summaryrefslogtreecommitdiff
path: root/src/include/optimizer
AgeCommit message (Expand)Author
2011-09-03Rearrange planner to save the whole PlannerInfo (subroot) for a subquery.Tom Lane
2011-09-01Remove unnecessary #include references, per pgrminclude script.Bruce Momjian
2011-08-09Fix nested PlaceHolderVar expressions that appear only in targetlists.Tom Lane
2011-08-03Make pgbench use erand48() rather than random().Robert Haas
2011-07-12Avoid listing ungrouped Vars in the targetlist of Agg-underneath-Window.Tom Lane
2011-06-09Pgindent run before 9.1 beta2.Bruce Momjian
2011-04-24Improve cost estimation for aggregates and window functions.Tom Lane
2011-04-16Clean up collation processing in prepunion.c.Tom Lane
2011-04-10pgindent run before PG 9.1 beta 1.Bruce Momjian
2011-03-22Reimplement planner's handling of MIN/MAX aggregate optimization (again).Tom Lane
2011-03-20Revise collation derivation method and expression-tree representation.Tom Lane
2011-02-25Support data-modifying commands (INSERT/UPDATE/DELETE) in WITH.Tom Lane
2011-02-20Implement an API to let foreign-data wrappers actually be functional.Tom Lane
2011-02-08Per-column collation supportPeter Eisentraut
2011-01-01Stamp copyrights for year 2011.Bruce Momjian
2010-12-03Create core infrastructure for KNNGIST.Tom Lane
2010-11-19Improve relation width estimation for subqueries.Tom Lane
2010-11-08Use appendrel planning logic for top-level UNION ALL structures.Tom Lane
2010-11-04Reimplement planner's handling of MIN/MAX aggregate optimization.Tom Lane
2010-10-29Avoid creation of useless EquivalenceClasses during planning.Tom Lane
2010-10-14Support MergeAppend plans, to allow sorted output from append relations.Tom Lane
2010-10-11Improve the planner's simplification of NOT constructs.Tom Lane
2010-10-08Teach CLUSTER to use seqscan-and-sort when it's faster than indexscan.Tom Lane
2010-09-28Fix PlaceHolderVar mechanism's interaction with outer joins.Tom Lane
2010-09-20Remove cvs keywords from all files.Magnus Hagander
2010-09-14Fix join-removal logic for pseudoconstant and outerjoin-delayed quals.Tom Lane
2010-07-12Make NestLoop plan nodes pass outer-relation variables into their innerTom Lane
2010-04-19Add an 'enable_material' GUC.Robert Haas
2010-03-28Rework join-removal logic as per recent discussion. In particular thisTom Lane
2010-02-26pgindent run for 9.0Bruce Momjian
2010-02-12Extend the set of frame options supported for window functions.Tom Lane
2010-01-15Do parse analysis of an EXPLAIN's contained statement during the normalTom Lane
2010-01-02Update copyright for the year 2010.Bruce Momjian
2010-01-01Support "x IS NOT NULL" clauses as indexscan conditions. This turns outTom Lane
2009-12-15Support ORDER BY within aggregate function calls, at long last providing aTom Lane
2009-11-28Eliminate a lot of list-management overhead within join_search_one_levelTom Lane
2009-11-15Improve planning of Materialize nodes inserted atop the inner input of aTom Lane
2009-10-26Re-implement EvalPlanQual processing to improve its performance and eliminateTom Lane
2009-10-12Move the handling of SELECT FOR UPDATE locking and rechecking out ofTom Lane
2009-10-10Split the processing of INSERT/UPDATE/DELETE operations out of execMain.c.Tom Lane
2009-09-17Implement "join removal" for cases where the inner side of a left joinTom Lane
2009-09-12Rewrite the planner's handling of materialized plan types so that there isTom Lane
2009-07-16Make GEQO's planning deterministic by having it start from a predictableTom Lane
2009-07-16Make backend header files C++ safePeter Eisentraut
2009-07-06Fix set_append_rel_pathlist() to deal intelligently with cases whereTom Lane
2009-06-118.4 pgindent run, with new combined Linux/FreeBSD/MinGW typedef listBruce Momjian
2009-05-12Do some minor code refactoring in preparation for changing the APIs ofTom Lane
2009-05-09Fix cost_nestloop and cost_hashjoin to model the behavior of semi and antiTom Lane
2009-04-28Improve pull_up_subqueries logic so that it doesn't insert unnecessaryTom Lane
2009-04-19Fix estimate_num_groups() to not fail on PlaceHolderVars, per report fromTom Lane