Bump org.postgresql:postgresql from 42.5.1 to 42.7.11 - #3
Open
dependabot[bot] wants to merge 1 commit into
Open
Conversation
Copy link
Copy Markdown
Sorry, something went wrong.
Bumps [org.postgresql:postgresql](https://github.com/pgjdbc/pgjdbc) from 42.5.1 to 42.7.11. - [Release notes](https://github.com/pgjdbc/pgjdbc/releases) - [Changelog](https://github.com/pgjdbc/pgjdbc/blob/master/CHANGELOG.md) - [Commits](pgjdbc/pgjdbc@REL42.5.1...REL42.7.11) --- updated-dependencies: - dependency-name: org.postgresql:postgresql dependency-version: 42.7.11 dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com>
fm4v
added a commit
to fm4v/sqlancer
that referenced
this pull request
Jun 13, 2026
Extends the existing TextIndexLike + JsonSkipIndex coverage toward the full ClickHouse text-index surface. Validated on dev-vm head 26.6.1.734: a 1-hour full-fleet run (167k queries, all ~45 oracles) produced 0 false positives from any of these units. Soundness rules below were established empirically there. ClickHouse#3 Pure-optimization toggle arm: TextIndexLike gains a DIRECT_READ_OFF arm (query_plan_direct_read_from_text_index=0, query_plan_text_index_add_hint=0). A pure optimization toggle must never change results; UNKNOWN_SETTING is already tolerated so it degrades on older builds. ClickHouse#4 Merge + delete-masked-part topology: TextIndexLike optionally issues a synchronous lightweight DELETE (lightweight_deletes_sync=2) and/or OPTIMIZE TABLE ... FINAL after the inserts, recording topology in assertion messages. The Java ground truth now counts over the live (non-deleted) rows. Targets the #107309 class (index honoring a delete-masked part). ClickHouse#6 General-fleet predicate injection: generateTextSearchPredicate emits startsWith / endsWith / multiSearchAny over plain String columns from a fixed vocabulary, gated by --text-search-predicate-emission (default on). These three are the functions proven index==scan-equivalent across ALL tokenizers (incl. array). hasToken/hasAllTokens/hasAnyTokens are deliberately NOT emitted here: hasToken diverges index-vs-scan on the array tokenizer (whole-value token) and hasAllTokens/hasAnyTokens diverge on ngrams (the multi-token needle's space-spanning grams) -- by-design tokenizer-semantics differences, not bugs. Sound for NoREC because renderSkipIndex never carries a preprocessor. #7a renderSkipIndex breadth: full tokenizer matrix (splitByNonAlpha, ngrams(N), array, asciiCJK, splitByString, sparseGrams) and text() eligibility extended to FixedString / LowCardinality(String) / Nullable(String) / Array(String) (bare) and Map(String,*) via mapKeys/mapValues expression indexes. New ClickHouseErrors.getTextIndexErrors() tolerances wired into table creation. ClickHouse#5 TextIndexPreprocessor oracle (NEW): builds a private table with INDEX(s) preprocessor=lower(s), then asserts that a forced direct read (force_data_skipping_indices + direct_read=1, add_hint=0) over a mixed-case corpus equals a Java lower()-token-membership ground truth. (The documented INDEX(lower(s)) equivalent form cannot be force-engaged for hasToken(s,...) on 26.6.1.734 -- it raises INDEX_NOT_USED -- so the ground-truth comparison replaces a second-table comparison.) #7b TextIndexContainer oracle (NEW): Array(String) + array tokenizer with exact has/hasAny/hasAll List ground truth, and Map(String,String) key-vs-value isolation via mapContainsKey/mapContainsValue, across DEFAULT / ignore_data_skipping_indices / use_skip_indexes=0 arms. ALTER lifecycle: ClickHouseAlterGenerator gains ADD_INDEX / MATERIALIZE_INDEX / CLEAR_INDEX / DROP_INDEX kinds. New ClickHouseTextIndexLifecycle oracle asserts CREATE-with-index == (index-free + ALTER ADD INDEX + MATERIALIZE INDEX SETTINGS mutations_sync=2) == use_skip_indexes=0 scan over a predicate battery (LIKE + hasToken always; hasAllTokens/hasAnyTokens only on the splitByNonAlpha arm, where they are index==scan-equivalent), exercising the distinct MATERIALIZE-over-historical-parts code path. All three new oracles registered in ClickHouseOracleFactory and the run / validity oracle-name lists. All oracles use private per-iteration tables, fixed token vocabulary, and integer-key/count comparisons to stay clear of the repo's known false-positive families.
fm4v
added a commit
that referenced
this pull request
Jun 24, 2026
…ide-int types + EET roundtrip identities) Implements the plan's Phase-1 P0 fast wins from docs/plans/2026-06-13-001-feat-clickhouse-coverage-backlog-30-ideas-plan.md. New differential / ground-truth oracles (each default-on, gated by a --flag, self-contained, soundness-checklist-compliant): - #2 PrewhereEquivalence WHERE == PREWHERE == optimize_move_to_prewhere=0 (multiset) - #3 ReadInOrderToggle optimize_read_in_order/aggregation_in_order on==off - #4 CountOptimization trivial/implicit/use_projections on==off + countIf cross-check + GROUP-BY-key count arm (hardens #106573, #106125) - #5 LazyMaterializationToggle query_plan_optimize_lazy_materialization on==off (positional) - #12 ReplacingDedup ReplacingMergeTree(ver) FINAL == argMax(val,ver) GROUP BY key - sqlancer#19 QuantileConsistency quantileExact==medianExact, monotone-in-level, Low<=Exact<=High - sqlancer#20 UniqExactness uniqExact == count(DISTINCT) == length(groupUniqArray) - sqlancer#21 ArgExtremum argMax / arraySort(groupArray) / groupArraySorted vs Java truth - sqlancer#28 MaterializedColumn MATERIALIZED/ALIAS col == defining expr (single-snapshot) Generator/catalog: - sqlancer#16 emit signed Int128/Int256 + Decimal256 (Decimal precision 39..76); Interval column emission deliberately NOT added (ClickHouse forbids storing Interval in tables); ClickHouseTypeParser gains an Interval read-back branch for free. - sqlancer#23 EET roundtrip identities: unhex(hex), base64Decode(base64Encode), tryBase64Decode(base64Encode), toIPv4(IPv4NumToString(toUInt32)), toIPv6(IPv6NumToString) -- each type-gated; +7 unit-test cases. Wired into ClickHouseOracleFactory, ClickHouseOptions (--*-oracle flags) and .claude/run-sqlancer.sh ALL_ORACLES. mvn compile green; EETIdentitiesTest 20/20. Probe-on-head items (recorded for dev-vm validation): query_plan_optimize_lazy_materialization, read_in_order_use_buffering, optimize_trivial_count_query family, groupArraySorted/array text format, IPv6 IPv4-mapped roundtrip.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bumps org.postgresql:postgresql from 42.5.1 to 42.7.11.
Release notesSourced from org.postgresql:postgresql's releases.
... (truncated)
ChangelogSourced from org.postgresql:postgresql's changelog.
... (truncated)
CommitsDependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.
Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
You can disable automated security fix PRs for this repo from the Security Alerts page.