Eventual-Inc/Daft - CodSpeed

Daft

Blog Docs Changelog

Performance History

Latest Results

feat(dashboard): render repartition spec in human-readable form

BABTUNA:feat/dashboard-human-readable-repartition

1 hour ago

test(list_filter): add non-Boolean predicate and nested-list cases

aaron-ang:expr-list-filter

2 hours ago

style(dashboard): use unwrap_or for static fallback string

claude/integrate-task-events-xazrE

2 hours ago

match rust docstring to python

gweaverbiodev:gw/hamming

3 hours ago

fix(checkpoint): use strip_file_uri_to_path in put_bytes for Windows paths Follow-up to #6791. The previous PR fixed the test-helper to produce a canonical file:///C:/Users/... URL on Windows, which unblocked the reader path (daft-io's glob + LocalSource::get both route through strip_file_uri_to_path which drops the leading slash before the drive letter via #[cfg(windows)]). But S3CheckpointStore::put_bytes still did its own path.strip_prefix("file://") and handed `/C:/Users/...` to Path::new(...), which Windows rejects with os error 123 ("filename, directory name, or volume label syntax is incorrect"). Result: reads worked but writes failed at the first create_dir_all, regressing the Windows coverage job from 3 passed / 11 failed → 0 passed / 14 failed. Switch to daft_io::strip_file_uri_to_path — the canonical helper that already handles both POSIX (file:///tmp/foo → /tmp/foo) and Windows (file:///C:/Users/... → C:/Users/...) via the existing strip_leading_slash_before_drive path. Hard to unit-test locally since the failing shape is Windows-specific; relying on the existing Windows rust-tests CI to verify. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

rohit/bugfix/s3-checkpoint-windows-put

4 hours ago

feat(functions): add great_circle_distance function (#6754) ## Changes Made Adds `great_circle_distance` function. This computes the great circle distance via haversine approximation. Couple of decisions: - Keep name `great_circle_distance` vs. `haversine_distance`, I see mixed usage of these names in other libraries. - I have not yet added to `expressions` because this seems like a strange method to have as an expression? - Does this live under `daft-functions/spatial` or new `daft-functions-spatial`. I know there is future work for geospatial support that would probably absorb this function at some point, so maybe it is ok as-is. ## Related Issues https://github.com/Eventual-Inc/Daft/issues/6719

main

5 hours ago

fix serde import

gweaverbiodev:gw/great-circle

6 hours ago

fix(udf): filter divider padding + use deque for pending lines Addresses Greptile feedback on PR #6793. - trace_output() now skips bare "\n" lines, which are the padding we inject before _OUTPUT_DIVIDER to guarantee the sentinel lands on its own line. Previously that padding surfaced as an empty line per batch and got printed as a spurious "[`udf` Worker #N] " prefix through the Rust caller. - _pending_lines is now a collections.deque with popleft() instead of a list with pop(0), which was O(n) per pop. Under a UDF that floods stderr, the wait() loop can build up a large queue and the subsequent trace_output() drain would be O(n^2). Both tests from the earlier commit (silent UDF, pipe-buffer, and divider-merge) pass. Existing use_process=True tests unaffected. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

rohit/bugfix/df-1963-use-process-deadlocks

6 hours ago

Latest Branches

0%

feat(dashboard): render Repartition spec in human-readable form#6798

2 hours ago

8f2c5ae

BABTUNA:feat/dashboard-human-readable-repartition

+1%

feat(list): add list_filter expression#6769

2 hours ago

1cb80a6

aaron-ang:expr-list-filter

0%

feat(dashboard): Draft tasks view for Flotilla#6783

2 hours ago

349aeac

claude/integrate-task-events-xazrE

© 2026 CodSpeed Technology

Home Terms Privacy Docs