Releases: paradedb/paradedb
Releases · paradedb/paradedb
v0.18.1
What's Changed
- chore: cherry pick #2975 into
0.18.x
by @rebasedming in #2991 - fix: Install
cargo machete
as locked. (cherry-pick of #2993) by @stuhood in #2995 - perf: improve FSM extend and pop/pop_many performance by @eeeebbbbrrrr in #2989
- fix: parallel vacuum process now runs correctly (#2987) by @eeeebbbbrrrr in #3001
- feat: support aggregation over JSON data in aggregate custom scan. (#2970) by @mdashti in #3004
- fix: fixed empty table aggregate errors in aggregate custom scan by @github-actions[bot] in #3012
- fix: restore the garbage list by @github-actions[bot] in #3022
- perf: optimize merging heuristics to prefer background merging by @philippemnoel in #3033
- fix: Only mark Docker images as
latest
if they actually are the latest ones by @github-actions[bot] in #3026 - fix: make IndexLayerSizes calculations more accurate by @philippemnoel in #3039
- fix: Remove the string/numeric classification from fast fields explain by @philippemnoel in #3044
- ci: Fix out-of-order benchmark workflow steps and broken Slack alerts by @philippemnoel in #3048
- perf: reduce FSM locking overhead by @eeeebbbbrrrr in #3064
- chore: port fsm fixes from main by @eeeebbbbrrrr in #3067
- fix: conversion error from non-string JSON group-by columns by @philippemnoel in #3068
- feat: heap filter pushdown works with sub-queries by @paradedb-bot in #3070
- fix: don't include target_segment_count in background_layer_sizes by @paradedb-bot in #3075
- perf: update tantivy dependency by @paradedb-bot in #3086
- perf: FSM conditional locking & other minor drive-bys by @paradedb-bot in #3087
- fix: fix failing test by @paradedb-bot in #3093
- chore: clean up + test null subselects by @paradedb-bot in #3090
- chore: add missing upgrade scripts from 0.17.x by @eeeebbbbrrrr in #3097
- feat: ability to apply a slop value to "phrase queries" by @paradedb-bot in #3105
- chore: cherry-pick PR3013 to main by @paradedb-bot in #3104
- feat: EXPLAIN metrics for parallel workers by @paradedb-bot in #3113
- fix: Fix handling of nulls in TopN by @paradedb-bot in #3114
- fix: Do not background merge during vacuum if background merge is disabled by @paradedb-bot in #3116
- chore: Cherry pick paradedb-bot to 0.18.x by @philippemnoel in #3123
- feat: added support for ORDER BY aggregates in aggregate custom scan by @paradedb-bot in #3126
- chore: Upgrade to
0.18.1
by @rebasedming in #3127 - feat: add
limit_fetch_multiplier
index setting to reduce Top N retries by @paradedb-bot in #3129
Full Changelog: v0.18.0...v0.18.1
v0.17.12
What's Changed
- feat: EXPLAIN metrics for parallel workers by @paradedb-bot in #3112
- fix: Do not background merge during vacuum if background merge is disabled by @paradedb-bot in #3117
- chore: Upgrade to
0.17.12
by @rebasedming in #3120
Full Changelog: v0.17.10...v0.17.12
v0.17.10
What's Changed
- perf: reduce FSM locking overhead by @eeeebbbbrrrr in #3065
- chore: port fsm fixes from main by @eeeebbbbrrrr in #3066
- fix: don't include target_segment_count in background_layer_sizes by @paradedb-bot in #3076
- perf: update tantivy dependency by @eeeebbbbrrrr in #3082
- fix: Patch null subselects by @rebasedming in #3084
- fix: cherry pick excluded sub-queries from filter-pushdown (on non-indexed fields… by @rebasedming in #3089
- perf: FSM conditional locking & other minor drive-bys by @paradedb-bot in #3088
- fix: fix failing test by @paradedb-bot in #3094
- chore: Upgrade to
0.17.10
by @rebasedming in #3091
Full Changelog: v0.17.9...v0.17.10
v0.17.9
What's Changed
- chore: Upgrade to
0.17.9
by @rebasedming in #3041 - fix: make IndexLayerSizes calculations more accurate by @philippemnoel in #3040
Full Changelog: v0.17.8...v0.17.9
v0.17.8
What's Changed
- fix: Fix the permissions in the ParadeDB Docker image #2954 by @philippemnoel in #3027
- chore: Cherrypick #3023 into 0.17.x by @philippemnoel in #3029
- fix: add missing upgrade script from 0.16.4 to 0.16.5 by @eeeebbbbrrrr in #3010
- perf: optimize merging heuristics to prefer background merging by @philippemnoel in #3032
- chore: Upgrade to
0.17.18
by @rebasedming in #3034
Full Changelog: v0.17.7...v0.17.8
v0.17.7
What's Changed
- fix: restore the garbage list by @github-actions[bot] in #3021
- chore: Upgrade to
0.17.7
by @rebasedming in #3024
Full Changelog: v0.17.6...v0.17.7
v0.17.6
What's Changed
- fix: don't ERROR if
paradedb.terms_with_operator
is missing by @eeeebbbbrrrr in #3013 - fix: fsm: we have space if any slots are invalid, no need for all (#3015) by @eeeebbbbrrrr in #3016
- chore: Upgrade to
0.17.6
by @rebasedming in #3017
Full Changelog: v0.17.5...v0.17.6
v0.17.5
What's Changed
- chore: cherry pick #2975 into
0.17.x
by @rebasedming in #2990 - fix: Install
cargo machete
as locked. (cherry-pick of #2993) by @stuhood in #2994 - perf: improve FSM extend and pop/pop_many performance by @eeeebbbbrrrr in #2988
- fix: parallel vacuum process now runs correctly (#2987) by @eeeebbbbrrrr in #3000
- chore: cherry pick #2955 by @rebasedming in #3003
- chore: Upgrade to
0.17.5
by @rebasedming in #3005
Full Changelog: v0.17.4...v0.17.5
v0.18.0
What's Changed
- fix: Fix
rescan
for top-n by @stuhood in #2868 - docs: Explain we won't take docs PRs by @philippemnoel in #2873
- ci: Make benchmark Slack notifs be a link by @philippemnoel in #2874
- chore: Remove debug output in aggregate scan by @stuhood in #2876
- fix: issue #2844 - Use
paradedb.with_index()
when necessary by @eeeebbbbrrrr in #2879 - test: added more join regression tests + improved when to use Join quals by @mdashti in #2808
- chore: add docs + changelog about heap pushdown by @rebasedming in #2880
- chore: Make commit be hyperlink on its name for stressgres Slack notifs by @philippemnoel in #2881
- chore: move code around by @eeeebbbbrrrr in #2889
- chore: Bump upgrade job to a floor of
v0.15.19
by @stuhood in #2890 - perf: Skip computing scores if not requested for top-n field queries by @stuhood in #2892
- chore: add
background-merge.toml
to CI by @rebasedming in #2895 - perf: Avoid a re-query when a top-n query is exhausted by @stuhood in #2888
- feat: text search operators by @eeeebbbbrrrr in #2877
- chore: Update Issue template to specify how to get the ParadeDB version by @philippemnoel in #2898
- perf: Lazily load bitpacked columns by @stuhood in #2894
- fix: add #[pg_guard] to extern C-unwind functions by @eeeebbbbrrrr in #2899
- chore: upgrade to
0.17.1
by @rebasedming in #2902 - fix: Custom scan fails if partitioned table is empty by @rebasedming in #2905
- ci: Clarify release version format by @philippemnoel in #2906
- ci: test pg_search upgrades starting from v0.16.1 by @eeeebbbbrrrr in #2910
- feat: initial group-by aggregate implementation (for single group and COUNT(*) aggregation) by @mdashti in #2893
- chore: upgrade to
0.17.2
by @rebasedming in #2911 - perf: Reduce buffering during mixed fast field scans by @stuhood in #2887
- feat: new SQL builder functions by @eeeebbbbrrrr in #2897
- feat: improved
bool
type support by @mdashti in #2907 - chore: Enable running CI for drafts. by @stuhood in #2913
- feat: "proximity search" support by @eeeebbbbrrrr in #2685
- docs: Correct docs about filtering by @rebasedming in #2883
- perf: Remove StringFastField exec method by @stuhood in #2901
- feat: added support for ORDER BY in GROUP BY aggregate queries (pushed down to custom scan) by @mdashti in #2909
- chore: upgrade to pgrx v0.15.0 by @meotimdihia in #2917
- docs: New operators by @rebasedming in #2916
- feat: GROUP BY without an aggregation function by @mdashti in #2914
- feat: added support for multiple group-by columns (to be pushed down to custom scan). by @mdashti in #2915
- feat: ability to boost query predicate scores via
::boost(N)
by @eeeebbbbrrrr in #2918 - feat: JSON expressions can be used with text search operators and pushed down as filters by @rebasedming in #2919
- fix: Node relabeling no longer defeats new operators by @eeeebbbbrrrr in #2925
- docs: Fix broken Slack link by @philippemnoel in #2926
- fix: relation extension cache invalidation (#2927) by @eeeebbbbrrrr in #2934
- feat: ability to specify term "fuzzy"-ness via a cast to
::fuzzy(N)
by @eeeebbbbrrrr in #2921 - chore: add regression test for queries with score expressions by @rebasedming in #2936
- chore: add missing tests from enterprise by @eeeebbbbrrrr in #2937
- ci: Add missing code from SchemaBot Workflow by @philippemnoel in #2943
- ci: Fix benchmark-backfill-auto by @philippemnoel in #2942
- chore: fix release information by @eeeebbbbrrrr in #2949
- docs: fuzzy and boost by @rebasedming in #2938
- ci: Update the SSH deploy key to a unified one in organization-wide secrets by @philippemnoel in #2946
- fix: Fix the permissions in the ParadeDB Docker image by @philippemnoel in #2954
- perf: Use Arrow for fast field execution by @stuhood in #2924
- chore: Remove NumericFastField exec method by @stuhood in #2908
- fix: shouldn't panic if there was an interrupt by @rebasedming in #2955
- fix: Push down order-by for multiple sort columns to TopN by @stuhood in #2947
- fix: Properly set a tag during a manual docker publish by @stuhood in #2960
- feat: add support for additional aggregate functions (
SUM
,AVG
,MIN
,MAX
) to custom scan by @mdashti in #2923 - fix: zero worker threads by @eeeebbbbrrrr in #2959
- fix: the
LinkedListData::npages
field is dead code by @eeeebbbbrrrr in #2957 - fix: custom fsm "MAX BLOCKS PER PAGE" calculation by @eeeebbbbrrrr in #2958
- test: print reproduction script on prop-test failures by @mdashti in #2950
- ci: Don't tag Docker images as
latest
if there's a-
in the tag by @philippemnoel in #2969 - ci: Check for
-
in tag rather than-rc
in all our publish workflows by @philippemnoel in #2974 - chore: cherry pick upgrade to
0.17.4
(#2976) by @rebasedming in #2977 - chore: merge 0.17.x to main (fix bugs with 0.17.x exposed by the custom FSM (#2971)) by @eeeebbbbrrrr in #2979
- chore: upgrade to
0.18.0
by @rebasedming in #2980
New Contributors
- @meotimdihia made their first contribution in #2917
Full Changelog: v0.17.0...v0.18.0
v0.17.4
What's Changed
- fix: Properly set a tag during a manual docker publish (#2960) by @stuhood in #2961
- ci: Fix benchmark-backfill-auto (#2942) by @stuhood in #2962
- fix: bugs with 0.17.x exposed by the custom FSM by @eeeebbbbrrrr in #2971
- chore: Upgrade to
0.17.4
by @rebasedming in #2976
Full Changelog: v0.17.3...v0.17.4