemming is counted for.
*/
static List *
orderRules(List *locks)
{
List *regular = NIL;
List *instead_rules = NIL;
List *instead_qualified = NIL;
List *i;
foreach(i, locks)
{
RewriteRule *rule_lock = (RewriteRule *) lfirst(i);
if (rule_lock->isInstead)
{
if (rule_lock->qual == NULL)
instead_rules = lappend(instead_rules, rule_lock);
else
instead_qualified = lappend(instead_qualified, rule_lock);
}
else
regular = lappend(regular, rule_lock);
}
regular = nconc(regular, instead_qualified);
return nconc(regular, instead_rules);
}
static Query *
CopyAndAddQual(Query *parsetree,
List *actions,
Node *rule_qual,
int rt_index,
CmdType event)
{
Query *new_tree = (Query *) copyObject(parsetree);
Node *new_qual = NULL;
Query *rule_action = NULL;
if (actions)
rule_action = lfirst(actions);
if (rule_qual != NULL)
new_qual = (Node *) copyObject(rule_qual);
if (rule_action != NULL)
{
List *rtable;
int rt_length;
rtable = new_tree->rtable;
rt_length = length(rtable);
rtable = append(rtable, listCopy(rule_action->rtable));
new_tree->rtable = rtable;
OffsetVarNodes(new_qual, rt_length);
1998-01-21 05:24:46 +01:00
ChangeVarNodes(new_qual, PRS2_CURRENT_VARNO + rt_length, rt_index, 0);
}
/* XXX -- where current doesn't work for instead nothing.... yet */
AddNotQual(new_tree, new_qual);
return new_tree;
}
/*
* fireRules -
* Iterate through rule locks applying rules.
* All rules create their own parsetrees. Instead rules
* with rule qualification save the original parsetree
* and add their negated qualification to it. Real instead
* rules finally throw away the original parsetree.
*
* remember: reality is for dead birds -- glass
*
*/
static List *
fireRules(Query *parsetree,
int rt_index,
CmdType event,
bool *instead_flag,
List *locks,
List **qual_products)
{
RewriteInfo *info;
List *results = NIL;
List *i;
/* choose rule to fire from list of rules */
if (locks == NIL)
{
return NIL;
}
locks = orderRules(locks); /* real instead rules last */
foreach(i, locks)
{
RewriteRule *rule_lock = (RewriteRule *) lfirst(i);
Node *qual,
*event_qual;
List *actions;
List *r;
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
/*
* Instead rules change the resultRelation of the query. So the
* permission checks on the initial resultRelation would never be
* done (this is normally done in the executor deep down). So we
* must do it here. The result relations resulting from earlier
* rewrites are already checked against the rules eventrelation
* owner (during matchLocks) and have the skipAcl flag set.
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 (rule_lock->isInstead &&
parsetree->commandType != CMD_SELECT)
{
RangeTblEntry *rte;
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
int32 acl_rc;
int32 reqperm;
switch (parsetree->commandType)
{
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 CMD_INSERT:
reqperm = ACL_AP;
break;
default:
reqperm = ACL_WR;
break;
}
rte = (RangeTblEntry *) nth(parsetree->resultRelation - 1,
parsetree->rtable);
if (!rte->skipAcl)
{
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
acl_rc = pg_aclcheck(rte->relname,
GetPgUserName(), reqperm);
if (acl_rc != ACLCHECK_OK)
{
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
elog(ERROR, "%s: %s",
rte->relname,
aclcheck_error_strings[acl_rc]);
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
}
}
}
/* multiple rule action time */
*instead_flag = rule_lock->isInstead;
event_qual = rule_lock->qual;
actions = rule_lock->actions;
if (event_qual != NULL && *instead_flag)
{
Query *qual_product;
RewriteInfo qual_info;
/* ----------
* If there are instead rules with qualifications,
* the original query is still performed. But all
* the negated rule qualifications of the instead
* rules are added so it does it's actions only
* in cases where the rule quals of all instead
* rules are false. Think of it as the default
* action in a case. We save this in *qual_products
* so deepRewriteQuery() can add it to the query
* list after we mangled it up enough.
* ----------
*/
if (*qual_products == NIL)
qual_product = parsetree;
else
qual_product = (Query *) nth(0, *qual_products);
qual_info.event = qual_product->commandType;
qual_info.new_varno = length(qual_product->rtable) + 2;
qual_product = CopyAndAddQual(qual_product,
actions,
event_qual,
rt_index,
event);
qual_info.rule_action = qual_product;
if (event == CMD_INSERT || event == CMD_UPDATE)
FixNew(&qual_info, qual_product);
*qual_products = lappend(NIL, qual_product);
}
foreach(r, actions)
{
Query *rule_action = lfirst(r);
Node *rule_qual = copyObject(event_qual);
if (rule_action->commandType == CMD_NOTHING)
continue;
/*--------------------------------------------------
* Step 1:
* Rewrite current.attribute or current to tuple variable
* this appears to be done in parser?
*--------------------------------------------------
*/
info = gatherRewriteMeta(parsetree, rule_action, rule_qual,
rt_index, event, instead_flag);
/* handle escapable cases, or those handled by other code */
if (info->nothing)
{
if (*instead_flag)
return NIL;
else
continue;
}
if (info->action == info->event &&
info->event == CMD_SELECT)
continue;
/*
* Event Qualification forces copying of parsetree and
* splitting into two queries one w/rule_qual, one w/NOT
* rule_qual. Also add user query qual onto rule action
*/
qual = parsetree->qual;
AddQual(info->rule_action, qual);
if (info->rule_qual != NULL)
AddQual(info->rule_action, info->rule_qual);
/*--------------------------------------------------
* Step 2:
* Rewrite new.attribute w/ right hand side of target-list
* entry for appropriate field name in insert/update
*--------------------------------------------------
*/
if ((info->event == CMD_INSERT) || (info->event == CMD_UPDATE))
FixNew(info, parsetree);
/*--------------------------------------------------
* Step 3:
* rewriting due to retrieve rules
*--------------------------------------------------
*/
info->rule_action->rtable = info->rt;
/*
ProcessRetrieveQuery(info->rule_action, info->rt,
&orig_instead_flag, TRUE);
*/
/*--------------------------------------------------
* Step 4
* Simplify? hey, no algorithm for simplification... let
* the planner do it.
*--------------------------------------------------
*/
results = lappend(results, info->rule_action);
pfree(info);
}
/* ----------
* If this was an unqualified instead rule,
* throw away an eventually saved 'default' parsetree
* ----------
*/
if (event_qual == NULL && *instead_flag)
*qual_products = NIL;
}
return results;
}
static List *
RewriteQuery(Query *parsetree, bool *instead_flag, List **qual_products)
{
CmdType event;
List *product_queries = NIL;
int result_relation = 0;
RangeTblEntry *rt_entry;
Relation rt_entry_relation = NULL;
RuleLock *rt_entry_locks = NULL;
Assert(parsetree != NULL);
event = parsetree->commandType;
/*
* SELECT rules are handled later when we have all the
* queries that should get executed
*/
if (event == CMD_SELECT)
return NIL;
/*
* Utilities aren't rewritten at all - why is this here?
*/
if (event == CMD_UTILITY)
return NIL;
/*
* only for a delete may the targetlist be NULL
*/
if (event != CMD_DELETE)
Assert(parsetree->targetList != NULL);
result_relation = parsetree->resultRelation;
/*
* the statement is an update, insert or delete - fire rules
* on it.
*/
rt_entry = rt_fetch(result_relation, parsetree->rtable);
rt_entry_relation = heap_openr(rt_entry->relname);
rt_entry_locks = rt_entry_relation->rd_rules;
heap_close(rt_entry_relation);
if (rt_entry_locks != NULL)
{
List *locks =
matchLocks(event, rt_entry_locks, result_relation, parsetree);
product_queries =
fireRules(parsetree,
result_relation,
event,
instead_flag,
locks,
qual_products);
}
return product_queries;
1998-01-09 06:48:22 +01:00
}
/*
* to avoid infinite recursion, we restrict the number of times a query
* can be rewritten. Detecting cycles is left for the reader as an excercise.
*/
#ifndef REWRITE_INVOKE_MAX
#define REWRITE_INVOKE_MAX 10
#endif
static int numQueryRewriteInvoked = 0;
/*
* deepRewriteQuery -
* rewrites the query and apply the rules again on the queries rewritten
*/
static List *
deepRewriteQuery(Query *parsetree)
{
List *n;
List *rewritten = NIL;
List *result = NIL;
bool instead;
List *qual_products = NIL;
if (++numQueryRewriteInvoked > REWRITE_INVOKE_MAX)
{
elog(ERROR, "query rewritten %d times, may contain cycles",
numQueryRewriteInvoked - 1);
}
instead = FALSE;
result = RewriteQuery(parsetree, &instead, &qual_products);
foreach(n, result)
{
Query *pt = lfirst(n);
List *newstuff = NIL;
newstuff = deepRewriteQuery(pt);
if (newstuff != NIL)
rewritten = nconc(rewritten, newstuff);
}
/* ----------
* qual_products are the original query with the negated
* rule qualification of an instead rule
* ----------
*/
if (qual_products != NIL)
rewritten = nconc(rewritten, qual_products);
/* ----------
* The original query is appended last if not instead
* because update and delete rule actions might not do
* anything if they are invoked after the update or
* delete is performed. The command counter increment
* between the query execution makes the deleted (and
* maybe the updated) tuples disappear so the scans
* for them in the rule actions cannot find them.
* ----------
*/
if (!instead)
rewritten = lappend(rewritten, parsetree);
return rewritten;
}
/*
* QueryOneRewrite -
* rewrite one query
*/
static List *
QueryRewriteOne(Query *parsetree)
{
numQueryRewriteInvoked = 0;
/*
* take a deep breath and apply all the rewrite rules - ay
*/
return deepRewriteQuery(parsetree);
}
/* ----------
* RewritePreprocessQuery -
* adjust details in the parsetree, the rule system
* depends on
* ----------
*/
static void
RewritePreprocessQuery(Query *parsetree)
{
/* ----------
* if the query has a resultRelation, reassign the
* result domain numbers to the attribute numbers in the
* target relation. FixNew() depends on it when replacing
* *new* references in a rule action by the expressions
* from the rewritten query.
* ----------
*/
if (parsetree->resultRelation > 0)
{
RangeTblEntry *rte;
Relation rd;
List *tl;
TargetEntry *tle;
int resdomno;
rte = (RangeTblEntry *) nth(parsetree->resultRelation - 1,
parsetree->rtable);
rd = heap_openr(rte->relname);
foreach(tl, parsetree->targetList)
{
tle = (TargetEntry *) lfirst(tl);
resdomno = attnameAttNum(rd, tle->resdom->resname);
tle->resdom->resno = resdomno;
}
heap_close(rd);
}
}
/*
* QueryRewrite -
* rewrite one query via query rewrite system, possibly returning 0
* or many queries
*/
List *
QueryRewrite(Query *parsetree)
{
List *querylist;
List *results = NIL;
List *l;
Query *query;
/*
* Step 1
*
* There still seems something broken with the resdom numbers
* so we reassign them first.
*/
RewritePreprocessQuery(parsetree);
/*
* Step 2
*
* Apply all non-SELECT rules possibly getting 0 or many queries
*/
querylist = QueryRewriteOne(parsetree);
/*
* Step 3
*
* Apply all the RIR rules on each query
*/
foreach (l, querylist) {
query = (Query *)lfirst(l);
results = lappend(results, fireRIRrules(query));
}
return results;
}