Add a logger#680
Open
Intron7 wants to merge 2 commits into
Open
Conversation
Member
Author
|
@coderabbitai review |
✅ Actions performedReview triggered.
|
📝 WalkthroughSummary by CodeRabbit
WalkthroughThis PR introduces a scverse-style logging system for rapids-singlecell with Pydantic settings-based verbosity control. It adds a new ChangesLogging infrastructure and instrumentation
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Possibly related PRs
Suggested reviewers
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/rapids_singlecell/logging.py`:
- Around line 104-116: The code currently uses truthiness checks on optional
formatter fields which skips valid falsy values (e.g., timedelta(0)); update the
conditionals to explicit None checks: replace "if record.time_passed:" with "if
record.time_passed is not None:" and keep the microseconds branch as an explicit
check (e.g., "if record.time_passed.microseconds != 0:" or similar) so
zero-duration is handled, and replace "if record.deep:" with "if record.deep is
not None:"; ensure the logic inside (replacing "{time_passed}" in record.msg,
adjusting self._style._fmt, and appending ": {deep}" behavior) remains unchanged
apart from these None checks.
In `@src/rapids_singlecell/preprocessing/_neighbors/__init__.py`:
- Around line 190-193: The auto-adjustment block that sets n_neighbors when
n_neighbors > adata.shape[0] needs an explicit guard for empty AnnData: check if
adata.shape[0] == 0 before adjusting and raise a clear ValueError (or log and
exit) instead of proceeding to set n_neighbors to 1; update the code around the
n_neighbors adjustment (the block that references adata.shape and n_neighbors in
src/rapids_singlecell/preprocessing/_neighbors/__init__.py) so empty input is
detected early and handled with an informative error message to avoid later KNN
failures.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: ce41bad9-ff8d-4e9c-ac3c-2dbc9a4ddc78
📒 Files selected for processing (20)
src/rapids_singlecell/__init__.pysrc/rapids_singlecell/_settings/__init__.pysrc/rapids_singlecell/_settings/verbosity.pysrc/rapids_singlecell/decoupler_gpu/_helper/_log.pysrc/rapids_singlecell/logging.pysrc/rapids_singlecell/preprocessing/_harmony_integrate.pysrc/rapids_singlecell/preprocessing/_hvg/__init__.pysrc/rapids_singlecell/preprocessing/_neighbors/__init__.pysrc/rapids_singlecell/preprocessing/_normalize.pysrc/rapids_singlecell/preprocessing/_pca.pysrc/rapids_singlecell/preprocessing/_regress_out.pysrc/rapids_singlecell/preprocessing/_scale.pysrc/rapids_singlecell/preprocessing/_scrublet/__init__.pysrc/rapids_singlecell/preprocessing/_scrublet/core.pysrc/rapids_singlecell/preprocessing/_simple.pysrc/rapids_singlecell/tools/_clustering.pysrc/rapids_singlecell/tools/_rank_genes_groups/__init__.pysrc/rapids_singlecell/tools/_tsne.pysrc/rapids_singlecell/tools/_umap.pytests/test_logging.py
Comment on lines
+104
to
+116
| if record.time_passed: | ||
| if record.time_passed.microseconds: | ||
| record.time_passed = timedelta( | ||
| seconds=int(record.time_passed.total_seconds()) | ||
| ) | ||
| if "{time_passed}" in record.msg: | ||
| record.msg = record.msg.replace( | ||
| "{time_passed}", str(record.time_passed) | ||
| ) | ||
| else: | ||
| self._style._fmt += " ({time_passed})" | ||
| if record.deep: | ||
| record.msg = f"{record.msg}: {record.deep}" |
There was a problem hiding this comment.
Use explicit None checks for optional formatter fields.
Please avoid truthiness checks on optional values here. This also fixes the zero-duration edge case (timedelta(0)) being skipped.
Suggested patch
- if record.time_passed:
+ if record.time_passed is not None:
if record.time_passed.microseconds:
record.time_passed = timedelta(
seconds=int(record.time_passed.total_seconds())
)
if "{time_passed}" in record.msg:
record.msg = record.msg.replace(
"{time_passed}", str(record.time_passed)
)
else:
self._style._fmt += " ({time_passed})"
- if record.deep:
+ if record.deep is not None:
record.msg = f"{record.msg}: {record.deep}"📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| if record.time_passed: | |
| if record.time_passed.microseconds: | |
| record.time_passed = timedelta( | |
| seconds=int(record.time_passed.total_seconds()) | |
| ) | |
| if "{time_passed}" in record.msg: | |
| record.msg = record.msg.replace( | |
| "{time_passed}", str(record.time_passed) | |
| ) | |
| else: | |
| self._style._fmt += " ({time_passed})" | |
| if record.deep: | |
| record.msg = f"{record.msg}: {record.deep}" | |
| if record.time_passed is not None: | |
| if record.time_passed.microseconds: | |
| record.time_passed = timedelta( | |
| seconds=int(record.time_passed.total_seconds()) | |
| ) | |
| if "{time_passed}" in record.msg: | |
| record.msg = record.msg.replace( | |
| "{time_passed}", str(record.time_passed) | |
| ) | |
| else: | |
| self._style._fmt += " ({time_passed})" | |
| if record.deep is not None: | |
| record.msg = f"{record.msg}: {record.deep}" |
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/rapids_singlecell/logging.py` around lines 104 - 116, The code currently
uses truthiness checks on optional formatter fields which skips valid falsy
values (e.g., timedelta(0)); update the conditionals to explicit None checks:
replace "if record.time_passed:" with "if record.time_passed is not None:" and
keep the microseconds branch as an explicit check (e.g., "if
record.time_passed.microseconds != 0:" or similar) so zero-duration is handled,
and replace "if record.deep:" with "if record.deep is not None:"; ensure the
logic inside (replacing "{time_passed}" in record.msg, adjusting
self._style._fmt, and appending ": {deep}" behavior) remains unchanged apart
from these None checks.
Comment on lines
+190
to
+193
| if n_neighbors > adata.shape[0]: | ||
| n_neighbors = 1 + int(0.5 * adata.shape[0]) | ||
| logg.warning(f"n_obs too small: adjusting to `n_neighbors = {n_neighbors}`") | ||
|
|
There was a problem hiding this comment.
Guard empty AnnData before neighbor auto-adjustment.
When adata.shape[0] == 0, the new adjustment produces n_neighbors = 1, which is still invalid and fails later in KNN with a less direct error. Add an explicit early check for empty input.
💡 Proposed fix
+ if adata.shape[0] == 0:
+ raise ValueError("Cannot compute neighbors on empty AnnData (`n_obs == 0`).")
+
if n_neighbors > adata.shape[0]:
n_neighbors = 1 + int(0.5 * adata.shape[0])
logg.warning(f"n_obs too small: adjusting to `n_neighbors = {n_neighbors}`")📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| if n_neighbors > adata.shape[0]: | |
| n_neighbors = 1 + int(0.5 * adata.shape[0]) | |
| logg.warning(f"n_obs too small: adjusting to `n_neighbors = {n_neighbors}`") | |
| if adata.shape[0] == 0: | |
| raise ValueError("Cannot compute neighbors on empty AnnData (`n_obs == 0`).") | |
| if n_neighbors > adata.shape[0]: | |
| n_neighbors = 1 + int(0.5 * adata.shape[0]) | |
| logg.warning(f"n_obs too small: adjusting to `n_neighbors = {n_neighbors}`") |
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/rapids_singlecell/preprocessing/_neighbors/__init__.py` around lines 190
- 193, The auto-adjustment block that sets n_neighbors when n_neighbors >
adata.shape[0] needs an explicit guard for empty AnnData: check if
adata.shape[0] == 0 before adjusting and raise a clear ValueError (or log and
exit) instead of proceeding to set n_neighbors to 1; update the code around the
n_neighbors adjustment (the block that references adata.shape and n_neighbors in
src/rapids_singlecell/preprocessing/_neighbors/__init__.py) so empty input is
detected early and handled with an informative error message to avoid later KNN
failures.
|
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
2 participants
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.
This adds a scanpy style logger to rsc.