Attribute index changes to their table in the dolt_diff summary#1656
Merged
Conversation
Index changes carry no named catalog entry, so the diff summary was blind to them: an index-only commit produced no row at all where Dolt reports the table with schema_change=1, for CREATE and DROP alike, in committed and working diffs, full scans and table_name-filtered scans. The summary now compares each catalog's index schema rows (name and sql) per table and flags the parent when they differ; the unnamed-entry branch in the full scan is also guarded to the master entry so index entries no longer take the dolt_schemas path against garbage roots. The diff oracle gains nine schema-object scenarios verified against real Dolt: index create/drop/mixed-with-data commits, working-set index changes, sibling-noise isolation, filtered scans, view-only and trigger-only commits (locking the already-correct dolt_schemas rows), and no-row-diff assertions for index-only commits. Statements whose syntax diverges between engines run through a new dual-setup helper. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Version-Control Performance CeilingsRuns: median of 3 executions per benchmark, excluding fixture setup. The
|
Sysbench-Style Benchmark: Doltlite vs SQLiteIn-MemoryReads
Writes
File-BackedReads
Writes
File-Backed (autocommit)Each statement runs as its own transaction — exposes per-commit ReadsReads have no commit cost; these are the same SQL files as the
Writes
100000 rows, median of 5 invocations per test; autocommit writes use 9, workload-only timing via host monotonic clock when available. Performance Ceiling Check (2.5x individual, 2x average; autocommit writes: 6x / 5x)All tests within ceilings. |
Sysbench-Style Benchmark (BLOB PK): Doltlite vs SQLiteCompanion to the classic Sysbench-Style Benchmark. Every workload here In-MemoryReads
Writes
File-BackedReads
Writes
File-Backed (autocommit)Each statement runs as its own transaction — exposes per-commit ReadsReads have no commit cost; these are the same SQL files as the
Writes
100000 rows, median of 5 invocations per test; autocommit writes use 9, workload-only timing via host monotonic clock when available. Performance Ceiling Check (2.5x individual, 2x average; autocommit writes: 6x / 5x)All tests within ceilings. |
Sysbench-Style Benchmark (TEXT PK): Doltlite vs SQLiteCompanion to the classic Sysbench-Style Benchmark. Every workload here In-MemoryReads
Writes
File-BackedReads
Writes
File-Backed (autocommit)Each statement runs as its own transaction — exposes per-commit ReadsReads have no commit cost; these are the same SQL files as the
Writes
100000 rows, median of 5 invocations per test; autocommit writes use 9, workload-only timing via host monotonic clock when available. Performance Ceiling Check (2.5x individual, 2x average; autocommit writes: 6x / 5x)All tests within ceilings. |
Sysbench-Style Benchmark (composite PK): Doltlite vs SQLiteCompanion to the classic Sysbench-Style Benchmark. Every workload here In-MemoryReads
Writes
File-BackedReads
Writes
File-Backed (autocommit)Each statement runs as its own transaction — exposes per-commit ReadsReads have no commit cost; these are the same SQL files as the
Writes
100000 rows, median of 5 invocations per test; autocommit writes use 9, workload-only timing via host monotonic clock when available. Performance Ceiling Check (2.5x individual, 2x average; autocommit writes: 6x / 5x)All tests within ceilings. |
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
1 participant
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.
From the schema-objects-through-add/diff oracle gap survey: the add oracle (516 lines) had zero schema-object coverage and the diff oracle (864 lines) nearly none. Probing the gaps immediately found that the
dolt_diffsummary is blind to index changes — an index-only commit produces no row at all, where Dolt reportst | 0 | 1:table_name-filtered scans — all silentdolt_schemas | 1 | 1) — now locked by oracle scenariosFix: the summary compares each catalog's index schema rows (name + sql) per table and sets
schema_change=1on the parent when they differ — index changes have no named entry, so attribution must go through schema rows, the same lesson as the staging fixes. Also guards the full scan's unnamed-entry branch to the master entry (index entries were previously taking the dolt_schemas comparison path against non-master roots).Coverage: nine new scenarios in
vc_oracle_diff_test.shverified against real Dolt, including a dual-setup helper for statements whose syntax diverges (DROP INDEX … ON t, MySQL trigger bodies). Fail-before: master fails exactly the six index scenarios (57/63); fixed: 63/63.Filed alongside: #1655 —
dolt_statushas the same index blindness (sibling of #1651); its fix targetsdoltlite_status.cand is sequenced after #1653 to avoid conflicts, and the dolt_add oracle scenarios for schema objects (which project dolt_status) queue behind both.Validation: diff oracle 63/63, diff-stat 49/49, diff-tvf 8/8, status 27/27 vs real Dolt; battery 88/88; matrix 38/38; C regression 309,830/309,830.
🤖 Generated with Claude Code