summaryrefslogtreecommitdiff
path: root/src/backend
Commit message (Collapse)AuthorAgeFilesLines
* Fix tuple chain moving bug found by "Hiroshi Inoue" <Inoue@tpf.co.jp>.Vadim B. Mikheev1999-05-231-17/+105
|
* Update commentary in sample GEQO config file.Tom Lane1999-05-221-9/+17
|
* Reduce default GEQO 'effort' setting to MEDIUM always.Tom Lane1999-05-221-5/+2
| | | | | This agrees with the documentation and seems like a more useful default anyhow ...
* Modify aset.c logic so that blocks requested from malloc getTom Lane1999-05-221-31/+114
| | | | | bigger the more space is used in an allocset. This reduces the malloc overhead very substantially on queries that need lots of memory.
* Improve error message from failed LOAD command (includeTom Lane1999-05-222-18/+12
| | | | kernel's error description when file is not accessible).
* Allow GEQO effort to be specified numerically, as well asTom Lane1999-05-221-1/+4
| | | | | symbolic LOW/MEDIUM/HIGH values --- needed for experiments with other effort levels ...
* Modify backend switch parsing to prevent 'insecure' switchesTom Lane1999-05-225-160/+200
| | | | | from being accepted when they are passed from client connection request. Get rid of a couple that no longer do anything (like -P).
* Fix for select 1;select 2 without trailing semi.Bruce Momjian1999-05-221-14/+9
|
* Fix for DEFAULT ''.Bruce Momjian1999-05-227-43/+37
|
* Make postgres prompt backend>, and remove PARSEDEBUG.Bruce Momjian1999-05-223-117/+7
|
* Disable fix. Didn't work.Bruce Momjian1999-05-211-2/+4
|
* Fix typo and attempt default fix.Bruce Momjian1999-05-212-9/+10
|
* Treat {} as special regex too.Bruce Momjian1999-05-211-5/+7
|
* add retest, a regex testing programTatsuo Ishii1999-05-212-2/+51
|
* Fix problem with | in ~ comparison using index.Bruce Momjian1999-05-211-35/+52
|
* Report strerror() rather than errno in low-level backend libpqTom Lane1999-05-211-9/+12
| | | | failure messages.
* Generate distinct error messages for trigger function not foundTom Lane1999-05-201-3/+5
| | | | and trigger function found but returns wrong type.
* Fixed shift/reduce conflictJan Wieck1999-05-201-21/+7
| | | | | | | | SelectStmt and CursorStmt tried to parse FOR UPDATE ... / FOR READ ONLY. Cursor now checks that it is read only by looking at forUpdate of Query. SelectStmt handles FOR READ ONLY too. Jan
* Remove 4096 string limited key on block sizeBruce Momjian1999-05-192-17/+12
|
* Upgrade to PyGreSQL (2.4)Bruce Momjian1999-05-194-14/+22
|
* Add Aggref and ArrayRef to the set of node types that transformExprTom Lane1999-05-181-7/+13
| | | | | | will pass through rather than spitting up. This is necessary to handle cases where coerce_type causes a subexpression to be retransformed, as in SELECT count(*) + 1.0 FROM table
* Now that hashjoin is reliable for large joins (knock on wood),Tom Lane1999-05-181-47/+30
| | | | | | remove optimizer's arbitrary limit on how large a join it will use hashing for. (The limit was too large to prevent the problems we'd been seeing, anyway...)
* Remove no-longer-used fields in Hash and HashJoin nodes.Tom Lane1999-05-184-53/+7
|
* Rewrite hash join to use simple linked lists instead of aTom Lane1999-05-182-782/+409
| | | | | | | fixed-size hashtable. This should prevent 'hashtable out of memory' errors, unless you really do run out of memory. Note: target size for hashtable is now taken from -S postmaster switch, not -B, since it is local memory in the backend rather than shared memory.
* All works on linux now by my tests and regression(with patch below).Bruce Momjian1999-05-171-2/+13
| | | | | ALTER TABLE RENAME with extents. Ole Gjerde
* Skip junk nodes when comparing UNION target list lengths.Bruce Momjian1999-05-172-7/+25
|
* Change resjunk to a boolean.Bruce Momjian1999-05-1713-50/+51
|
* This is actually more of a fundamental problem with mdtruncate. ItBruce Momjian1999-05-171-8/+19
| | | | | | | | | | | | looks like someone just didn't add support for multiple segments for truncation. The following patch seems to do the right thing, for me at least. It passed my tests, my data looks right(no data that shouldn't be in there) and regression is ok. Ole Gjerde
* Prior patch added 2 more characters to string allocatedTom Lane1999-05-171-2/+2
| | | | for SERIAL column's constraint, but forgot to increase space palloc'd...
* SELECT * error message fix.Bruce Momjian1999-05-171-4/+3
|
* Apply freebsd specific patches dealign with ELF system from FreeBSD'sMarc G. Fournier1999-05-171-0/+2
| | | | ports collection ...
* Move IN to proper place.Bruce Momjian1999-05-171-4/+4
|
* CleanupBruce Momjian1999-05-171-2/+2
|
* Fix typo in change.Bruce Momjian1999-05-171-2/+2
|
* Tighten coding in new_join_pathkey, which seems to be a hotspotTom Lane1999-05-171-96/+53
| | | | for GEQO ...
* Change GEQO optimizer to release memory after each geneTom Lane1999-05-173-34/+74
| | | | | is evaluated. This bounds memory usage to something reasonable even when many tables are being joined.
* Require IN in LOCK syntax.Bruce Momjian1999-05-171-2/+2
|
* Change md* call to smgr*.Bruce Momjian1999-05-171-2/+2
|
* Minor code cleanup in optimizer.Tom Lane1999-05-162-21/+13
|
* Fix some typos in geqo optimizer --- it now generatesTom Lane1999-05-161-37/+14
| | | | | reasonable plans again. Still eats memory like there's no tomorrow, however :-(.
* I made it so it rolled over files at 1MB. My table ended up with 120Bruce Momjian1999-05-151-2/+2
| | | | | | | | | | | | | | | | | | | segments, and my indexes had 3(Yes, it DOES work!). DROP TABLE removed ALL segments from the table, but only the main index segment. So it looks like removing the table itself is using mdunlink in md.c, while removing indexes uses FileNameUnlink() which only unlinks 1 file. As far as I can tell, calling FileNameUnlink() and mdunlink() is basically the same, except mdunlink() deletes any extra segments. I've done some testing and it seems to work. It also passes regression tests(except float8, geometry and rules, but that's normal). If this patch is right, this fixes all known multi-segment problems on Linux. Ole Gjerde
* Hi, Bruce!Bruce Momjian1999-05-132-61/+185
| | | | | | | These are my last changes to lmgr fixing deadlock handling. Please apply them to cvs... Vadim
* Add double quotes around the sequence name generated to support theThomas G. Lockhart1999-05-131-45/+18
| | | | | | SERIAL data type DEFAULT clause. This fixes a problem finding the sequence name when mixed case table names are involved.
* Surround a variable declaration with ENABLE_OUTER_JOINS to suppressThomas G. Lockhart1999-05-131-1/+3
| | | | compiler warnings about an unused variable.
* set client_encoding to <nothing> crashes backend.Tatsuo Ishii1999-05-132-4/+13
|
* Rip out QueryTreeList structure, root and branch. QuerytreeTom Lane1999-05-1311-264/+175
| | | | | | | | | | lists are now plain old garden-variety Lists, allocated with palloc, rather than specialized expansible-array data allocated with malloc. This substantially simplifies their handling and eliminates several sources of memory leakage. Several basic types of erroneous queries (syntax error, attempt to insert a duplicate key into a unique index) now demonstrably leak zero bytes per query.
* Release allocated memory during AtAbort_Memory.Tom Lane1999-05-131-6/+35
|
* Fixed small bug in ruleutils and added output of pg_views andJan Wieck1999-05-121-2/+2
| | | | | | pg_rules to rules regression test. Jan
* Fixed wrong hasAggs when aggregate columns of view aren'tJan Wieck1999-05-122-4/+146
| | | | | | | | selected. Disabled ability of defining DISTINCT or ORDER BY on views. Jan
* Replaced targetlist entry in GroupClause by reference numberJan Wieck1999-05-1215-188/+223
| | | | | | | in Resdom and GroupClause so changing of resno's doesn't confuse the grouping any more. Jan
View Lorry Controller Status