perf(sql): improve performance of linear as-of join algo by bluestreak01 · Pull Request #6338 · questdb/questdb

bluestreak01 · 2025-10-31T01:30:08Z

optimisation for as-of joins on single symbol, such as:

SELECT /*+ ASOF_LINEAR_SEARCH(t p) */ avg(bid), max(bid) 
FROM (trades) t 
ASOF JOIN prices p on (t.symbol=p.sym);

It is activated with "ASOF_LINEAR_SEARCH" hint and is non-default, as it is a variant of linear as-of algo.

The performance improvement comes from re-indexing symbols and joining on "int" rather than symbol "text"

coderabbitai · 2025-10-31T01:30:34Z

Important

Review skipped

Auto reviews are disabled on this repository.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Walkthrough

Adds a single-symbol ASOF join cursor and integrates it into SQL code generation, refactors switch blocks to switch expressions, adds column-access helper overloads for RecordCursor, updates pattern matching and exception flow in AsyncFilteredRecordCursor, and adds tests for the single-symbol ASOF path.

Changes

Cohort / File(s)	Summary
ASOF single-symbol join + wiring `core/src/main/java/io/questdb/griffin/engine/join/AsOfJoinSingleSymbolRecordCursorFactory.java`, `core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java`	New specialized ASOF join factory for single-symbol joins; SqlCodeGenerator now selects a single-symbol fast path (AsOfJoinSingleSymbolRecordCursorFactory) when applicable and inlines previous createAsOfJoin logic into join-selection flow.
Column access helper overloads `core/src/main/java/io/questdb/griffin/engine/join/AsofJoinColumnAccessHelper.java`, `core/src/main/java/io/questdb/griffin/engine/join/SingleSymbolColumnAccessHelper.java`, `core/src/main/java/io/questdb/griffin/engine/join/NoopColumnAccessHelper.java`	Adds `of(RecordCursor)` overloads (and a default unsupported overload) so column-access helpers support non-TimeFrame slave cursors; minor formatting cleanup.
ASOF join cursor tests `core/src/test/java/io/questdb/test/griffin/AsOfJoinTest.java`	Adds tests covering single-symbol ASOF join behavior and dynamic-slave symbol error path.
Switch-expression refactor (date utilities) `core/src/main/java/io/questdb/griffin/engine/functions/date/GenerateSeriesTimestampStringRecordCursorFactory.java`	Replaced several traditional switch blocks with Java switch expressions across metadata, unit handling, random access support, and stride adjustment; no behavioral changes.
Exception handling pattern match `core/src/main/java/io/questdb/griffin/engine/table/AsyncFilteredRecordCursor.java`	Uses pattern matching for CairoException in catch block and rethrows bound CairoException for non-interruption cases instead of wrapping.
Minor formatting `core/src/main/java/io/questdb/std/Hash.java`	Added trailing newline.

Sequence Diagram(s)

sequenceDiagram
%%{init: {"themeVariables":{"sequenceActorBackground":"#f6f8fa","actorBorder":"#cbd5e1"}}}%%
actor Client
participant SqlCodeGenerator
participant MasterFactory
participant SlaveFactory
participant AsOfSingleFactory
participant AsOfLightFactory
participant AsOfCursor as "AsOfSingleSymbolJoinCursor"
Client->>SqlCodeGenerator: compile SELECT ... ASOF ... WITH HINT ASOF_LINEAR_SEARCH
SqlCodeGenerator->>MasterFactory: build master factory
SqlCodeGenerator->>SlaveFactory: analyze slave, detect single-symbol?
alt single-symbol and linear hint
    SqlCodeGenerator->>AsOfSingleFactory: instantiate AsOfJoinSingleSymbolRecordCursorFactory
    AsOfSingleFactory->>AsOfCursor: create inner cursor, build joinKeyMap from slave
    AsOfCursor-->>MasterFactory: drive master rows, probe joinKeyMap for symbol key
else fallback
    SqlCodeGenerator->>AsOfLightFactory: instantiate AsOfJoinLightRecordCursorFactory
    AsOfLightFactory->>SlaveFactory: build slave time frame structure
end
SqlCodeGenerator->>Client: return compiled plan using chosen factory

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes

Areas needing focused review:
- AsOfJoinSingleSymbolRecordCursorFactory.java: correctness of joinKeyMap usage, timestamp scaling and tolerance logic, cursor lifecycle and resource cleanup.
- SqlCodeGenerator.java: join-selection logic ensuring metadata and timestamp index propagation and correctness across ASOF/non-ASOF and CROSS join branches.
- AsyncFilteredRecordCursor.java: exception handling change (rethrowing CairoException) and its effects on callers and higher-level error handling.

Possibly related PRs

feat(sql): optimized ASOF JOIN on single symbol key where RHS symbol is low-frequency #6208 — modifies ASOF join code paths and single-symbol/memoized ASOF implementations; strong overlap with join-selection and factories.
feat(sql): speed up ASOF JOIN on indexed symbol column #6158 — touches ASOF single-symbol fast paths and SqlCodeGenerator join-selection; related implementation details.
feat(sql): join enhancements #6076 — adjusts SqlCodeGenerator join-construction paths and specific join factories (NestedLoop right/full); related control-flow changes.

Suggested labels

SQL, Performance

Suggested reviewers

mtopolnik
RaphDal

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 4.55% which is insufficient. The required threshold is 80.00%.	You can run `@coderabbitai generate docstrings` to improve docstring coverage.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title Check	✅ Passed	The pull request title "perf(sql): improve performance of linear as-of join algo" directly and accurately summarizes the main change in the changeset. The primary contribution is a new specialized ASOF join cursor factory (AsOfJoinSingleSymbolRecordCursorFactory) that optimizes single-symbol ASOF joins by re-indexing symbols and performing joins on integer IDs rather than text. The title is concise, uses domain-specific terminology ("sql"), indicates the nature of the change ("perf" for performance), and is clear enough for developers scanning PR history to understand the primary objective without requiring detailed inspection.
Description Check	✅ Passed	The pull request description is directly related to the changeset and provides meaningful context about the optimization. It explains that the change targets as-of joins on single symbols, provides a concrete example query, specifies the ASOF_LINEAR_SEARCH hint as the activation mechanism, clarifies that it's non-default, and articulates the performance benefit (re-indexing symbols and joining on integers rather than text). The description accurately reflects the new AsOfJoinSingleSymbolRecordCursorFactory and modifications to SqlCodeGenerator to implement this optimization.

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

bluestreak01 · 2025-10-31T01:34:47Z

@CodeRabbit review

coderabbitai · 2025-10-31T01:38:40Z

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

coderabbitai

Actionable comments posted: 2

🧹 Nitpick comments (5)

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java (2)
314-316: Remove redundant static import

import static io.questdb.griffin.model.QueryModel.*; already brings QUERY; the extra explicit static import is unnecessary.
- import static io.questdb.griffin.model.QueryModel.QUERY;
+ // QUERY already available via wildcard static import above
5298-5302: Pattern-matching instanceof introduces unused af

af isn’t referenced; either use the bound var or keep classic instanceof.
- if (!(f instanceof WindowFunction af)) {
+ if (!(f instanceof WindowFunction)) {
core/src/main/java/io/questdb/griffin/engine/join/AsOfJoinSingleSymbolRecordCursorFactory.java (3)
230-243: Consider adding circuit breaker check for long-running queries.

The while loop could iterate through many slave records without checking for query cancellation. For large slave cursors, this might delay cancellation.

Consider adding a circuit breaker check:
 final Record rec = slaveCursor.getRecord();
 while (slaveCursor.hasNext()) {
+    circuitBreaker.statefulThrowExceptionIfTripped();
     slaveTimestamp = scaleTimestamp(rec.getTimestamp(slaveTimestampIndex), slaveTimestampScale);
Note: You'll need to pass the circuit breaker to the hasNext method or store it as a field if this approach is adopted.

215-218: Consider extracting map update logic to reduce duplication.

The map update logic appears twice (lines 215-218 and 235-238) with identical structure.

Consider extracting to a helper method:
private void updateJoinKeyMap(Record rec, int symbolColumnIndex) {
    MapKey key = joinKeyMap.withKey();
    key.putInt(rec.getInt(symbolColumnIndex));
    MapValue value = key.createValue();
    value.putLong(0, rec.getRowId());
}
Then call it from both locations.

Also applies to: 235-238

249-258: Tolerance recheck is correct but subtle.

The tolerance check on line 255 is necessary because map entries may have been added during previous master record iterations with different tolerance windows. A brief comment would help future maintainers understand why this recheck is required.

Consider adding a clarifying comment:
 key = joinKeyMap.withKey();
 key.putInt(slaveSymbolKey);
 value = key.findValue();
 if (value != null) {
+    // Recheck tolerance: map entry may have been added for a previous master record
     slaveCursor.recordAt(slaveRecord, value.getLong(0));
     slaveTimestamp = scaleTimestamp(slaveRecord.getTimestamp(slaveTimestampIndex), slaveTimestampScale);
     record.hasSlave(toleranceInterval == Numbers.LONG_NULL || slaveTimestamp >= masterTimestamp - toleranceInterval);

📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between d979ed8 and 8f3e640.

📒 Files selected for processing (8)

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java (9 hunks)
core/src/main/java/io/questdb/griffin/engine/functions/date/GenerateSeriesTimestampStringRecordCursorFactory.java (3 hunks)
core/src/main/java/io/questdb/griffin/engine/join/AsOfJoinSingleSymbolRecordCursorFactory.java (1 hunks)
core/src/main/java/io/questdb/griffin/engine/join/AsofJoinColumnAccessHelper.java (2 hunks)
core/src/main/java/io/questdb/griffin/engine/join/SingleSymbolColumnAccessHelper.java (2 hunks)
core/src/main/java/io/questdb/griffin/engine/table/AsyncFilteredRecordCursor.java (1 hunks)
core/src/main/java/io/questdb/std/Hash.java (1 hunks)
core/src/test/java/io/questdb/test/griffin/AsOfJoinTest.java (1 hunks)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)

GitHub Check: New pull request (Check Changes Check changes)
GitHub Check: build
GitHub Check: build

🔇 Additional comments (13)

core/src/main/java/io/questdb/griffin/engine/functions/date/GenerateSeriesTimestampStringRecordCursorFactory.java (4)

83-87: LGTM! Clean switch expression refactor.

The switch expression correctly maps timestamp types to their corresponding metadata. The refactor improves readability while maintaining the same behavior.

205-209: LGTM! Switch expression correctly identifies fixed-length time units.

The logic appropriately returns false for variable-length units ('M', 'y') that don't support random access, and true for all fixed-length units. The refactor is clean and maintains existing semantics.

217-228: LGTM! Switch expression correctly maps time units to timestamp conversions.

The refactor cleanly converts each unit to its corresponding timestamp value. The method is only invoked when supportsRandomAccess() returns true (excluding 'M' and 'y'), and upstream validation at line 148-151 ensures only valid units reach this point.

235-249: LGTM! Switch expression provides clear validation for time unit characters.

The refactor correctly identifies all valid time units and returns the appropriate boolean. The inclusion of both 'h' and 'H' for hours, and 'u' and 'U' for micros provides good compatibility coverage.

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java (3)

196-196: Import for single‑symbol ASOF path

Importing AsOfJoinSingleSymbolRecordCursorFactory looks correct and scoped.

2672-2700: Switch expression simplification for CROSS OUTER joins LGTM

Cleaner and equivalent control flow; null‑row factories preserved per side.

4388-4402: Switch expression for select model LGTM

Branching is clearer; fall‑through to joins/no‑select is preserved.

core/src/main/java/io/questdb/griffin/engine/join/AsOfJoinSingleSymbolRecordCursorFactory.java (6)

55-89: LGTM! Excellent resource management.

The constructor properly handles resource cleanup in the exception path, ensuring the joinKeyMap is freed before calling close() and rethrowing.

96-109: LGTM! Proper cursor lifecycle management.

The getCursor method correctly handles cleanup of both master and slave cursors in the exception path.

129-135: LGTM! Complete resource cleanup.

All resources are properly freed.

152-173: LGTM! Efficient timestamp scaling logic.

The optimization to avoid timestamp scaling when types match is good for performance.

175-194: LGTM! Standard cursor lifecycle methods.

The close() method properly guards against double-close and resets state.

266-301: LGTM! Complete state management.

The toTop() method correctly resets all state, and the of() method properly handles map reopening for cursor reuse.

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java

core/src/test/java/io/questdb/test/griffin/AsOfJoinTest.java

bluestreak01 · 2025-10-31T01:58:36Z

@CodeRabbit review

coderabbitai · 2025-10-31T01:59:02Z

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java

bluestreak01 · 2025-11-01T04:30:27Z

@CodeRabbit review. Make sure NoOpAccessor cannot be passed to asof join implementations that expect a functioning accessor

coderabbitai · 2025-11-01T04:32:48Z

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Find NoOpAccessor definition and usage
rg -nP "class\s+NoOpAccessor" --type=java -A 10

# Find where NoOpAccessor is instantiated or used
rg -nP "NoOpAccessor" --type=java -C 3

Length of output: 0

🏁 Script executed:

#!/bin/bash
# Find AsofJoinColumnAccessHelper implementations and usages
rg -nP "(implements|extends).*AsofJoinColumnAccessHelper" --type=java -C 5

# Find all asof join factory implementations
rg -nP "class.*AsOf.*Join.*Factory" --type=java -C 3

Length of output: 17413

🏁 Script executed:

#!/bin/bash
# Look at the SqlCodeGenerator to see how it creates AsOfJoinSingleSymbolRecordCursorFactory
ast-grep --pattern 'new AsOfJoinSingleSymbolRecordCursorFactory($$$)'

# Also check AsOfJoinLightRecordCursorFactory creation
ast-grep --pattern 'new AsOfJoinLightRecordCursorFactory($$$)'

Length of output: 3577

🏁 Script executed:

#!/bin/bash
# Look at AsOfJoinSingleSymbolRecordCursorFactory constructor to see what parameters it takes
rg -nP "public\s+AsOfJoinSingleSymbolRecordCursorFactory" --type=java -A 20

Length of output: 3094

🏁 Script executed:

#!/bin/bash
# Find where NoopColumnAccessHelper.INSTANCE is used
rg -nP "NoopColumnAccessHelper\.INSTANCE" --type=java -C 5

Length of output: 6299

🏁 Script executed:

#!/bin/bash
# Look at the context around where columnAccessHelper is determined in SqlCodeGenerator before creating AsOfJoinSingleSymbolRecordCursorFactory
rg -nP "columnAccessHelper\s*=" --type=java -C 10 core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java

Length of output: 7285

🏁 Script executed:

#!/bin/bash
# Check the code path that determines when to use AsOfJoinSingleSymbolRecordCursorFactory vs AsOfJoinLightRecordCursorFactory
rg -nP "AsOfJoinSingleSymbolRecordCursorFactory|AsOfJoinLightRecordCursorFactory" --type=java -B 20 -A 5 core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java | head -100

Length of output: 5828

🏁 Script executed:

#!/bin/bash
# Look at NoopColumnAccessHelper implementation to see what methods throw
cat core/src/main/java/io/questdb/griffin/engine/join/NoopColumnAccessHelper.java

Length of output: 2119

🏁 Script executed:

#!/bin/bash
# Check which AsOf join factories take columnAccessHelper as a constructor parameter
rg -nP "public\s+\w+\s*\(" --type=java -A 15 core/src/main/java/io/questdb/griffin/engine/join/AsOfJoin*Factory.java | rg -A 15 "columnAccessHelper"

Length of output: 10966

🏁 Script executed:

#!/bin/bash
# Look at the full context of AsOfJoinLightRecordCursorFactory to see if it uses column access helper
cat core/src/main/java/io/questdb/griffin/engine/join/AsOfJoinLightRecordCursorFactory.java | head -100

Length of output: 4065

🏁 Script executed:

#!/bin/bash
# Check all places where join factories that take columnAccessHelper are instantiated
rg -nP "new AsOfJoin(Fast|Indexed|Memoized|SingleSymbol)RecordCursorFactory" --type=java -B 5 -A 1 core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java

Length of output: 2641

🏁 Script executed:

#!/bin/bash
# Let's see how columnAccessHelper is used in AsOfJoinSingleSymbolRecordCursorFactory
rg -nP "columnAccessHelper\." --type=java core/src/main/java/io/questdb/griffin/engine/join/AsOfJoinSingleSymbolRecordCursorFactory.java

Length of output: 138

🏁 Script executed:

#!/bin/bash
# Search for any other instantiations of AsOfJoinSingleSymbolRecordCursorFactory in the codebase
rg -nP "new\s+AsOfJoinSingleSymbolRecordCursorFactory" --type=java

Length of output: 170

🏁 Script executed:

#!/bin/bash
# Check if there are any assertions in the AsOfJoinSingleSymbolRecordCursorFactory constructor
rg -nP "assert.*columnAccessHelper" --type=java -A 2 -B 2 core/src/main/java/io/questdb/griffin/engine/join/

Length of output: 0

🏁 Script executed:

#!/bin/bash
# Check the same for other factories that use columnAccessHelper
rg -nP "(AsOfJoinFast|AsOfJoinIndexed|AsOfJoinMemoized)RecordCursorFactory.*assert" --type=java -A 3 core/src/main/java/io/questdb/griffin/engine/join/

Length of output: 0

🏁 Script executed:

#!/bin/bash
# Look at the test to see if it validates the single symbol optimization
cat core/src/test/java/io/questdb/test/griffin/AsOfJoinTest.java | grep -A 30 "testSingleSymbolAsOf"

Length of output: 3420

@bluestreak01, I've analyzed the code paths and found that while there is protection in place, it could be strengthened with explicit assertions. Here's what I found:

Current Protection

The code already prevents NoopColumnAccessHelper from being passed to AsOfJoinSingleSymbolRecordCursorFactory through a runtime check in SqlCodeGenerator.java:

// Line 2882-2884
AsofJoinColumnAccessHelper columnAccessHelper = createAsofColumnAccessHelper(masterMetadata, slaveMetadata, selfJoin);
boolean isOptimizable = columnAccessHelper != NoopColumnAccessHelper.INSTANCE;
if (isOptimizable) {
    // Only creates AsOfJoinSingleSymbolRecordCursorFactory when not NoopColumnAccessHelper
    master = new AsOfJoinSingleSymbolRecordCursorFactory(..., columnAccessHelper, ...);
}

If not optimizable, it falls back to AsOfJoinLightRecordCursorFactory (line 2908), which doesn't use columnAccessHelper at all.

Recommended Improvement

However, to fail fast and prevent future bugs, I recommend adding defensive assertions in the constructors:

AsOfJoinSingleSymbolRecordCursorFactory.java:

public AsOfJoinSingleSymbolRecordCursorFactory(
        CairoConfiguration configuration,
        RecordMetadata metadata,
        RecordCursorFactory masterFactory,
        RecordCursorFactory slaveFactory,
        @Transient ColumnTypes joinColumnTypes,
        @Transient ColumnTypes valueTypes,
        int columnSplit,
        int slaveSymbolColumnIndex,
        AsofJoinColumnAccessHelper columnAccessHelper,
        JoinContext joinContext,
        long toleranceInterval
) {
    super(metadata, joinContext, masterFactory, slaveFactory);
    assert columnAccessHelper != NoopColumnAccessHelper.INSTANCE : "NoopColumnAccessHelper cannot be used with AsOfJoinSingleSymbolRecordCursorFactory";
    this.slaveSymbolColumnIndex = slaveSymbolColumnIndex;
    this.columnAccessHelper = columnAccessHelper;
    // ... rest of constructor

The same assertion should be added to:

AsOfJoinFastRecordCursorFactory (line 64)
AsOfJoinIndexedRecordCursorFactory (line 69)
AsOfJoinMemoizedRecordCursorFactory (line 102)

These factories all use columnAccessHelper.getSlaveKey() or similar methods that would throw UnsupportedOperationException if NoopColumnAccessHelper were passed.

Why This Matters

Without the assertion, if a bug were introduced or the code were refactored, NoopColumnAccessHelper could be passed and would only fail at query execution time when getSlaveKey() is called (line 205 in the new factory). With the assertion, it would fail immediately during factory construction in development/testing.

Would you like me to proceed with adding these assertions?

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

…sof-linear-symbol

…nto mt_asof-linear-symbol" This reverts commit 55625e8, reversing changes made to d8ed506.

glasstiger · 2025-11-03T10:27:52Z

[PR Coverage check]

😍 pass : 179 / 200 (89.50%)

file detail

	path	covered line	new line	coverage
🔵	io/questdb/griffin/engine/join/AsofJoinColumnAccessHelper.java	0	1	00.00%
🔵	io/questdb/griffin/engine/table/AsyncFilteredRecordCursor.java	1	2	50.00%
🔵	io/questdb/griffin/engine/functions/date/GenerateSeriesTimestampStringRecordCursorFactory.java	14	20	70.00%
🔵	io/questdb/griffin/engine/join/AsOfJoinSingleSymbolRecordCursorFactory.java	116	128	90.62%
🔵	io/questdb/griffin/SqlCodeGenerator.java	44	45	97.78%
🔵	io/questdb/griffin/engine/join/SingleSymbolColumnAccessHelper.java	4	4	100.00%

mtopolnik and others added 5 commits October 30, 2025 15:24

Prototype

714570a

Small fix

3e1a500

Modified hashcode

98776a7

bugfix, integration, test

ffa59a8

Merge branch 'master' into mt_asof-linear-symbol

1835239

remove debris

8f3e640

coderabbitai bot reviewed Oct 31, 2025

View reviewed changes

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java Show resolved Hide resolved

core/src/test/java/io/questdb/test/griffin/AsOfJoinTest.java Show resolved Hide resolved

bluestreak01 added 2 commits October 31, 2025 01:58

fixes

8e29a58

Merge remote-tracking branch 'origin/master' into mt_asof-linear-symbol

fa6710d

bluestreak01 and others added 4 commits October 31, 2025 08:58

Merge branch 'master' into mt_asof-linear-symbol

38a1926

AsOfJoinDenseRecordCursorFactory

b65ba09

Fix log line

6a01908

Simplify expression in Ligt record cursor

19a6f62

RaphDal reviewed Oct 31, 2025

View reviewed changes

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java Outdated Show resolved Hide resolved

RaphDal reviewed Oct 31, 2025

View reviewed changes

core/src/main/java/io/questdb/griffin/SqlCodeGenerator.java Show resolved Hide resolved

bluestreak01 and others added 3 commits October 31, 2025 20:01

Merge branch 'master' into mt_asof-linear-symbol

760b58a

fix bugs found in review

d3b969c

oops

d8ed506

mtopolnik added 5 commits November 1, 2025 14:55

Fix some bugs

901fabe

Update test

a8c3bb7

Fix some bugs

5eceb55

Fix AsOfJoinFuzzTest bug

acaebfc

Fix bug

ae54424

mtopolnik added 2 commits November 2, 2025 16:15

Merge remote-tracking branch 'origin/mt_asof-linear-symbol' into mt_a…

55625e8

…sof-linear-symbol

Revert "Merge remote-tracking branch 'origin/mt_asof-linear-symbol' i…

48f7ab5

…nto mt_asof-linear-symbol" This reverts commit 55625e8, reversing changes made to d8ed506.

RaphDal approved these changes Nov 3, 2025

View reviewed changes

bluestreak01 merged commit d4016f5 into master Nov 3, 2025
37 checks passed

bluestreak01 deleted the mt_asof-linear-symbol branch November 3, 2025 11:26

coderabbitai bot mentioned this pull request Nov 6, 2025

chore(core): refactor and clean up ASOF/LT JOIN code #6348

Merged

coderabbitai bot mentioned this pull request Nov 23, 2025

fix(sql): fix a bug that made some ASOF JOIN queries fail with an internal error #6433

Merged

coderabbitai bot mentioned this pull request Dec 4, 2025

fix(sql): fix empty result set from limit queries #6504

Merged

coderabbitai bot mentioned this pull request Jan 13, 2026

fix(sql): fix ASOF JOIN crash when ON clause has symbol and other columns #6634

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

perf(sql): improve performance of linear as-of join algo#6338

perf(sql): improve performance of linear as-of join algo#6338
bluestreak01 merged 22 commits intomasterfrom
mt_asof-linear-symbol

bluestreak01 commented Oct 31, 2025 •

edited

Loading

Uh oh!

coderabbitai bot commented Oct 31, 2025 •

edited

Loading

Review skipped

Uh oh!

bluestreak01 commented Oct 31, 2025

Uh oh!

coderabbitai bot commented Oct 31, 2025

Uh oh!

coderabbitai bot left a comment

Uh oh!

Uh oh!

Uh oh!

bluestreak01 commented Oct 31, 2025

Uh oh!

coderabbitai bot commented Oct 31, 2025

Uh oh!

Uh oh!

Uh oh!

bluestreak01 commented Nov 1, 2025

Uh oh!

coderabbitai bot commented Nov 1, 2025

Uh oh!

glasstiger commented Nov 3, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

bluestreak01 commented Oct 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

coderabbitai bot commented Oct 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Walkthrough

Changes

Sequence Diagram(s)

Estimated code review effort

Possibly related PRs

Suggested labels

Suggested reviewers

Pre-merge checks and finishing touches

Uh oh!

bluestreak01 commented Oct 31, 2025

Uh oh!

coderabbitai bot commented Oct 31, 2025

Uh oh!

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

bluestreak01 commented Oct 31, 2025

Uh oh!

coderabbitai bot commented Oct 31, 2025

Uh oh!

Uh oh!

Uh oh!

bluestreak01 commented Nov 1, 2025

Uh oh!

coderabbitai bot commented Nov 1, 2025

Current Protection

Recommended Improvement

Why This Matters

Uh oh!

glasstiger commented Nov 3, 2025

[PR Coverage check]

file detail

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

bluestreak01 commented Oct 31, 2025 •

edited

Loading

coderabbitai bot commented Oct 31, 2025 •

edited

Loading