perf: Element-wise comparison only for tolerance-requiring data types#26
Open
Marius Merkle (MariusMerkleQC) wants to merge 8 commits intobenchmarkfrom
Open
perf: Element-wise comparison only for tolerance-requiring data types#26Marius Merkle (MariusMerkleQC) wants to merge 8 commits intobenchmarkfrom
Marius Merkle (MariusMerkleQC) wants to merge 8 commits intobenchmarkfrom
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## benchmark #26 +/- ##
===========================================
Coverage 100.00% 100.00%
===========================================
Files 10 10
Lines 758 780 +22
===========================================
+ Hits 758 780 +22 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Copilot started reviewing on behalf of
Marius Merkle (MariusMerkleQC)
March 27, 2026 23:03
View session
There was a problem hiding this comment.
Pull request overview
This PR optimizes condition_equal_columns for nested list/array columns by avoiding the expensive element-wise comparison path when tolerances/special handling aren’t needed, and updates the performance benchmark accordingly.
Changes:
- Add
_needs_element_wise_comparison()(plus helpers) to decide when list/array columns require element-wise comparison. - Shortcut list/array comparisons to
eq_missing()when element-wise handling is deemed unnecessary. - Update the performance test to assert comparable performance for
list<i64>comparisons.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
diffly/_conditions.py |
Introduces dtype-based gating to skip element-wise list/array comparisons unless tolerances/special handling are needed. |
tests/test_performance.py |
Updates benchmark expectations to ensure the optimized path is not significantly slower than direct eq_missing(). |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Marius Merkle (MariusMerkleQC)
requested review from
EgeKaraismailogluQC and
Oliver Borchert (borchero)
as code owners
Comment on lines
124
to
139
| # Otherwise, we simply compare all fields independently | ||
| return pl.all_horizontal( | ||
| [ | ||
| _compare_columns( | ||
| col_left=col_left.struct[field], | ||
| col_right=col_right.struct[field], | ||
| dtype_left=fields_left[field], | ||
| dtype_right=fields_right[field], | ||
| max_list_length=max_list_length, | ||
| abs_tol=abs_tol, | ||
| rel_tol=rel_tol, | ||
| abs_tol_temporal=abs_tol_temporal, | ||
| ) | ||
| for field in fields_left | ||
| ] | ||
| ) |
Collaborator
Author
There was a problem hiding this comment.
Should we also shortcut the comparison here?
# Otherwise, we simply compare all fields independently
if _needs_element_wise_comparison(dtype_left, dtype_right):
return pl.all_horizontal(
[
_compare_columns(
col_left=col_left.struct[field],
col_right=col_right.struct[field],
dtype_left=fields_left[field],
dtype_right=fields_right[field],
max_list_length=max_list_length,
abs_tol=abs_tol,
rel_tol=rel_tol,
abs_tol_temporal=abs_tol_temporal,
)
for field in fields_left
]
)
return col_left.eq_missing(col_right)
Collaborator
Author
There was a problem hiding this comment.
Ok, I added a performance test in #25, which already passes without the above optimization, so we should definitely not add it.
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
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.
Motivation
See this comment.
Changes
Introduce a function
_needs_element_wise_comparisonthat checks whether element-wise comparison needs to be performed; this is the case for(1) float vs numeric columns -> absolute and relative tolerances apply (->
_is_float_numeric_pair())(2) temporal columns -> absolute temporal tolerance applies (->
_is_temporal_pair())In all other cases, naive comparison suffices, and this shortcut is taken if the above helper returns
False. This avoids the expensive_compare_sequence_columns(). The performance improvement can be seen in the benchmark test.