database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-16 15:02:33 +03:00
Code Activity
5,987 Commits 38 Branches 669 Tags
70dd9e81af3d6f285d94098fc6dc249f2ad20e85
Commit Graph

373 Commits

Author SHA1 Message Date
Bruce Momjian
a9591ce66a Change #include's to use <> and "" as appropriate. 1999-07-15 23:04:24 +00:00
Bruce Momjian
2e6b1e63a3 Remove unused #includes in *.c files. 1999-07-15 22:40:16 +00:00
Bruce Momjian
4b2c2850bf Clean up #include in /include directory. Add scripts for checking includes. 1999-07-15 15:21:54 +00:00
Tom Lane
8aea617c03 Several routines failed to cope with CASE expressions, and 
indeed some of 'em were missing support for more node types than that...
 1999-07-15 01:52:09 +00:00
Bruce Momjian
0cf1b79528 Cleanup of /include #include's, for 6.6 only. 1999-07-14 01:20:30 +00:00
Bruce Momjian
db15dc05ad Fix for \do and ceil()/float. 1999-07-07 16:09:33 +00:00
Bruce Momjian
e9c977da7d Fix spelling of variable name. 1999-07-07 09:36:45 +00:00
Bruce Momjian
9f7ac20e57 Cleanup of min tuple size. 1999-07-07 09:27:28 +00:00
Bruce Momjian
1391098851 Fix misspelling. 1999-07-07 09:11:15 +00:00
Bruce Momjian
97dfff832c Fix to prevent too large tuple from being created. 1999-07-03 00:33:04 +00:00
Tom Lane
fd8e580bb7 Clean up problems with sublinks + grouping in planner. Not 
sure if they are all fixed, because rewriter is now the stumbling block,
but at least some cases work that did not work before.
 1999-06-21 01:20:57 +00:00
Tom Lane
974bdd94f9 On second thought, expression_tree_walker should handle bare 
SubLink nodes after all ...
 1999-06-21 01:18:02 +00:00
Bruce Momjian
326d8658ad Change form() to varargform() to prevent portability problems. 1999-06-19 04:54:23 +00:00
Tom Lane
86f36719db Create a generic expression-tree-walker subroutine, which 
will gradually replace all of the boilerplate tree-walk-recursion code that
currently exists in O(N) slightly different forms in N subroutines.
I've had it with adding missing cases to these subroutines...
 1999-06-19 03:41:45 +00:00
Tom Lane
aaf2442472 Remove query_planner's overhasty rejection of cases where 
tlist and qual are NULL.  It ought to handle these the same as the cases
where tlist contains only constant expressions, ie, be willing to generate
a Result-node plan.  This did not use to matter, but it does now because
union_planner will flatten the tlist when aggregates are present.  Thus,
'select count(1) from table' now causes query_planner to be given a null
tlist, and to duplicate 6.4's behavior we need it to give back a Result
plan rather than refusing the query.  6.4 was arguably doing the Wrong
Thing for this query, but I'm not going to open a semantics issue right
before 6.5 release ... can revisit that problem later.
 1999-06-12 19:38:30 +00:00
Tom Lane
acf242da97 Plug hole in dike: planner would coredump if query_planner 
returned NULL, which it will do in some cases where an elog(ERROR) would
probably be more appropriate.  For the moment, generate a not-very-
informative error message rather than proceeding to certain coredump.
Probably ought to think about making query_planner elog instead of
returning NULL, but this is at least a safe change for now.
 1999-06-12 19:27:41 +00:00
Tom Lane
c37ecaf8d5 Fix errors in SELECT ... GROUP BY ... UNION SELECT ... 
ye proverbial one-line patch (not counting five lines of comment so's
maybe it won't happen again)
 1999-06-10 06:55:40 +00:00
Tom Lane
b4210ae0f0 Fix problems with grouping/aggregation in queries that use 
inheritance ... basically it was completely busted :-(
 1999-06-06 17:38:11 +00:00
Tom Lane
b325dab67a new_relation_targetlist used to cause about 8 separate (and 
redundant) SearchSysCache searches per table column in an INSERT, which
accounted for a good percentage of the CPU time for INSERT ... VALUES().
Now it only does two searches in the typical case.
 1999-05-29 01:48:06 +00:00
Bruce Momjian
278bbf4572 Make functions static or NOT_USED as appropriate. 1999-05-26 12:57:23 +00:00
Bruce Momjian
fcff1cdf4e Another pgindent run. Sorry folks. 1999-05-25 22:43:53 +00:00
Bruce Momjian
4eadfe8754 Make 0x007f -> (unsigned)0x7f to make pgindent happy. 1999-05-25 22:04:56 +00:00
Bruce Momjian
07842084fe pgindent run over code. 1999-05-25 16:15:34 +00:00
Tom Lane
795f6ca66a Update commentary in sample GEQO config file. 1999-05-22 23:59:59 +00:00
Tom Lane
b2f14e11ec Reduce default GEQO 'effort' setting to MEDIUM always. 
This agrees with the documentation and seems like a more useful default
anyhow ...
 1999-05-22 23:27:19 +00:00
Tom Lane
b21005fa7c Allow GEQO effort to be specified numerically, as well as 
symbolic LOW/MEDIUM/HIGH values --- needed for experiments with other
effort levels ...
 1999-05-22 19:29:01 +00:00
Tom Lane
c2f0d565f3 Now that hashjoin is reliable for large joins (knock on wood), 
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...)
 1999-05-18 21:36:10 +00:00
Tom Lane
353d36f979 Remove no-longer-used fields in Hash and HashJoin nodes. 1999-05-18 21:34:29 +00:00
Bruce Momjian
585c967720 Change resjunk to a boolean. 1999-05-17 17:03:51 +00:00
Tom Lane
0b8b1fe3aa Tighten coding in new_join_pathkey, which seems to be a hotspot 
for GEQO ...
 1999-05-17 00:26:33 +00:00
Tom Lane
1332c1e144 Change GEQO optimizer to release memory after each gene 
is evaluated.  This bounds memory usage to something reasonable even
when many tables are being joined.
 1999-05-17 00:25:34 +00:00
Tom Lane
fecb2b0024 Minor code cleanup in optimizer. 1999-05-16 19:45:37 +00:00
Tom Lane
f2ed835baf Fix some typos in geqo optimizer --- it now generates 
reasonable plans again.  Still eats memory like there's no tomorrow,
however :-(.
 1999-05-16 19:45:00 +00:00
Tom Lane
507a0a2ab0 Rip out QueryTreeList structure, root and branch. Querytree 
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.
 1999-05-13 07:29:22 +00:00
Jan Wieck
79c2576f77 Replaced targetlist entry in GroupClause by reference number 
in Resdom and GroupClause so changing of resno's doesn't confuse
the grouping any more.

Jan
 1999-05-12 15:02:39 +00:00
Bruce Momjian
4853495e03 Change error messages to oids come out as %u and not %d. Change has no 
real affect now.
 1999-05-10 00:46:32 +00:00
Tom Lane
ec1f5f78b9 Fix oversights in flatten_tlistentry and replace_clause_joinvar_refs 
that led to CASE expressions not working very well in joined queries.
 1999-05-06 23:07:33 +00:00
Tom Lane
5729c3503d fix_indxqual_references didn't cope with ArrayRef nodes, 
meaning that this failed:
select proname,typname,prosrc from pg_proc,pg_type
where proname = 'float8' and pg_proc.proargtypes[0] = pg_type.oid;
 1999-05-06 01:30:58 +00:00
Tom Lane
5da466c597 Make sure targetlist generated for subplan does not share 
nodes with HAVING qualifier of upper plan.  Have not seen any failures,
just being a little bit paranoid...
 1999-05-04 00:00:20 +00:00
Tom Lane
da5f1dd722 Revise union_planner and associated routines to clean up breakage 
from EXCEPT/HAVING patch.  Cases involving nontrivial GROUP BY expressions
now work again.  Also, the code is at least somewhat better documented...
 1999-05-03 00:38:44 +00:00
Tom Lane
605d84941d Clean up cost_sort some more: most callers were double-counting 
the cost of reading the source data.
 1999-05-01 19:47:42 +00:00
Tom Lane
17dbeebb5f Fill in reasonable-looking cost estimates in inserted nodes. 
This makes no difference to the optimizer, which has already decided what
it's gonna do, but it makes the output of EXPLAIN much more plausible.
 1999-04-30 04:04:27 +00:00
Tom Lane
7a7ba33536 Clean up some bogosities in path cost estimation, like 
sometimes estimating an index scan of a table to be cheaper than a
sequential scan of the same tuples...
 1999-04-30 04:01:44 +00:00
Tom Lane
11a0027e28 Fix nasty little typo that prevented get_cheapest_path_for_joinkeys 
from ever returning a path.  This put a bit of a crimp in the system's
ability to generate intelligent merge-join plans...
 1999-04-30 03:59:06 +00:00
Tom Lane
970583ab4f Several routines in setrefs.c would crash on array refs 
due to lack of check for recursing into a null subexpression.
 1999-04-29 00:20:27 +00:00
Tom Lane
eb00bdf237 Insert a test for missing targetlist entry in replace_agg_clause. 1999-04-26 00:37:46 +00:00
Tom Lane
4438b70b94 Repair some problems in planner's handling of HAVING clauses. 
This fixes a few of the problems Hiroshi Inoue complained of, but
I have not touched the rewrite-related issues.
 1999-04-19 01:43:12 +00:00
Tom Lane
e91f43a122 Fix potential overflow problems when relation size exceeds 
2gig.  Fix failure to reliably put the smaller relation on the inside of
a hashjoin.
 1999-04-05 02:07:07 +00:00
Tom Lane
ff38837fe9 Fix nasty bug in optimization of multiway joins: optimizer 
would sometimes generate a plan that omitted a sort step before merge.
 1999-04-03 00:18:28 +00:00
Bruce Momjian
bd6f98af31 I suggest the following portability patch, which does not 
change functionality, but makes the code more ANSI C'ish.
My AIX xlc compiler barfs on all of these. Can someone please
review and apply to current.

 <<port.patch>>
Thanks
Andreas
 1999-03-19 18:56:43 +00:00