[SPARK-57053][SQL][TESTS] Expand time-window.sql SQL test coverage#56098
Open
vladimirg-db wants to merge 1 commit into
Open
[SPARK-57053][SQL][TESTS] Expand time-window.sql SQL test coverage#56098vladimirg-db wants to merge 1 commit into
vladimirg-db wants to merge 1 commit into
Conversation
### What changes were proposed in this pull request? Adds ~70 new positive and negative test cases to `time-window.sql`, covering gaps in the existing tumbling/sliding/session window coverage: - Tumbling and sliding window edge cases: NTZ time column, subsecond precision, null timestamps, negative/zero/non-literal/unparseable duration, slide >= duration, startTime > slide, ROLLUP/CUBE/GROUPING SETS variants for sliding windows. - `window_time()` function: basic event-time extraction, NTZ, wrong argument type/arity, and struct-input rejection paths. - `session_window` parity with tumbling/sliding coverage: basic aggregation, conditional gap duration (from the docs example), column-valued gap duration, NTZ, null timestamps, subsecond, negative/zero gap, wrong arg count, multiple-window error, and acceptance/rejection of session_window in WHERE / QUALIFY / SELECT *. - Nested and stacked windows: nested window in GROUP BY, nested literal windows in SELECT/GROUP BY ALL, nested window via subquery, stacked sibling windows. - Window placement: in simple projection, in SELECT with GROUP BY ALL, in select+group by combinations, in EXISTS subquery, in PIVOT aggregate. Most query texts are imported verbatim from the Reyden `time-window-tests` golden file. One Reyden-only case (`window_overlap_exceeds_max`, an internal cap that Spark does not enforce) is skipped because Spark accepts the query and generates an extremely large `Expand` plan. Two cases whose Reyden names imply rejection but which Spark accepts are renamed to drop the `_rejected` suffix. ### Why are the changes needed? `time-window.sql` previously covered only ~23 cases, all GROUP BY variants of `window()` / `session_window()`. Resolver work on the single-pass analyzer surfaced multiple uncovered code paths (`window_time()` extraction, NTZ inputs, struct-input handling, session window with conditional/column gap). This PR expands the suite so future analyzer changes have a broader regression surface. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? Existing test (`time-window.sql` + `time-window.sql_analyzer_test`). Goldens regenerated with `SPARK_GENERATE_GOLDEN_FILES=1`. ### Was this patch authored or co-authored using generative AI tooling? Generated-by: Claude Code (claude-opus-4-7)
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.
What changes were proposed in this pull request?
Adds ~70 new positive and negative test cases to
time-window.sql, covering gaps in the existing tumbling/sliding/session window coverage:window_time()function: basic event-time extraction, NTZ, wrong argument type/arity, and struct-input rejection paths.session_windowparity with tumbling/sliding coverage: basic aggregation, conditional gap duration (from the docs example), column-valued gap duration, NTZ, null timestamps, subsecond, negative/zero gap, wrong arg count, multiple-window error, and acceptance/rejection of session_window in WHERE / QUALIFY / SELECT *.Most query texts are imported verbatim from the Reyden
time-window-testsgolden file. One Reyden-only case (window_overlap_exceeds_max, an internal cap that Spark does not enforce) is skipped because Spark accepts the query and generates an extremely largeExpandplan. Two cases whose Reyden names imply rejection but which Spark accepts are renamed to drop the_rejectedsuffix.Why are the changes needed?
time-window.sqlpreviously covered only ~23 cases, all GROUP BY variants ofwindow()/session_window(). Resolver work on the single-pass analyzer surfaced multiple uncovered code paths (window_time()extraction, NTZ inputs, struct-input handling, session window with conditional/column gap). This PR expands the suite so future analyzer changes have a broader regression surface.Does this PR introduce any user-facing change?
No.
How was this patch tested?
Existing test (
time-window.sql+time-window.sql_analyzer_test). Goldens regenerated withSPARK_GENERATE_GOLDEN_FILES=1.Was this patch authored or co-authored using generative AI tooling?
Generated-by: Claude Code (claude-opus-4-7)
What changes were proposed in this pull request?
Why are the changes needed?
Does this PR introduce any user-facing change?
How was this patch tested?
Was this patch authored or co-authored using generative AI tooling?