fix: Preserve sort before take in CTE#5638
Merged
Conversation
Replace internal compiler error with user-facing error message when a bare lambda like `x -> y` is used at the top level. Now produces: "expected a table, but found a function" Fixes PRQL#4280 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
When using `group {} (sort a | take 1)`, the inner sort was being
incorrectly dropped because `sort_undone` was unconditionally set
to true for all groups. Now only set `sort_undone` when there's an
actual partition (non-empty group keys).
Fixes PRQL#5100
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The error message for bare lambda expressions (e.g., `x -> y`) was fixed in PR PRQL#5634. Move the test from bad_error_messages.rs to error_messages.rs since the error is now clear and helpful. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
…QL#4432 Add inline snapshot tests to prevent regressions for bugs that have been fixed: - test_source_column_name: Using `source` as column name no longer causes "Ambiguous name" error (PRQL#5094) - test_column_inference_with_into: Column inference works correctly with `into` statement (PRQL#4723) - test_distinct_on_columns_propagated: DISTINCT ON propagates necessary columns to CTE (PRQL#4432) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
When `sort | take` was lowered to RQ, the sort was embedded in the Take struct. But postprocessing only tracked explicit SqlTransform::Sort, ignoring the embedded sort. This caused ORDER BY to be lost in CTEs. Now extracts `take.sort` for simple takes (no partition) where the sort was merged during RQ lowering. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
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.
Summary
sort | takebefore an aggregation should enforce the sort in the CTE #5401 wheresort | takebefore an aggregation lost the ORDER BY in CTEssort | takeintoTake { sort: [...] }, postprocessing now extracts the embedded sort for simple takesTest plan
test_sort_take_before_aggregate🤖 Generated with Claude Code