Mercurial > hg > MonetDB

log

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
less more | rev 80892: (0) -30000 -10000 -3000 -1000 -300 -100 -60 +60 +100 +300 +1000 +3000 +10000 tip
age author description
2021-04-08 Pedro Ferreira Merged with default pushcands
2021-04-08 Pedro Ferreira More missing backups pushcands
2021-04-08 Pedro Ferreira Delay backend backup copy to where is needed pushcands
2021-04-08 Pedro Ferreira Code cleanup pushcands
2021-04-08 Niels Nes fix small issue with semijoins, which already pushed the candidates into the inner expression. pushcands
2021-04-08 Pedro Ferreira We can't push candidates through in values lists yet pushcands
2021-04-08 Pedro Ferreira More cleaning pushcands
2021-04-08 Pedro Ferreira Cleanup affected row count generation. If simply there are multiple insert/update/delete statements in the plan, update the accumulator pushcands
2021-04-08 Niels Nes include monetdbe_* manual_pages, please check the result
2021-04-08 Pedro Ferreira Cleanning up pushcands
2021-04-08 Niels Nes small perf test, ie use special loop for dense candidates in the string full scan with double elim (for now only for offset size == 1).
2021-04-08 Pedro Ferreira Reset backend before compiling a function pushcands
2021-04-08 svetlin preserve inline query comments
2021-04-07 Martin van Dinther Disabled two tests which produce non-stable output when summertime (or wintertime) changes timezone. Oct2020
2021-04-07 Pedro Ferreira I shouldn't have changed this pushcands
2021-04-07 Sjoerd Mullender Only print used algorithm in TRACE output.
2021-04-07 Panagiotis Koutsourakis Fix header construction string_imprints
2021-04-07 Panagiotis Koutsourakis Merge with default string_imprints
2021-04-07 Pedro Ferreira Merged with default pushcands
2021-04-07 Pedro Ferreira Added row counts on merge statement tests
2021-04-07 Pedro Ferreira Cleaning my old mess, generate the row count accumulator only when required and don't generate empty ddl exception statements pushcands
2021-04-07 Niels Nes small improvement, ie don't project already projected cols pushcands
2021-04-07 Niels Nes fallback on for aggregations, ie use current aggregated columns pushcands
2021-04-07 Pedro Ferreira More st_table fixes pushcands
2021-04-07 Pedro Ferreira Be more restrictive to stmt_affected_rows generation. More fixes for table statements pushcands
2021-04-07 Sjoerd Mullender Use different .tail files for different width string offsets.
2021-04-07 Pedro Ferreira More fixes pushcands
2021-04-07 Pedro Ferreira st_table fixes pushcands
2021-04-07 Niels Nes improved passing of the inner candidates pushcands
2021-04-07 Niels Nes also pass candidates from inner down into new rel_bin_stmt (in rel2bin_select) pushcands
2021-04-07 Niels Nes don't update inner sub. pushcands
2021-04-07 Niels Nes make sure we pass the selections during rel2bin_select, properly handling and lists. pushcands
2021-04-07 Niels Nes pass cols, candidates using a new rel_bin_stmt (ie solving problems with referenced relations) pushcands
2021-04-07 svetlin merge with default qcancel
2021-04-07 Niels Nes look at the right rel_bin_stmt for the group,ext and cnts pushcands
2021-04-07 Pedro Ferreira I need the number of affected rows on update statements to test the code generation changes on pushcands branch
2021-04-07 Pedro Ferreira More cleanup pushcands
2021-04-07 Pedro Ferreira Defensive lines pushcands
2021-04-07 Pedro Ferreira Added rel_bin_stmt in code generation (not finished yet). By grouping statments into relations will allow further optimizations at this phase such as pushing candidates. Some tests don't run yet. More cleanup needed pushcands
2021-04-07 Niels Nes optimizing the logger a bit.
2021-04-07 martin kersten Documentation of a failed experiment
2021-04-06 Niels Nes small fixes for sqlancer bufs
2021-04-06 Panagiotis Koutsourakis Construct strimp descriptor correctly string_imprints
2021-04-06 Panagiotis Koutsourakis Take into account negative numbers string_imprints
2021-04-06 Panagiotis Koutsourakis Fix bitmask construction string_imprints
2021-04-06 Panagiotis Koutsourakis Update comment string_imprints
2021-04-06 Sjoerd Mullender Approve.
2021-04-06 Pedro Ferreira Simplified version of buggy query. Right now I can't figure out a clean way to fix it
2021-04-04 martin kersten Avoid crash on wrong optimizer
2021-04-04 Pedro Ferreira There's no need to have two optimizers to push inner join expressions down, also attempt to push expressions on left and right joins on the possible sides. Removed unused rel_has_cmp_exp function
2021-04-04 martin kersten Remove history because nowadays we don't use cached queries anymore.
2021-04-03 Pedro Ferreira More wrong results
2021-04-03 Pedro Ferreira Also swap subquery to rhs on values lists
2021-04-03 Pedro Ferreira More wrong results
2021-04-03 Pedro Ferreira If the grouping relation has more than one reference, then the semijoin cannot be pushed down. This fixes sqlancer infinite optimization loop
2021-04-02 Pedro Ferreira Make sure exists rewriter returns a comparison for depth=0 expression on selections
2021-04-02 Sjoerd Mullender Powerpc approvals.
2021-04-02 Sjoerd Mullender 32 bit approvals.
2021-04-02 Sjoerd Mullender Approvals.
2021-04-02 Pedro Ferreira Merged with default properties
less more | rev 80892: (0) -30000 -10000 -3000 -1000 -300 -100 -60 +60 +100 +300 +1000 +3000 +10000 tip