rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
postgresql/src/backend/commands/user.c

1189 lines
33 KiB
C
Raw Normal View History

Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
/*-------------------------------------------------------------------------
*
Change my-function-name-- to my_function_name, and optimizer renames.
1999-02-14 00:22:53 +01:00
* user.c
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Commands for manipulating roles (formerly called users).
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
*
Tag appropriate files for rc3 Also performed an initial run through of upgrading our Copyright date to extend to 2005 ... first run here was very simple ... change everything where: grep 1996-2004 && the word 'Copyright' ... scanned through the generated list with 'less' first, and after, to make sure that I only picked up the right entries ...
2004-12-31 23:04:05 +01:00
* Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
Add: * Portions Copyright (c) 1996-2000, PostgreSQL, Inc to all files copyright Regents of Berkeley. Man, that's a lot of files.
2000-01-26 06:58:53 +01:00
* Portions Copyright (c) 1994, Regents of the University of California
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
*
Fix up problems in write_auth_file and parsing of the auth file. In particular, make hba.c cope with zero-length tokens, which it never did properly before. Also, enforce rolcanlogin.
2005-06-29 00:16:45 +02:00
* $PostgreSQL: pgsql/src/backend/commands/user.c,v 1.154 2005/06/28 22:16:44 tgl Exp $
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
*
*-------------------------------------------------------------------------
*/
Clean up various to-do items associated with system indexes: pg_database now has unique indexes on oid and on datname. pg_shadow now has unique indexes on usename and on usesysid. pg_am now has unique index on oid. pg_opclass now has unique index on oid. pg_amproc now has unique index on amid+amopclaid+amprocnum. Remove pg_rewrite's unnecessary index on oid, delete unused RULEOID syscache. Remove index on pg_listener and associated syscache for performance reasons (caching rows that are certain to change before you need 'em again is rather pointless). Change pg_attrdef's nonunique index on adrelid into a unique index on adrelid+adnum. Fix various incorrect settings of pg_class.relisshared, make that the primary reference point for whether a relation is shared or not. IsSharedSystemRelationName() is now only consulted to initialize relisshared during initial creation of tables and indexes. In theory we might now support shared user relations, though it's not clear how one would get entries for them into pg_class &etc of multiple databases. Fix recently reported bug that pg_attribute rows created for an index all have the same OID. (Proof that non-unique OID doesn't matter unless it's actually used to do lookups ;-)) There's no need to treat pg_trigger, pg_attrdef, pg_relcheck as bootstrap relations. Convert them into plain system catalogs without hardwired entries in pg_class and friends. Unify global.bki and template1.bki into a single init script postgres.bki, since the alleged distinction between them was misleading and pointless. Not to mention that it didn't work for setting up indexes on shared system relations. Rationalize locking of pg_shadow, pg_group, pg_attrdef (no need to use AccessExclusiveLock where ExclusiveLock or even RowExclusiveLock will do). Also, hold locks until transaction commit where necessary.
2001-06-12 07:55:50 +02:00
#include "postgres.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "access/heapam.h"
Group-manipulating code neglected to fill in array element type, which is now required.
2002-08-30 03:01:02 +02:00
#include "catalog/indexing.h"
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
#include "catalog/pg_auth_members.h"
#include "catalog/pg_authid.h"
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "catalog/pg_database.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "commands/user.h"
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "libpq/crypt.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "miscadmin.h"
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
#include "utils/builtins.h"
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
#include "utils/catcache.h"
Add code to prevent transaction ID wraparound by enforcing a safe limit in GetNewTransactionId(). Since the limit value has to be computed before we run any real transactions, this requires adding code to database startup to scan pg_database and determine the oldest datfrozenxid. This can conveniently be combined with the first stage of an attack on the problem that the 'flat file' copies of pg_shadow and pg_group are not properly updated during WAL recovery. The code I've added to startup resides in a new file src/backend/utils/init/flatfiles.c, and it is responsible for rewriting the flat files as well as initializing the XID wraparound limit value. This will eventually allow us to get rid of GetRawDatabaseInfo too, but we'll need an initdb so we can add a trigger to pg_database.
2005-02-20 03:22:07 +01:00
#include "utils/flatfiles.h"
First round of changes for new fmgr interface. fmgr itself and the key call sites are changed, but most called functions are still oldstyle. An exception is that the PL managers are updated (so, for example, NULL handling now behaves as expected in plperl and plpgsql functions). NOTE initdb is forced due to added column in pg_proc.
2000-05-28 19:56:29 +02:00
#include "utils/fmgroids.h"
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
#include "utils/guc.h"
has_table_privilege functions from Joe Conway (with some kibitzing from Tom Lane). For the moment, only the OID/name variants are provided. I didn't force initdb, but the additions to the 'privileges' regress test won't pass until you do one.
2001-06-14 03:09:22 +02:00
#include "utils/lsyscache.h"
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "utils/syscache.h"
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
Clean up various to-do items associated with system indexes: pg_database now has unique indexes on oid and on datname. pg_shadow now has unique indexes on usename and on usesysid. pg_am now has unique index on oid. pg_opclass now has unique index on oid. pg_amproc now has unique index on amid+amopclaid+amprocnum. Remove pg_rewrite's unnecessary index on oid, delete unused RULEOID syscache. Remove index on pg_listener and associated syscache for performance reasons (caching rows that are certain to change before you need 'em again is rather pointless). Change pg_attrdef's nonunique index on adrelid into a unique index on adrelid+adnum. Fix various incorrect settings of pg_class.relisshared, make that the primary reference point for whether a relation is shared or not. IsSharedSystemRelationName() is now only consulted to initialize relisshared during initial creation of tables and indexes. In theory we might now support shared user relations, though it's not clear how one would get entries for them into pg_class &etc of multiple databases. Fix recently reported bug that pg_attribute rows created for an index all have the same OID. (Proof that non-unique OID doesn't matter unless it's actually used to do lookups ;-)) There's no need to treat pg_trigger, pg_attrdef, pg_relcheck as bootstrap relations. Convert them into plain system catalogs without hardwired entries in pg_class and friends. Unify global.bki and template1.bki into a single init script postgres.bki, since the alleged distinction between them was misleading and pointless. Not to mention that it didn't work for setting up indexes on shared system relations. Rationalize locking of pg_shadow, pg_group, pg_attrdef (no need to use AccessExclusiveLock where ExclusiveLock or even RowExclusiveLock will do). Also, hold locks until transaction commit where necessary.
2001-06-12 07:55:50 +02:00
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
extern bool Password_encryption;
Many more cleanups...
1998-12-14 07:50:32 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
static List *roleNamesToIds(List *memberNames);
static void AddRoleMems(const char *rolename, Oid roleid,
List *memberNames, List *memberIds,
Oid grantorId, bool admin_opt);
static void DelRoleMems(const char *rolename, Oid roleid,
List *memberNames, List *memberIds,
bool admin_opt);
Authentication improvements: A new pg_hba.conf column, USER Allow specifiction of lists of users separated by commas Allow group names specified by + Allow include files containing lists of users specified by @ Allow lists of databases, and database files Allow samegroup in database column to match group name matching dbname Removal of secondary password files Remove pg_passwd utility Lots of code cleanup in user.c and hba.c New data/global/pg_pwd format New data/global/pg_group file
2002-04-04 06:25:54 +02:00
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* CREATE ROLE
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
*/
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
CreateRole(CreateRoleStmt *stmt)
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Relation pg_authid_rel;
TupleDesc pg_authid_dsc;
pgindent run over code.
1999-05-25 18:15:34 +02:00
HeapTuple tuple;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Datum new_record[Natts_pg_authid];
char new_record_nulls[Natts_pg_authid];
Oid roleid;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
ListCell *item;
ListCell *option;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
char *password = NULL; /* user password */
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
bool encrypt_password = Password_encryption; /* encrypt password? */
char encrypted_password[MD5_PASSWD_LEN + 1];
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
bool issuper = false; /* Make the user a superuser? */
bool createrole = false; /* Can this user create roles? */
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
bool createdb = false; /* Can the user create databases? */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
bool canlogin = false; /* Can this user login? */
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
List *addroleto = NIL; /* roles to make this a member of */
List *rolemembers = NIL; /* roles to be members of this role */
List *adminmembers = NIL; /* roles to be admins of this role */
char *validUntil = NULL; /* time the login is valid until */
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
DefElem *dpassword = NULL;
DefElem *dcreatedb = NULL;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
DefElem *dcreaterole = NULL;
DefElem *dcanlogin = NULL;
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
DefElem *daddroleto = NULL;
DefElem *drolemembers = NULL;
DefElem *dadminmembers = NULL;
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
DefElem *dvalidUntil = NULL;
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
/* Extract options from the statement node tree */
foreach(option, stmt->options)
{
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
DefElem *defel = (DefElem *) lfirst(option);
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
Replace useless strcasecmp's by strcmp's.
2001-09-19 11:48:42 +02:00
if (strcmp(defel->defname, "password") == 0 ||
strcmp(defel->defname, "encryptedPassword") == 0 ||
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
strcmp(defel->defname, "unencryptedPassword") == 0)
{
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
if (dpassword)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
dpassword = defel;
Replace useless strcasecmp's by strcmp's.
2001-09-19 11:48:42 +02:00
if (strcmp(defel->defname, "encryptedPassword") == 0)
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
encrypt_password = true;
Replace useless strcasecmp's by strcmp's.
2001-09-19 11:48:42 +02:00
else if (strcmp(defel->defname, "unencryptedPassword") == 0)
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
encrypt_password = false;
}
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
else if (strcmp(defel->defname, "sysid") == 0)
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
ereport(WARNING,
(errmsg("SYSID can no longer be specified")));
}
else if (strcmp(defel->defname, "createrole") == 0)
{
if (dcreaterole)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
dcreaterole = defel;
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
}
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
else if (strcmp(defel->defname, "createdb") == 0)
{
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
if (dcreatedb)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
dcreatedb = defel;
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
else if (strcmp(defel->defname, "canlogin") == 0)
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (dcanlogin)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
dcanlogin = defel;
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
}
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
else if (strcmp(defel->defname, "addroleto") == 0)
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
{
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
if (daddroleto)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
daddroleto = defel;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
}
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
else if (strcmp(defel->defname, "rolemembers") == 0)
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
{
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
if (drolemembers)
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
drolemembers = defel;
}
else if (strcmp(defel->defname, "adminmembers") == 0)
{
if (dadminmembers)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
dadminmembers = defel;
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
}
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
else if (strcmp(defel->defname, "validUntil") == 0)
{
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
if (dvalidUntil)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
dvalidUntil = defel;
}
else
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
elog(ERROR, "option \"%s\" not recognized",
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
defel->defname);
}
if (dcreatedb)
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
createdb = intVal(dcreatedb->arg) != 0;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (dcreaterole)
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
createrole = intVal(dcreaterole->arg) != 0;
/* XXX issuper is implied by createrole for now */
issuper = createrole;
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (dcanlogin)
canlogin = intVal(dcanlogin->arg) != 0;
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
if (dvalidUntil)
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
validUntil = strVal(dvalidUntil->arg);
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
if (dpassword)
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
password = strVal(dpassword->arg);
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
if (daddroleto)
addroleto = (List *) daddroleto->arg;
if (drolemembers)
rolemembers = (List *) drolemembers->arg;
if (dadminmembers)
adminmembers = (List *) dadminmembers->arg;
Repair problems with omitted password and VALID UNTIL parameters in CREATE USER.
1999-04-02 08:16:36 +02:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/* Check some permissions first */
if (!superuser())
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("must be superuser to create roles")));
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (strcmp(stmt->role, "public") == 0)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_RESERVED_NAME),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role name \"%s\" is reserved",
stmt->role)));
Make PUBLIC an unreserved word (in fact, not a keyword at all), per previous discussion.
2002-04-18 23:16:16 +02:00
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Check the pg_authid relation to be certain the role doesn't
* already exist. Note we secure exclusive lock because
* we need to protect our eventual update of the flat auth file.
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
pg_authid_rel = heap_open(AuthIdRelationId, ExclusiveLock);
pg_authid_dsc = RelationGetDescr(pg_authid_rel);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
tuple = SearchSysCache(AUTHNAME,
PointerGetDatum(stmt->role),
0, 0, 0);
if (HeapTupleIsValid(tuple))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_DUPLICATE_OBJECT),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" already exists",
stmt->role)));
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
* Build a tuple to insert
*/
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
MemSet(new_record, 0, sizeof(new_record));
MemSet(new_record_nulls, ' ', sizeof(new_record_nulls));
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolname - 1] =
DirectFunctionCall1(namein, CStringGetDatum(stmt->role));
new_record[Anum_pg_authid_rolsuper - 1] = BoolGetDatum(issuper);
new_record[Anum_pg_authid_rolcreaterole - 1] = BoolGetDatum(createrole);
new_record[Anum_pg_authid_rolcreatedb - 1] = BoolGetDatum(createdb);
/* superuser gets catupdate right by default */
new_record[Anum_pg_authid_rolcatupdate - 1] = BoolGetDatum(issuper);
new_record[Anum_pg_authid_rolcanlogin - 1] = BoolGetDatum(canlogin);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (password)
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
{
if (!encrypt_password || isMD5(password))
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolpassword - 1] =
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
DirectFunctionCall1(textin, CStringGetDatum(password));
else
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (!EncryptMD5(password, stmt->role, strlen(stmt->role),
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
encrypted_password))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
elog(ERROR, "password encryption failed");
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolpassword - 1] =
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
DirectFunctionCall1(textin, CStringGetDatum(encrypted_password));
}
}
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
else
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record_nulls[Anum_pg_authid_rolpassword - 1] = 'n';
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (validUntil)
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolvaliduntil - 1] =
DirectFunctionCall3(timestamptz_in,
CStringGetDatum(validUntil),
ObjectIdGetDatum(InvalidOid),
Int32GetDatum(-1));
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
else
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record_nulls[Anum_pg_authid_rolvaliduntil - 1] = 'n';
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record_nulls[Anum_pg_authid_rolconfig - 1] = 'n';
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
tuple = heap_formtuple(pg_authid_dsc, new_record, new_record_nulls);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Insert new record in the pg_authid table
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
roleid = simple_heap_insert(pg_authid_rel, tuple);
Assert(OidIsValid(roleid));
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Restructure system-catalog index updating logic. Instead of having hardwired lists of index names for each catalog, use the relcache's mechanism for caching lists of OIDs of indexes of any table. This reduces the common case of updating system catalog indexes to a single line, makes it much easier to add a new system index (in fact, you can now do so on-the-fly if you want to), and as a nice side benefit improves performance a little. Per recent pghackers discussion.
2002-08-05 05:29:17 +02:00
/* Update indexes */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
CatalogUpdateIndexes(pg_authid_rel, tuple);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Add the new role to the specified existing roles.
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
*/
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
foreach(item, addroleto)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
char *oldrolename = strVal(lfirst(item));
Oid oldroleid = get_roleid_checked(oldrolename);
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
AddRoleMems(oldrolename, oldroleid,
list_make1(makeString(stmt->role)),
list_make1_oid(roleid),
GetUserId(), false);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
/*
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
* Add the specified members to this new role. adminmembers get the
* admin option, rolemembers don't.
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
*/
AddRoleMems(stmt->role, roleid,
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
adminmembers, roleNamesToIds(adminmembers),
GetUserId(), true);
AddRoleMems(stmt->role, roleid,
rolemembers, roleNamesToIds(rolemembers),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
GetUserId(), false);
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
/*
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
* Now we can clean up; but keep lock until commit (to avoid possible
* deadlock when commit code tries to acquire lock).
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
heap_close(pg_authid_rel, NoLock);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Set flag to update flat auth file at commit.
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
auth_file_update_needed();
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
}
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* ALTER ROLE
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
*/
Move renametrig() from tablecmds.c to trigger.c --- if we're going to divide backend/commands by object type, let's try to pay at least minimal attention to respecting that structure, eh? Also reorder the contents of tablecmds.c; it seems odd to me to put ALTER commands before creation/deletion commands.
2002-04-26 21:29:47 +02:00
void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
AlterRole(AlterRoleStmt *stmt)
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Datum new_record[Natts_pg_authid];
char new_record_nulls[Natts_pg_authid];
char new_record_repl[Natts_pg_authid];
Relation pg_authid_rel;
TupleDesc pg_authid_dsc;
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
HeapTuple tuple,
new_tuple;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
ListCell *option;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
char *password = NULL; /* user password */
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
bool encrypt_password = Password_encryption; /* encrypt password? */
char encrypted_password[MD5_PASSWD_LEN + 1];
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
int issuper = -1; /* Make the user a superuser? */
int createrole = -1; /* Can this user create roles? */
int createdb = -1; /* Can the user create databases? */
int canlogin = -1; /* Can this user login? */
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
List *rolemembers = NIL; /* roles to be added/removed */
char *validUntil = NULL; /* time the login is valid until */
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
DefElem *dpassword = NULL;
DefElem *dcreatedb = NULL;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
DefElem *dcreaterole = NULL;
DefElem *dcanlogin = NULL;
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
DefElem *dvalidUntil = NULL;
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
DefElem *drolemembers = NULL;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Oid roleid;
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
/* Extract options from the statement node tree */
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
foreach(option, stmt->options)
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
{
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
DefElem *defel = (DefElem *) lfirst(option);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Replace useless strcasecmp's by strcmp's.
2001-09-19 11:48:42 +02:00
if (strcmp(defel->defname, "password") == 0 ||
strcmp(defel->defname, "encryptedPassword") == 0 ||
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
strcmp(defel->defname, "unencryptedPassword") == 0)
{
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (dpassword)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
dpassword = defel;
Replace useless strcasecmp's by strcmp's.
2001-09-19 11:48:42 +02:00
if (strcmp(defel->defname, "encryptedPassword") == 0)
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
encrypt_password = true;
Replace useless strcasecmp's by strcmp's.
2001-09-19 11:48:42 +02:00
else if (strcmp(defel->defname, "unencryptedPassword") == 0)
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
encrypt_password = false;
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
}
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
else if (strcmp(defel->defname, "createdb") == 0)
{
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (dcreatedb)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
dcreatedb = defel;
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
else if (strcmp(defel->defname, "createrole") == 0)
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (dcreaterole)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
dcreaterole = defel;
}
else if (strcmp(defel->defname, "canlogin") == 0)
{
if (dcanlogin)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
dcanlogin = defel;
}
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
else if (strcmp(defel->defname, "validUntil") == 0)
{
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (dvalidUntil)
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
dvalidUntil = defel;
}
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
else if (strcmp(defel->defname, "rolemembers") == 0 &&
stmt->action != 0)
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
{
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
if (drolemembers)
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("conflicting or redundant options")));
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
drolemembers = defel;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
}
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
else
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
elog(ERROR, "option \"%s\" not recognized",
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
defel->defname);
}
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (dcreatedb)
createdb = intVal(dcreatedb->arg);
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (dcreaterole)
{
createrole = intVal(dcreaterole->arg);
/* XXX createrole implies issuper for now */
issuper = createrole;
}
if (dcanlogin)
canlogin = intVal(dcanlogin->arg);
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (dvalidUntil)
validUntil = strVal(dvalidUntil->arg);
if (dpassword)
password = strVal(dpassword->arg);
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
if (drolemembers)
rolemembers = (List *) drolemembers->arg;
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/* must be superuser or just want to change your own password */
if (!superuser() &&
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
!(issuper < 0 &&
createrole < 0 &&
createdb < 0 &&
canlogin < 0 &&
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
!validUntil &&
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
!rolemembers &&
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
password &&
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
strcmp(GetUserNameFromId(GetUserId()), stmt->role) == 0))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("permission denied")));
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Scan the pg_authid relation to be certain the user exists. Note we
* secure exclusive lock to protect our update of the flat auth file.
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
pg_authid_rel = heap_open(AuthIdRelationId, ExclusiveLock);
pg_authid_dsc = RelationGetDescr(pg_authid_rel);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
tuple = SearchSysCache(AUTHNAME,
PointerGetDatum(stmt->role),
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
0, 0, 0);
heap_fetch requires buffer pointer, must be released; heap_getnext no longer returns buffer pointer, can be gotten from scan; descriptor; bootstrap can create multi-key indexes; pg_procname index now is multi-key index; oidint2, oidint4, oidname are gone (must be removed from regression tests); use System Cache rather than sequential scan in many places; heap_modifytuple no longer takes buffer parameter; remove unused buffer parameter in a few other functions; oid8 is not index-able; remove some use of single-character variable names; cleanup Buffer variables usage and scan descriptor looping; cleaned up allocation and freeing of tuples; 18k lines of diff;
1998-08-19 04:04:17 +02:00
if (!HeapTupleIsValid(tuple))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" does not exist", stmt->role)));
roleid = HeapTupleGetOid(tuple);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
* Build an updated tuple, perusing the information just obtained
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
MemSet(new_record, 0, sizeof(new_record));
MemSet(new_record_nulls, ' ', sizeof(new_record_nulls));
MemSet(new_record_repl, ' ', sizeof(new_record_repl));
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolname - 1] = DirectFunctionCall1(namein,
CStringGetDatum(stmt->role));
new_record_repl[Anum_pg_authid_rolname - 1] = 'r';
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Make the world somewhat safe for (not from) DELETE FROM pg_shadow; Assign the fixed user id 1 to the user created by initdb. A stand-alone backend will always set the user id to 1. (Consequently, the name of that user is no longer important.) In stand-alone mode, the user id 1 will have implicit superuser status, to allow repairs even if there are no users defined. Print a warning message when starting in stand-alone mode when no users are defined. Disallow dropping the current user and session user. Granting/revoking superuser status also grants/revokes usecatupd. (Previously, it would never grant it back. This could lead to "deadlocks".) CREATE USER and CREATE GROUP will start allocating user ids at 100 (unless explicitly specified), to prevent accidental creation of a superuser (plus some room for future extensions).
2001-09-08 17:24:00 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* issuper/createrole/catupdate/etc
Make the world somewhat safe for (not from) DELETE FROM pg_shadow; Assign the fixed user id 1 to the user created by initdb. A stand-alone backend will always set the user id to 1. (Consequently, the name of that user is no longer important.) In stand-alone mode, the user id 1 will have implicit superuser status, to allow repairs even if there are no users defined. Print a warning message when starting in stand-alone mode when no users are defined. Disallow dropping the current user and session user. Granting/revoking superuser status also grants/revokes usecatupd. (Previously, it would never grant it back. This could lead to "deadlocks".) CREATE USER and CREATE GROUP will start allocating user ids at 100 (unless explicitly specified), to prevent accidental creation of a superuser (plus some room for future extensions).
2001-09-08 17:24:00 +02:00
*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* XXX It's rather unclear how to handle catupdate. It's probably best to
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
* keep it equal to the superuser status, otherwise you could end up
* with a situation where no existing superuser can alter the
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* catalogs, including pg_authid!
Make the world somewhat safe for (not from) DELETE FROM pg_shadow; Assign the fixed user id 1 to the user created by initdb. A stand-alone backend will always set the user id to 1. (Consequently, the name of that user is no longer important.) In stand-alone mode, the user id 1 will have implicit superuser status, to allow repairs even if there are no users defined. Print a warning message when starting in stand-alone mode when no users are defined. Disallow dropping the current user and session user. Granting/revoking superuser status also grants/revokes usecatupd. (Previously, it would never grant it back. This could lead to "deadlocks".) CREATE USER and CREATE GROUP will start allocating user ids at 100 (unless explicitly specified), to prevent accidental creation of a superuser (plus some room for future extensions).
2001-09-08 17:24:00 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (issuper >= 0)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolsuper - 1] = BoolGetDatum(issuper > 0);
new_record_repl[Anum_pg_authid_rolsuper - 1] = 'r';
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolcatupdate - 1] = BoolGetDatum(issuper > 0);
new_record_repl[Anum_pg_authid_rolcatupdate - 1] = 'r';
}
if (createrole >= 0)
{
new_record[Anum_pg_authid_rolcreaterole - 1] = BoolGetDatum(createrole > 0);
new_record_repl[Anum_pg_authid_rolcreaterole - 1] = 'r';
}
if (createdb >= 0)
{
new_record[Anum_pg_authid_rolcreatedb - 1] = BoolGetDatum(createdb > 0);
new_record_repl[Anum_pg_authid_rolcreatedb - 1] = 'r';
}
if (canlogin >= 0)
{
new_record[Anum_pg_authid_rolcanlogin - 1] = BoolGetDatum(canlogin > 0);
new_record_repl[Anum_pg_authid_rolcanlogin - 1] = 'r';
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
/* password */
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (password)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
if (!encrypt_password || isMD5(password))
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolpassword - 1] =
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
DirectFunctionCall1(textin, CStringGetDatum(password));
else
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (!EncryptMD5(password, stmt->role, strlen(stmt->role),
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
encrypted_password))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
elog(ERROR, "password encryption failed");
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolpassword - 1] =
Use MD5 for wire protocol encryption for >= 7.2 client/server. Allow pg_shadow to be MD5 encrypted. Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user. Add password_encryption postgresql.conf option. Update wire protocol version to 2.1.
2001-08-15 20:42:16 +02:00
DirectFunctionCall1(textin, CStringGetDatum(encrypted_password));
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record_repl[Anum_pg_authid_rolpassword - 1] = 'r';
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
/* valid until */
Changes from Vince Vielhaber to allow the optional clauses of CREATE USER and ALTER USER to appear in any order, not only the fixed order they used to be required to appear in. Also, some changes from Tom Lane to create a FULL option for VACUUM; it doesn't do anything yet, but I needed to change many of the same files to make that happen, so now seemed like a good time.
2001-07-11 00:09:29 +02:00
if (validUntil)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_authid_rolvaliduntil - 1] =
DirectFunctionCall3(timestamptz_in,
CStringGetDatum(validUntil),
ObjectIdGetDatum(InvalidOid),
Int32GetDatum(-1));
new_record_repl[Anum_pg_authid_rolvaliduntil - 1] = 'r';
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_tuple = heap_modifytuple(tuple, pg_authid_dsc, new_record,
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
new_record_nulls, new_record_repl);
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
simple_heap_update(pg_authid_rel, &tuple->t_self, new_tuple);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/* Update indexes */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
CatalogUpdateIndexes(pg_authid_rel, new_tuple);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
ReleaseSysCache(tuple);
heap_freetuple(new_tuple);
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
/*
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
* Now we can clean up; but keep lock until commit (to avoid possible
* deadlock when commit code tries to acquire lock).
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
heap_close(pg_authid_rel, NoLock);
if (stmt->action == +1) /* add members to role */
AddRoleMems(stmt->role, roleid,
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
rolemembers, roleNamesToIds(rolemembers),
GetUserId(), false);
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
else if (stmt->action == -1) /* drop members from role */
DelRoleMems(stmt->role, roleid,
Bring syntax of role-related commands into SQL compliance. To avoid syntactic conflicts, both privilege and role GRANT/REVOKE commands have to use the same production for scanning the list of tokens that might eventually turn out to be privileges or role names. So, change the existing GRANT/REVOKE code to expect a list of strings not pre-reduced AclMode values. Fix a couple other minor issues while at it, such as InitializeAcl function name conflicting with a Windows system function.
2005-06-28 21:51:26 +02:00
rolemembers, roleNamesToIds(rolemembers),
false);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Set flag to update flat auth file at commit.
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
auth_file_update_needed();
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
}
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* ALTER ROLE ... SET
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
*/
void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
AlterRoleSet(AlterRoleSetStmt *stmt)
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
{
char *valuestr;
HeapTuple oldtuple,
newtuple;
Relation rel;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Datum repl_val[Natts_pg_authid];
char repl_null[Natts_pg_authid];
char repl_repl[Natts_pg_authid];
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
int i;
Merge the last few variable.c configuration variables into the generic GUC support. It's now possible to set datestyle, timezone, and client_encoding from postgresql.conf and per-database or per-user settings. Also, implement rollback of SET commands that occur in a transaction that later fails. Create a SET LOCAL var = value syntax that sets the variable only for the duration of the current transaction. All per previous discussions in pghackers.
2002-05-17 03:19:19 +02:00
valuestr = flatten_set_variable_args(stmt->variable, stmt->value);
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
/*
pgindent run.
2002-09-04 22:31:48 +02:00
* RowExclusiveLock is sufficient, because we don't need to update the
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* flat auth file.
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
rel = heap_open(AuthIdRelationId, RowExclusiveLock);
oldtuple = SearchSysCache(AUTHNAME,
PointerGetDatum(stmt->role),
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
0, 0, 0);
if (!HeapTupleIsValid(oldtuple))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" does not exist", stmt->role)));
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
if (!(superuser() ||
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
(HeapTupleGetOid(oldtuple) == GetUserId())))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("permission denied")));
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
for (i = 0; i < Natts_pg_authid; i++)
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
repl_repl[i] = ' ';
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_repl[Anum_pg_authid_rolconfig - 1] = 'r';
pgindent run.
2002-09-04 22:31:48 +02:00
if (strcmp(stmt->variable, "all") == 0 && valuestr == NULL)
Repair core dump when trying to delete an entry from an already-NULL datconfig or useconfig field. Per report from Dustin Sallings.
2002-12-02 06:20:47 +01:00
{
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
/* RESET ALL */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_null[Anum_pg_authid_rolconfig - 1] = 'n';
Repair core dump when trying to delete an entry from an already-NULL datconfig or useconfig field. Per report from Dustin Sallings.
2002-12-02 06:20:47 +01:00
}
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
else
{
pgindent run.
2002-09-04 22:31:48 +02:00
Datum datum;
bool isnull;
ArrayType *array;
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_null[Anum_pg_authid_rolconfig - 1] = ' ';
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
datum = SysCacheGetAttr(AUTHNAME, oldtuple,
Anum_pg_authid_rolconfig, &isnull);
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
More janitorial work: remove the explicit casting of NULL literals to a pointer type when it is not necessary to do so. For future reference, casting NULL to a pointer type is only necessary when (a) invoking a function AND either (b) the function has no prototype OR (c) the function is a varargs function.
2004-01-07 19:56:30 +01:00
array = isnull ? NULL : DatumGetArrayTypeP(datum);
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
if (valuestr)
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
array = GUCArrayAdd(array, stmt->variable, valuestr);
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
else
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
array = GUCArrayDelete(array, stmt->variable);
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
Repair core dump when trying to delete an entry from an already-NULL datconfig or useconfig field. Per report from Dustin Sallings.
2002-12-02 06:20:47 +01:00
if (array)
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_val[Anum_pg_authid_rolconfig - 1] = PointerGetDatum(array);
Repair core dump when trying to delete an entry from an already-NULL datconfig or useconfig field. Per report from Dustin Sallings.
2002-12-02 06:20:47 +01:00
else
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_null[Anum_pg_authid_rolconfig - 1] = 'n';
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
}
Change heap_modifytuple() to require a TupleDesc rather than a Relation. Patch from Alvaro Herrera, minor editorializing by Neil Conway.
2005-01-28 00:24:11 +01:00
newtuple = heap_modifytuple(oldtuple, RelationGetDescr(rel), repl_val, repl_null, repl_repl);
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
simple_heap_update(rel, &oldtuple->t_self, newtuple);
Restructure system-catalog index updating logic. Instead of having hardwired lists of index names for each catalog, use the relcache's mechanism for caching lists of OIDs of indexes of any table. This reduces the common case of updating system catalog indexes to a single line, makes it much easier to add a new system index (in fact, you can now do so on-the-fly if you want to), and as a nice side benefit improves performance a little. Per recent pghackers discussion.
2002-08-05 05:29:17 +02:00
CatalogUpdateIndexes(rel, newtuple);
User and database-specific session defaults for run-time configuration variables. New commands ALTER DATABASE ... SET and ALTER USER ... SET.
2002-03-01 23:45:19 +01:00
ReleaseSysCache(oldtuple);
heap_close(rel, RowExclusiveLock);
}
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* DROP ROLE
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
*/
void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
DropRole(DropRoleStmt *stmt)
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Relation pg_authid_rel, pg_auth_members_rel;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
ListCell *item;
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
if (!superuser())
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("must be superuser to drop roles")));
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Scan the pg_authid relation to find the Oid of the role to be
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
* deleted. Note we secure exclusive lock, because we need to protect
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* our update of the flat auth file.
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
pg_authid_rel = heap_open(AuthIdRelationId, ExclusiveLock);
pg_auth_members_rel = heap_open(AuthMemRelationId, ExclusiveLock);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
foreach(item, stmt->roles)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
const char *role = strVal(lfirst(item));
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
HeapTuple tuple,
tmp_tuple;
Relation pg_rel;
TupleDesc pg_dsc;
ScanKeyData scankey;
HeapScanDesc scan;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
CatCList *auth_mem_list;
Oid roleid;
int i;
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
tuple = SearchSysCache(AUTHNAME,
PointerGetDatum(role),
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
0, 0, 0);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
if (!HeapTupleIsValid(tuple))
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" does not exist", role)));
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
roleid = HeapTupleGetOid(tuple);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (roleid == GetUserId())
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_OBJECT_IN_USE),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("current role cannot be dropped")));
if (roleid == GetSessionUserId())
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_OBJECT_IN_USE),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("session role cannot be dropped")));
Make the world somewhat safe for (not from) DELETE FROM pg_shadow; Assign the fixed user id 1 to the user created by initdb. A stand-alone backend will always set the user id to 1. (Consequently, the name of that user is no longer important.) In stand-alone mode, the user id 1 will have implicit superuser status, to allow repairs even if there are no users defined. Print a warning message when starting in stand-alone mode when no users are defined. Disallow dropping the current user and session user. Granting/revoking superuser status also grants/revokes usecatupd. (Previously, it would never grant it back. This could lead to "deadlocks".) CREATE USER and CREATE GROUP will start allocating user ids at 100 (unless explicitly specified), to prevent accidental creation of a superuser (plus some room for future extensions).
2001-09-08 17:24:00 +02:00
Remove dashes in comments that don't need them, rewrap with pgindent.
2001-03-22 07:16:21 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Check if role still owns a database. If so, error out.
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*
Remove dashes in comments that don't need them, rewrap with pgindent.
2001-03-22 07:16:21 +01:00
* (It used to be that this function would drop the database
* automatically. This is not only very dangerous for people that
* don't read the manual, it doesn't seem to be the behaviour one
* would expect either.) -- petere 2000/01/14)
*/
Completion of project to use fixed OIDs for all system catalogs and indexes. Replace all heap_openr and index_openr calls by heap_open and index_open. Remove runtime lookups of catalog OID numbers in various places. Remove relcache's support for looking up system catalogs by name. Bulky but mostly very boring patch ...
2005-04-14 22:03:27 +02:00
pg_rel = heap_open(DatabaseRelationId, AccessShareLock);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
pg_dsc = RelationGetDescr(pg_rel);
Cross-data-type comparisons are now indexable by btrees, pursuant to my pghackers proposal of 8-Nov. All the existing cross-type comparison operators (int2/int4/int8 and float4/float8) have appropriate support. The original proposal of storing the right-hand-side datatype as part of the primary key for pg_amop and pg_amproc got modified a bit in the event; it is easier to store zero as the 'default' case and only store a nonzero when the operator is actually cross-type. Along the way, remove the long-since-defunct bigbox_ops operator class.
2003-11-12 22:15:59 +01:00
ScanKeyInit(&scankey,
Anum_pg_database_datdba,
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
BTEqualStrategyNumber, F_OIDEQ,
roleid);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Restructure indexscan API (index_beginscan, index_getnext) per yesterday's proposal to pghackers. Also remove unnecessary parameters to heap_beginscan, heap_rescan. I modified pg_proc.h to reflect the new numbers of parameters for the AM interface routines, but did not force an initdb because nothing actually looks at those fields.
2002-05-21 01:51:44 +02:00
scan = heap_beginscan(pg_rel, SnapshotNow, 1, &scankey);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Restructure indexscan API (index_beginscan, index_getnext) per yesterday's proposal to pghackers. Also remove unnecessary parameters to heap_beginscan, heap_rescan. I modified pg_proc.h to reflect the new numbers of parameters for the AM interface routines, but did not force an initdb because nothing actually looks at those fields.
2002-05-21 01:51:44 +02:00
if ((tmp_tuple = heap_getnext(scan, ForwardScanDirection)) != NULL)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
char *dbname;
Clean up various to-do items associated with system indexes: pg_database now has unique indexes on oid and on datname. pg_shadow now has unique indexes on usename and on usesysid. pg_am now has unique index on oid. pg_opclass now has unique index on oid. pg_amproc now has unique index on amid+amopclaid+amprocnum. Remove pg_rewrite's unnecessary index on oid, delete unused RULEOID syscache. Remove index on pg_listener and associated syscache for performance reasons (caching rows that are certain to change before you need 'em again is rather pointless). Change pg_attrdef's nonunique index on adrelid into a unique index on adrelid+adnum. Fix various incorrect settings of pg_class.relisshared, make that the primary reference point for whether a relation is shared or not. IsSharedSystemRelationName() is now only consulted to initialize relisshared during initial creation of tables and indexes. In theory we might now support shared user relations, though it's not clear how one would get entries for them into pg_class &etc of multiple databases. Fix recently reported bug that pg_attribute rows created for an index all have the same OID. (Proof that non-unique OID doesn't matter unless it's actually used to do lookups ;-)) There's no need to treat pg_trigger, pg_attrdef, pg_relcheck as bootstrap relations. Convert them into plain system catalogs without hardwired entries in pg_class and friends. Unify global.bki and template1.bki into a single init script postgres.bki, since the alleged distinction between them was misleading and pointless. Not to mention that it didn't work for setting up indexes on shared system relations. Rationalize locking of pg_shadow, pg_group, pg_attrdef (no need to use AccessExclusiveLock where ExclusiveLock or even RowExclusiveLock will do). Also, hold locks until transaction commit where necessary.
2001-06-12 07:55:50 +02:00
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
dbname = NameStr(((Form_pg_database) GETSTRUCT(tmp_tuple))->datname);
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
ereport(ERROR,
(errcode(ERRCODE_OBJECT_IN_USE),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" cannot be dropped", role),
errdetail("The role owns database \"%s\".", dbname)));
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
heap_endscan(scan);
Clean up various to-do items associated with system indexes: pg_database now has unique indexes on oid and on datname. pg_shadow now has unique indexes on usename and on usesysid. pg_am now has unique index on oid. pg_opclass now has unique index on oid. pg_amproc now has unique index on amid+amopclaid+amprocnum. Remove pg_rewrite's unnecessary index on oid, delete unused RULEOID syscache. Remove index on pg_listener and associated syscache for performance reasons (caching rows that are certain to change before you need 'em again is rather pointless). Change pg_attrdef's nonunique index on adrelid into a unique index on adrelid+adnum. Fix various incorrect settings of pg_class.relisshared, make that the primary reference point for whether a relation is shared or not. IsSharedSystemRelationName() is now only consulted to initialize relisshared during initial creation of tables and indexes. In theory we might now support shared user relations, though it's not clear how one would get entries for them into pg_class &etc of multiple databases. Fix recently reported bug that pg_attribute rows created for an index all have the same OID. (Proof that non-unique OID doesn't matter unless it's actually used to do lookups ;-)) There's no need to treat pg_trigger, pg_attrdef, pg_relcheck as bootstrap relations. Convert them into plain system catalogs without hardwired entries in pg_class and friends. Unify global.bki and template1.bki into a single init script postgres.bki, since the alleged distinction between them was misleading and pointless. Not to mention that it didn't work for setting up indexes on shared system relations. Rationalize locking of pg_shadow, pg_group, pg_attrdef (no need to use AccessExclusiveLock where ExclusiveLock or even RowExclusiveLock will do). Also, hold locks until transaction commit where necessary.
2001-06-12 07:55:50 +02:00
heap_close(pg_rel, AccessShareLock);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
* Somehow we'd have to check for tables, views, etc. owned by the
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* role as well, but those could be spread out over all sorts of
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
* databases which we don't have access to (easily).
*/
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Remove the role from the pg_authid table
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
simple_heap_delete(pg_authid_rel, &tuple->t_self);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
ReleaseSysCache(tuple);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Remove role from roles
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* scan pg_auth_members and remove tuples which have
* roleid == member or roleid == role
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
auth_mem_list = SearchSysCacheList(AUTHMEMROLEMEM, 1,
ObjectIdGetDatum(roleid),
0, 0, 0);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
for (i = 0; i < auth_mem_list->n_members; i++)
{
HeapTuple authmemtup = &auth_mem_list->members[i]->tuple;
simple_heap_delete(pg_auth_members_rel, &authmemtup->t_self);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
ReleaseSysCacheList(auth_mem_list);
pgindent run. Make it all clean.
2001-03-22 05:01:46 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
auth_mem_list = SearchSysCacheList(AUTHMEMMEMROLE, 1,
ObjectIdGetDatum(roleid),
0, 0, 0);
for (i = 0; i < auth_mem_list->n_members; i++)
{
HeapTuple authmemtup = &auth_mem_list->members[i]->tuple;
simple_heap_delete(pg_auth_members_rel, &authmemtup->t_self);
}
ReleaseSysCacheList(auth_mem_list);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
/*
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
* Now we can clean up; but keep lock until commit (to avoid possible
* deadlock when commit code tries to acquire lock).
Mega-commit to make heap_open/heap_openr/heap_close take an additional argument specifying the kind of lock to acquire/release (or 'NoLock' to do no lock processing). Ensure that all relations are locked with some appropriate lock level before being examined --- this ensures that relevant shared-inval messages have been processed and should prevent problems caused by concurrent VACUUM. Fix several bugs having to do with mismatched increment/decrement of relation ref count and mismatched heap_open/close (which amounts to the same thing). A bogus ref count on a relation doesn't matter much *unless* a SI Inval message happens to arrive at the wrong time, which is probably why we got away with this sloppiness for so long. Repair missing grab of AccessExclusiveLock in DROP TABLE, ALTER/RENAME TABLE, etc, as noted by Hiroshi. Recommend 'make clean all' after pulling this update; I modified the Relation struct layout slightly. Will post further discussion to pghackers list shortly.
1999-09-18 21:08:25 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
heap_close(pg_auth_members_rel, NoLock);
heap_close(pg_authid_rel, NoLock);
pgindent run before 6.3 release, with Thomas' requested changes.
1998-02-26 05:46:47 +01:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Set flag to update flat auth file at commit.
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
auth_file_update_needed();
Missed a few files from Todd's patch...oops :)
1997-12-04 01:34:01 +01:00
}
Password fix. Now people have to do the REVOKE themselves.
1998-02-19 18:20:01 +01:00
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Rename role
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
*/
void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
RenameRole(const char *oldname, const char *newname)
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
{
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
HeapTuple oldtuple,
newtuple;
TupleDesc dsc;
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
Relation rel;
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
Datum datum;
bool isnull;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Datum repl_val[Natts_pg_authid];
char repl_null[Natts_pg_authid];
char repl_repl[Natts_pg_authid];
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
int i;
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Oid roleid;
Pgindent run for 8.0.
2004-08-29 07:07:03 +02:00
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
/* ExclusiveLock because we need to update the password file */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
rel = heap_open(AuthIdRelationId, ExclusiveLock);
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
dsc = RelationGetDescr(rel);
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
oldtuple = SearchSysCache(AUTHNAME,
Pgindent run for 8.0.
2004-08-29 07:07:03 +02:00
CStringGetDatum(oldname),
0, 0, 0);
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
if (!HeapTupleIsValid(oldtuple))
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
ereport(ERROR,
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
(errcode(ERRCODE_UNDEFINED_OBJECT),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" does not exist", oldname)));
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
/*
pgindent run.
2003-08-04 02:43:34 +02:00
* XXX Client applications probably store the session user somewhere,
* so renaming it could cause confusion. On the other hand, there may
* not be an actual problem besides a little confusion, so think about
* this and decide.
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
roleid = HeapTupleGetOid(oldtuple);
if (roleid == GetSessionUserId())
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
ereport(ERROR,
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("session role may not be renamed")));
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
/* make sure the new name doesn't exist */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (SearchSysCacheExists(AUTHNAME,
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
CStringGetDatum(newname),
0, 0, 0))
ereport(ERROR,
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
(errcode(ERRCODE_DUPLICATE_OBJECT),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("role \"%s\" already exists", newname)));
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
/* must be superuser */
if (!superuser())
ereport(ERROR,
First bits of work on error message editing.
2003-07-19 01:20:33 +02:00
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
errmsg("must be superuser to rename roles")));
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
for (i = 0; i < Natts_pg_authid; i++)
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
repl_repl[i] = ' ';
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_repl[Anum_pg_authid_rolname - 1] = 'r';
repl_val[Anum_pg_authid_rolname - 1] = DirectFunctionCall1(namein,
Pgindent run for 8.0.
2004-08-29 07:07:03 +02:00
CStringGetDatum(newname));
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_null[Anum_pg_authid_rolname - 1] = ' ';
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
datum = heap_getattr(oldtuple, Anum_pg_authid_rolpassword, dsc, &isnull);
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
if (!isnull && isMD5(DatumGetCString(DirectFunctionCall1(textout, datum))))
{
/* MD5 uses the username as salt, so just clear it on a rename */
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
repl_repl[Anum_pg_authid_rolpassword - 1] = 'r';
repl_null[Anum_pg_authid_rolpassword - 1] = 'n';
Pgindent run for 8.0.
2004-08-29 07:07:03 +02:00
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
ereport(NOTICE,
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
(errmsg("MD5 password cleared because of role rename")));
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
}
Pgindent run for 8.0.
2004-08-29 07:07:03 +02:00
Change heap_modifytuple() to require a TupleDesc rather than a Relation. Patch from Alvaro Herrera, minor editorializing by Neil Conway.
2005-01-28 00:24:11 +01:00
newtuple = heap_modifytuple(oldtuple, dsc, repl_val, repl_null, repl_repl);
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
simple_heap_update(rel, &oldtuple->t_self, newtuple);
Pgindent run for 8.0.
2004-08-29 07:07:03 +02:00
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
CatalogUpdateIndexes(rel, newtuple);
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
Erase MD5 user passwords when a user is renamed because the username is used as salt for the MD5 password.
2004-05-06 18:59:16 +02:00
ReleaseSysCache(oldtuple);
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
heap_close(rel, NoLock);
Fix up problems in write_auth_file and parsing of the auth file. In particular, make hba.c cope with zero-length tokens, which it never did properly before. Also, enforce rolcanlogin.
2005-06-29 00:16:45 +02:00
/*
* Set flag to update flat auth file at commit.
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
auth_file_update_needed();
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
}
Password fix. Now people have to do the REVOKE themselves.
1998-02-19 18:20:01 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* GrantRoleStmt
Password fix. Now people have to do the REVOKE themselves.
1998-02-19 18:20:01 +01:00
*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Grant/Revoke roles to/from roles
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
*/
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
GrantRole(GrantRoleStmt *stmt)
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Oid grantor;
List *grantee_ids;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
ListCell *item;
Remove grammar restrictions on order of optional clauses in CREATE GROUP. From Vince Vielhaber.
2001-07-12 20:03:00 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (stmt->grantor)
grantor = get_roleid_checked(stmt->grantor);
else
grantor = GetUserId();
Remove grammar restrictions on order of optional clauses in CREATE GROUP. From Vince Vielhaber.
2001-07-12 20:03:00 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
grantee_ids = roleNamesToIds(stmt->grantee_roles);
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Step through all of the granted roles and add/remove
* entries for the grantees, or, if admin_opt is set, then
* just add/remove the admin option.
*
* Note: Permissions checking is done by AddRoleMems/DelRoleMems
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
foreach(item, stmt->granted_roles)
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
char *rolename = strVal(lfirst(item));
Oid roleid = get_roleid_checked(rolename);
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (stmt->is_grant)
AddRoleMems(rolename, roleid,
stmt->grantee_roles, grantee_ids,
grantor, stmt->admin_opt);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
else
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
DelRoleMems(rolename, roleid,
stmt->grantee_roles, grantee_ids,
stmt->admin_opt);
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
}
Fix up problems in write_auth_file and parsing of the auth file. In particular, make hba.c cope with zero-length tokens, which it never did properly before. Also, enforce rolcanlogin.
2005-06-29 00:16:45 +02:00
/*
* Set flag to update flat auth file at commit.
*/
auth_file_update_needed();
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
}
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
/*
* roleNamesToIds
*
* Given a list of role names (as String nodes), generate a list of role OIDs
* in the same order.
*/
static List *
roleNamesToIds(List *memberNames)
{
List *result = NIL;
ListCell *l;
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
foreach(l, memberNames)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
char *rolename = strVal(lfirst(l));
Oid roleid = get_roleid_checked(rolename);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
result = lappend_oid(result, roleid);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
return result;
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
}
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* AddRoleMems -- Add given members to the specified role
*
* rolename: name of role to add to (used only for error messages)
* roleid: OID of role to add to
* memberNames: list of names of roles to add (used only for error messages)
* memberIds: OIDs of roles to add
* grantorId: who is granting the membership
* admin_opt: granting admin option?
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
static void
AddRoleMems(const char *rolename, Oid roleid,
List *memberNames, List *memberIds,
Oid grantorId, bool admin_opt)
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Relation pg_authmem_rel;
TupleDesc pg_authmem_dsc;
ListCell *nameitem;
ListCell *iditem;
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Assert(list_length(memberNames) == list_length(memberIds));
* User management commands no longer user pg_exec_query_dest -> more robust * Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14 23:11:38 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
/* Skip permission check if nothing to do */
if (!memberIds)
return;
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Check permissions: must be superuser or have admin option on the
* role to be changed.
*
* XXX: The admin option is not considered to be inherited through
* multiple roles, unlike normal 'is_member_of_role' privilege checks.
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (!superuser())
{
HeapTuple authmem_chk_tuple;
Form_pg_auth_members authmem_chk;
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (grantorId != GetUserId())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser to set grantor ID")));
authmem_chk_tuple = SearchSysCache(AUTHMEMROLEMEM,
ObjectIdGetDatum(roleid),
ObjectIdGetDatum(grantorId),
0, 0);
if (!HeapTupleIsValid(authmem_chk_tuple))
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser or have admin option on role \"%s\"",
rolename)));
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
authmem_chk = (Form_pg_auth_members) GETSTRUCT(authmem_chk_tuple);
if (!authmem_chk->admin_option)
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser or have admin option on role \"%s\"",
rolename)));
ReleaseSysCache(authmem_chk_tuple);
}
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Secure exclusive lock to protect our update of the flat auth file.
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
pg_authmem_rel = heap_open(AuthMemRelationId, ExclusiveLock);
pg_authmem_dsc = RelationGetDescr(pg_authmem_rel);
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
forboth(nameitem, memberNames, iditem, memberIds)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
const char *membername = strVal(lfirst(nameitem));
Oid memberid = lfirst_oid(iditem);
HeapTuple authmem_tuple;
HeapTuple tuple;
Datum new_record[Natts_pg_auth_members];
char new_record_nulls[Natts_pg_auth_members];
char new_record_repl[Natts_pg_auth_members];
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Check if entry for this role/member already exists;
* if so, give warning unless we are adding admin option.
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
authmem_tuple = SearchSysCache(AUTHMEMROLEMEM,
ObjectIdGetDatum(roleid),
ObjectIdGetDatum(memberid),
0, 0);
if (HeapTupleIsValid(authmem_tuple) && !admin_opt)
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
ereport(NOTICE,
(errmsg("role \"%s\" is already a member of role \"%s\"",
membername, rolename)));
ReleaseSysCache(authmem_tuple);
continue;
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
/* Build a tuple to insert or update */
MemSet(new_record, 0, sizeof(new_record));
MemSet(new_record_nulls, ' ', sizeof(new_record_nulls));
MemSet(new_record_repl, ' ', sizeof(new_record_repl));
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record[Anum_pg_auth_members_roleid - 1] = ObjectIdGetDatum(roleid);
new_record[Anum_pg_auth_members_member - 1] = ObjectIdGetDatum(memberid);
new_record[Anum_pg_auth_members_grantor - 1] = ObjectIdGetDatum(grantorId);
new_record[Anum_pg_auth_members_admin_option - 1] = BoolGetDatum(admin_opt);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (HeapTupleIsValid(authmem_tuple))
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
new_record_repl[Anum_pg_auth_members_grantor - 1] = 'r';
new_record_repl[Anum_pg_auth_members_admin_option - 1] = 'r';
tuple = heap_modifytuple(authmem_tuple, pg_authmem_dsc,
new_record,
new_record_nulls, new_record_repl);
simple_heap_update(pg_authmem_rel, &tuple->t_self, tuple);
CatalogUpdateIndexes(pg_authmem_rel, tuple);
ReleaseSysCache(authmem_tuple);
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
}
else
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
tuple = heap_formtuple(pg_authmem_dsc,
new_record, new_record_nulls);
simple_heap_insert(pg_authmem_rel, tuple);
CatalogUpdateIndexes(pg_authmem_rel, tuple);
}
}
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
Authentication improvements: A new pg_hba.conf column, USER Allow specifiction of lists of users separated by commas Allow group names specified by + Allow include files containing lists of users specified by @ Allow lists of databases, and database files Allow samegroup in database column to match group name matching dbname Removal of secondary password files Remove pg_passwd utility Lots of code cleanup in user.c and hba.c New data/global/pg_pwd format New data/global/pg_group file
2002-04-04 06:25:54 +02:00
/*
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
* Now we can clean up; but keep lock until commit (to avoid possible
* deadlock when commit code tries to acquire lock).
Authentication improvements: A new pg_hba.conf column, USER Allow specifiction of lists of users separated by commas Allow group names specified by + Allow include files containing lists of users specified by @ Allow lists of databases, and database files Allow samegroup in database column to match group name matching dbname Removal of secondary password files Remove pg_passwd utility Lots of code cleanup in user.c and hba.c New data/global/pg_pwd format New data/global/pg_group file
2002-04-04 06:25:54 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
heap_close(pg_authmem_rel, NoLock);
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
}
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* DelRoleMems -- Remove given members from the specified role
*
* rolename: name of role to del from (used only for error messages)
* roleid: OID of role to del from
* memberNames: list of names of roles to del (used only for error messages)
* memberIds: OIDs of roles to del
* admin_opt: remove admin option only?
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
*/
static void
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
DelRoleMems(const char *rolename, Oid roleid,
List *memberNames, List *memberIds,
bool admin_opt)
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Relation pg_authmem_rel;
TupleDesc pg_authmem_dsc;
ListCell *nameitem;
ListCell *iditem;
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
Assert(list_length(memberNames) == list_length(memberIds));
/* Skip permission check if nothing to do */
if (!memberIds)
return;
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Check permissions: must be superuser or have admin option on the
* role to be changed.
*
* XXX: The admin option is not considered to be inherited through
* multiple roles, unlike normal 'is_member_of_role' privilege checks.
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (!superuser())
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
{
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
HeapTuple authmem_chk_tuple;
Form_pg_auth_members authmem_chk;
authmem_chk_tuple = SearchSysCache(AUTHMEMROLEMEM,
ObjectIdGetDatum(roleid),
ObjectIdGetDatum(GetUserId()),
0, 0);
if (!HeapTupleIsValid(authmem_chk_tuple))
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser or have admin option on role \"%s\"",
rolename)));
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
authmem_chk = (Form_pg_auth_members) GETSTRUCT(authmem_chk_tuple);
if (!authmem_chk->admin_option)
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser or have admin option on role \"%s\"",
rolename)));
ReleaseSysCache(authmem_chk_tuple);
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
}
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
/*
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
* Secure exclusive lock to protect our update of the flat auth file.
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
pg_authmem_rel = heap_open(AuthMemRelationId, ExclusiveLock);
pg_authmem_dsc = RelationGetDescr(pg_authmem_rel);
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
forboth(nameitem, memberNames, iditem, memberIds)
{
const char *membername = strVal(lfirst(nameitem));
Oid memberid = lfirst_oid(iditem);
HeapTuple authmem_tuple;
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
/*
* Find entry for this role/member
*/
authmem_tuple = SearchSysCache(AUTHMEMROLEMEM,
ObjectIdGetDatum(roleid),
ObjectIdGetDatum(memberid),
0, 0);
if (!HeapTupleIsValid(authmem_tuple))
{
ereport(WARNING,
(errmsg("role \"%s\" is not a member of role \"%s\"",
membername, rolename)));
continue;
}
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
if (!admin_opt)
{
/* Remove the entry altogether */
simple_heap_delete(pg_authmem_rel, &authmem_tuple->t_self);
}
else
{
/* Just turn off the admin option */
HeapTuple tuple;
Datum new_record[Natts_pg_auth_members];
char new_record_nulls[Natts_pg_auth_members];
char new_record_repl[Natts_pg_auth_members];
/* Build a tuple to update with */
MemSet(new_record, 0, sizeof(new_record));
MemSet(new_record_nulls, ' ', sizeof(new_record_nulls));
MemSet(new_record_repl, ' ', sizeof(new_record_repl));
new_record[Anum_pg_auth_members_admin_option - 1] = BoolGetDatum(false);
new_record_repl[Anum_pg_auth_members_admin_option - 1] = 'r';
tuple = heap_modifytuple(authmem_tuple, pg_authmem_dsc,
new_record,
new_record_nulls, new_record_repl);
simple_heap_update(pg_authmem_rel, &tuple->t_self, tuple);
CatalogUpdateIndexes(pg_authmem_rel, tuple);
}
Here's the Create/Alter/Drop Group stuff that's been really overdue. I didn't have time for documentation yet, but I'll write some. There are still some things to work out what happens when you alter or drop users, but the group stuff in and by itself is done. -- Peter Eisentraut Sernanders väg 10:115
1999-12-16 18:24:19 +01:00
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
ReleaseSysCache(authmem_tuple);
}
Support toasting of shared system relations, and provide toast tables for pg_database, pg_shadow, pg_group, all of which now have potentially-long fields. Along the way, get rid of SharedSystemRelationNames list: shared rels are now identified in their include/pg_catalog/*.h files by a BKI_SHARED_RELATION macro, while indexes and toast rels inherit sharedness automatically from their parent table. Fix some bugs with failure to detoast pg_group.grolist during ALTER GROUP.
2002-04-27 23:24:34 +02:00
Make CREATE/ALTER/DROP USER/GROUP transaction-safe, or at least pretty nearly so, by postponing write of flat password file until transaction commit.
2002-10-21 21:46:45 +02:00
/*
* Now we can clean up; but keep lock until commit (to avoid possible
* deadlock when commit code tries to acquire lock).
*/
Replace pg_shadow and pg_group by new role-capable catalogs pg_authid and pg_auth_members. There are still many loose ends to finish in this patch (no documentation, no regression tests, no pg_dump support for instance). But I'm going to commit it now anyway so that Alvaro can make some progress on shared dependencies. The catalog changes should be pretty much done.
2005-06-28 07:09:14 +02:00
heap_close(pg_authmem_rel, NoLock);
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
}