test(consistent): replace Cartesian product with curated matrices for descriptor tests#5378
Open
njzjz-bot wants to merge 9 commits intodeepmodeling:masterfrom
Open
Conversation
… descriptor tests Add an explicit parameterized_cases helper for heavy consistency suites and switch the descriptor DPA1, DPA2, and SeA tests to reviewable curated matrices. This keeps the existing parameterized Cartesian-product helper unchanged while reducing the default PR CI matrix to baseline, OFAT, and selected interaction cases. Authored by OpenClaw (model: gpt-5.4)
Contributor
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughAdds a shared subclass generator Changes
Sequence Diagram(s)sequenceDiagram
participant Author as Test Author
participant OldDec as "@parameterized"
participant NewDec as "@parameterized_cases"
participant Gen as _parameterized_with_cases
participant Module as Test Module Namespace
rect rgba(100,150,255,0.5)
Note over OldDec,Module: Old flow (computed Cartesian product)
Author->>OldDec: provide per-argument lists
OldDec->>Gen: compute Cartesian product -> list of tuples
Gen->>Gen: create subclass per tuple, set ClassVar param
Gen->>Module: inject subclass with sanitized name
end
rect rgba(100,200,100,0.5)
Note over NewDec,Module: New flow (explicit curated tuples)
Author->>NewDec: provide explicit case tuples (*CURATED_CASES)
NewDec->>Gen: forward provided tuples
Gen->>Gen: create subclass per tuple, set ClassVar param
Gen->>Module: inject subclass with sanitized name
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes 🚥 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 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}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
source/tests/consistent/common.py (1)
684-696: Moveimport reto module level.The
remodule is imported inside the loop on every iteration, which is inefficient and violates standard Python import conventions. Move the import to the top of the file with other standard library imports.♻️ Proposed fix
Add at the top of the file (around line 2):
import reThen update the function:
# generate a safe name for the class parts = [] for x in pp: s = str(x) # replace non-alnum with underscore, collapse multiple underscores - import re s = re.sub(r'[^a-zA-Z0-9_]', '_', s) s = re.sub(r'_+', '_', s)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/common.py` around lines 684 - 696, The loop in which pp is processed imports re on every iteration; move the import to module level (add a single "import re" with the other standard imports at the top of the file) and remove the inline "import re" inside the loop; update the block that builds parts (variables: pp, s, parts, name, base_class) to use the module-level re for re.sub calls and leave the rest of the logic unchanged.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@source/tests/consistent/common.py`:
- Around line 684-696: The loop in which pp is processed imports re on every
iteration; move the import to module level (add a single "import re" with the
other standard imports at the top of the file) and remove the inline "import re"
inside the loop; update the block that builds parts (variables: pp, s, parts,
name, base_class) to use the module-level re for re.sub calls and leave the rest
of the logic unchanged.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: aef3ec2d-e1af-44aa-ae63-415d691fa1a3
📒 Files selected for processing (4)
source/tests/consistent/common.pysource/tests/consistent/descriptor/test_dpa1.pysource/tests/consistent/descriptor/test_dpa2.pysource/tests/consistent/descriptor/test_se_e2_a.py
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #5378 +/- ##
==========================================
- Coverage 82.37% 80.39% -1.98%
==========================================
Files 812 812
Lines 83560 83476 -84
Branches 4091 4050 -41
==========================================
- Hits 68832 67113 -1719
- Misses 13507 15139 +1632
- Partials 1221 1224 +3 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Contributor
There was a problem hiding this comment.
Pull request overview
This PR reduces CI cost in the descriptor consistency suites by replacing large Cartesian-product parameter grids with explicit curated test matrices, enabled by a new parameterized_cases decorator helper.
Changes:
- Added
parameterized_cases(*cases)to support explicit curated parameter tuples without changing existingparameterized(*attrs)behavior. - Replaced Cartesian-product parametrization with curated case lists in
test_se_e2_a.py,test_dpa1.py, andtest_dpa2.py. - Reused the curated matrices for corresponding descriptor API consistency tests.
Reviewed changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
source/tests/consistent/common.py |
Adds parameterized_cases and refactors shared parametrization implementation into _parameterized_with_cases. |
source/tests/consistent/descriptor/test_se_e2_a.py |
Introduces curated matrices for SE_A descriptor and API tests to avoid full Cartesian expansion. |
source/tests/consistent/descriptor/test_dpa1.py |
Replaces Cartesian parametrization with curated DPA1 case tuples for both descriptor and API tests. |
source/tests/consistent/descriptor/test_dpa2.py |
Replaces a very large Cartesian grid with curated DPA2 case tuples for both descriptor and API tests. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Match the curated DPA2 case tuples with the unpack order used by the descriptor tests so direct_dist and update_g1_has_conv map to the intended fields. Also move the regex import used for parameterized test-class name sanitization to module scope. Authored by OpenClaw (model: gpt-5.4)
Contributor
Author
|
Thanks — both comments were valid, and I pushed a follow-up fix. What changed:
Validation:
— OpenClaw 2026.3.8 (model: gpt-5.4) |
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (2)
source/tests/consistent/descriptor/test_dpa2.py (2)
423-451: Optional: Extract param unpacking to reduce duplication.This 27-variable unpacking pattern is repeated 9 times across the file. For properties that only need a subset (e.g.,
skip_pdonly usesprecision), the full unpacking adds noise. Consider extracting frequently-needed fields as cached properties or using a helper dataclass.💡 Example: Extract precision as a property
`@property` def _precision(self) -> str: """Extract precision from param tuple (position 23).""" return self.param[23]Then simplify
skip_pd:`@property` def skip_pd(self) -> bool: - ( - repinit_tebd_input_mode, - ... # 27 variables - ) = self.param - return not INSTALLED_PD or precision == "bfloat16" + return not INSTALLED_PD or self._precision == "bfloat16"🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/descriptor/test_dpa2.py` around lines 423 - 451, The test file repeatedly unpacks a 27-element tuple from self.param (e.g., the large unpack in the shown block) which is duplicated across many tests; extract frequently-used fields into small cached properties (for example add properties like _precision -> return self.param[23], _repformer_update_style -> return self.param[12], etc.) or define a small dataclass mapping tuple positions to named attributes and instantiate it once (e.g., self.params = Params.from_tuple(self.param)); then replace the repeated full unpacking and uses in methods like skip_pd and other helpers to reference these properties/attributes (self._precision or self.params.precision) to reduce duplication and improve readability. Ensure tests that relied on tuple order still pass by mapping the same indices when creating the properties/dataclass.
67-96: Consider documenting the tuple field order for maintainability.The curated cases use bare positional tuples with 27 elements each. Without inline documentation, maintainers must cross-reference the unpacking code (e.g., lines 423-451) to understand what each position represents. A comment header above the tuple would clarify the field ordering.
📝 Suggested documentation
+# DPA2_CURATED_CASES field order: +# (repinit_tebd_input_mode, repinit_set_davg_zero, repinit_type_one_side, +# repinit_use_three_body, repformer_direct_dist, repformer_update_g1_has_conv, +# repformer_update_g1_has_drrd, repformer_update_g1_has_grrg, repformer_update_g1_has_attn, +# repformer_update_g2_has_g1g1, repformer_update_g2_has_attn, repformer_update_h2, +# repformer_attn2_has_gate, repformer_update_style, repformer_update_residual_init, +# repformer_set_davg_zero, repformer_trainable_ln, repformer_ln_eps, +# repformer_use_sqrt_nnei, repformer_g1_out_conv, repformer_g1_out_mlp, +# smooth, exclude_types, precision, add_tebd_to_repinit_out, use_econf_tebd, use_tebd_bias) DPA2_CURATED_CASES = ( ( "concat",🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/descriptor/test_dpa2.py` around lines 67 - 96, Add an inline comment header above DPA2_CURATED_CASES that documents the field order for the 27-tuple (e.g., list each index with the corresponding semantic name used when the tuple is unpacked in the test — reference the unpacking variables in the test at lines ~423-451) so future maintainers can read the mapping without cross-referencing code; include each position number and its name/type (e.g., 0: "mode"/"concat", 1: booleanFlagX, ... 26: booleanFlagY) and mention expected types like "str", "bool", "float", "list" to make the tuple structure explicit.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@source/tests/consistent/descriptor/test_dpa2.py`:
- Around line 423-451: The test file repeatedly unpacks a 27-element tuple from
self.param (e.g., the large unpack in the shown block) which is duplicated
across many tests; extract frequently-used fields into small cached properties
(for example add properties like _precision -> return self.param[23],
_repformer_update_style -> return self.param[12], etc.) or define a small
dataclass mapping tuple positions to named attributes and instantiate it once
(e.g., self.params = Params.from_tuple(self.param)); then replace the repeated
full unpacking and uses in methods like skip_pd and other helpers to reference
these properties/attributes (self._precision or self.params.precision) to reduce
duplication and improve readability. Ensure tests that relied on tuple order
still pass by mapping the same indices when creating the properties/dataclass.
- Around line 67-96: Add an inline comment header above DPA2_CURATED_CASES that
documents the field order for the 27-tuple (e.g., list each index with the
corresponding semantic name used when the tuple is unpacked in the test —
reference the unpacking variables in the test at lines ~423-451) so future
maintainers can read the mapping without cross-referencing code; include each
position number and its name/type (e.g., 0: "mode"/"concat", 1: booleanFlagX,
... 26: booleanFlagY) and mention expected types like "str", "bool", "float",
"list" to make the tuple structure explicit.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: b55ab950-4d03-4269-a5eb-85b4b608f888
📒 Files selected for processing (2)
source/tests/consistent/common.pysource/tests/consistent/descriptor/test_dpa2.py
🚧 Files skipped from review as they are similar to previous changes (1)
- source/tests/consistent/common.py
Run pre-commit on the PR branch and apply the resulting formatting-only changes from isort and ruff-format. Authored by OpenClaw (model: gpt-5.4)
Replace anonymous curated tuple lists in the heavy consistency descriptor suites with named baseline dictionaries plus explicit override helpers. This keeps the reduced-matrix strategy while making each curated case reviewable and documenting the intended coverage. It also makes the DPA2 case ordering explicit so repformer_direct_dist and repformer_update_g1_has_conv cannot be silently confused when editing boolean-heavy cases. Authored by OpenClaw (model: gpt-5.4)
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
source/tests/consistent/descriptor/test_se_e2_a.py (1)
14-23:⚠️ Potential issue | 🔴 CriticalKeep importing
parameterizedhere.
TestSeAStatstill uses@parameterized(...)at Line 382, so removing that symbol makes this module fail during import before any tests are collected.🛠️ Proposed fix
from ..common import ( INSTALLED_ARRAY_API_STRICT, INSTALLED_JAX, INSTALLED_PD, INSTALLED_PT, INSTALLED_PT_EXPT, INSTALLED_TF, CommonTest, + parameterized, parameterized_cases, )🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/descriptor/test_se_e2_a.py` around lines 14 - 23, The module removed the imported symbol parameterized but TestSeAStat still uses the `@parameterized` decorator, causing import-time failure; restore parameterized in the import list (alongside INSTALLED_ARRAY_API_STRICT, INSTALLED_JAX, etc.) so that the TestSeAStat decorator resolves correctly during module import and tests can be collected.
♻️ Duplicate comments (1)
source/tests/consistent/descriptor/test_dpa2.py (1)
67-161:⚠️ Potential issue | 🟠 MajorStop encoding DPA2 cases as anonymous 27-slot tuples.
This helper forces every consumer to destructure the full case just to read one field. In the current patch,
skip_pt,skip_dp,skip_tf, andsetUpdo not use any of those locals, whileskip_pd,rtol, andatolonly needprecision; Ruff is already flagging that asRUF059at Lines 274, 307, 340, 373, 450, 556, and 595. Returning a keyed case object here would make the field-order fix local, remove the current lint failures, and make future boolean reorder mistakes much harder. As per coding guidelines**/*.py: Always runruff check .andruff format .before committing changes or CI will fail.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/descriptor/test_dpa2.py` around lines 67 - 161, The test currently encodes DPA2 cases as 27-slot tuples (DPA2_CASE_FIELDS, DPA2_BASELINE_CASE, dpa2_case, DPA2_CURATED_CASES) which forces callers to destructure and is causing RUF059 lint failures for unused tuple slots; change dpa2_case to return a keyed mapping (e.g., a dict or lightweight dataclass) instead of a tuple, keep DPA2_BASELINE_CASE as the canonical dict, update DPA2_CURATED_CASES to call dpa2_case(...) and yield dicts, and update all consumers to access fields by name (e.g., case["precision"] or case.precision) rather than positional unpacking; after making these changes run ruff check . and ruff format . to clear the lint errors.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@source/tests/consistent/common.py`:
- Around line 676-699: The generated test-class names in
_parameterized_with_cases can collide after sanitization and silently overwrite
earlier TestClass entries in class_module; update the decorator to detect name
collisions and avoid silent overwrites by either appending a stable unique
suffix (e.g., an incrementing counter or a hash derived from pp) to the computed
name or by raising a clear error when the target name already exists in
class_module; implement the check around the assignment to class_module[name] so
that TestClass instances are preserved and collisions are handled
deterministically (refer to symbols _parameterized_with_cases, TestClass,
class_module, name, and pp).
---
Outside diff comments:
In `@source/tests/consistent/descriptor/test_se_e2_a.py`:
- Around line 14-23: The module removed the imported symbol parameterized but
TestSeAStat still uses the `@parameterized` decorator, causing import-time
failure; restore parameterized in the import list (alongside
INSTALLED_ARRAY_API_STRICT, INSTALLED_JAX, etc.) so that the TestSeAStat
decorator resolves correctly during module import and tests can be collected.
---
Duplicate comments:
In `@source/tests/consistent/descriptor/test_dpa2.py`:
- Around line 67-161: The test currently encodes DPA2 cases as 27-slot tuples
(DPA2_CASE_FIELDS, DPA2_BASELINE_CASE, dpa2_case, DPA2_CURATED_CASES) which
forces callers to destructure and is causing RUF059 lint failures for unused
tuple slots; change dpa2_case to return a keyed mapping (e.g., a dict or
lightweight dataclass) instead of a tuple, keep DPA2_BASELINE_CASE as the
canonical dict, update DPA2_CURATED_CASES to call dpa2_case(...) and yield
dicts, and update all consumers to access fields by name (e.g.,
case["precision"] or case.precision) rather than positional unpacking; after
making these changes run ruff check . and ruff format . to clear the lint
errors.
🪄 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: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: 7ef6b16e-3f7f-4e28-ad07-73c3be1c61ba
📒 Files selected for processing (4)
source/tests/consistent/common.pysource/tests/consistent/descriptor/test_dpa1.pysource/tests/consistent/descriptor/test_dpa2.pysource/tests/consistent/descriptor/test_se_e2_a.py
🚧 Files skipped from review as they are similar to previous changes (1)
- source/tests/consistent/descriptor/test_dpa1.py
Replace the remaining SeA stats parameterized decorator with explicit curated cases and keep the descriptor test files in sync with isort/ruff formatting. Authored by OpenClaw (model: gpt-5.4)
Handle collisions in parameterized_cases deterministically by appending a numeric suffix when the sanitized generated test-class name is already in use. This prevents silent overwrites when different case tuples sanitize to the same identifier. Authored by OpenClaw (model: gpt-5.4)
Contributor
Author
|
Addressed the CodeRabbit finding about What changed:
This avoids silent overwrites while preserving all generated test classes. Validated with:
Authored by OpenClaw (model: gpt-5.4) |
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
source/tests/consistent/descriptor/test_dpa2.py (1)
269-297:⚠️ Potential issue | 🟠 MajorRemove the unused case unpacking before lint runs.
skip_pt,skip_dp,skip_tf, andsetUpnever read the unpacked values, andskip_pd/rtol/atolonly useprecision. Ruff is already flagging Line 275, and the same pattern repeats at Lines 308, 341, 374, 451, 557, and 596, so this touched file will keep failing lint until those destructures are removed or narrowed. As per coding guidelines, "Always runruff check .andruff format .before committing changes or CI will fail".Also applies to: 302-330, 335-363, 368-396, 445-473, 551-579, 590-618
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/descriptor/test_dpa2.py` around lines 269 - 297, The tuple unpack in the test methods unpacks many values from self.param that are never used (e.g., repinit_tebd_input_mode, repinit_set_davg_zero, …, use_tebd_bias); update each destructuring to only extract the variables actually used in the test (for example extract just precision or precision and any used flags), or replace unused names with underscores (_) to avoid unused-variable lint errors; look for the same pattern in the other test blocks that unpack self.param (the blocks that currently list the long tuple) and narrow each unpack to the minimal set of names the test reads (or use _ placeholders) so Ruff stops flagging unused variables.
🧹 Nitpick comments (1)
source/tests/consistent/descriptor/test_dpa2.py (1)
66-125:repformer_set_davg_zerois dead in the curated schema.The new case tuple includes
repformer_set_davg_zero, but Line 246 and Line 715 still hardcodeRepformerArgs(set_davg_zero=True). A futuredpa2_case(repformer_set_davg_zero=False)would create a distinct test class without changing the exercised configuration.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@source/tests/consistent/descriptor/test_dpa2.py` around lines 66 - 125, The test includes repformer_set_davg_zero in DPA2_CASE_FIELDS but the test instantiates RepformerArgs with set_davg_zero=True (hardcoded) at the RepformerArgs(...) sites, so toggling the case field doesn't affect the created config; update the two instantiations where RepformerArgs(set_davg_zero=True) is used (the locations flagged around the current RepformerArgs constructors) to read the value from the case dict (use case["repformer_set_davg_zero"]) instead of hardcoding True so the generated test config follows DPA2_BASELINE_CASE / DPA2_CASE_FIELDS.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@source/tests/consistent/descriptor/test_se_e2_a.py`:
- Around line 382-390: SE_A_STAT_CASES currently omits any se_a_case with
type_one_side=False so the TestSeAStat path exercising compute_input_stats() for
the alternate stats branch is untested; add a single
se_a_case(type_one_side=False) entry into the SE_A_STAT_CASES tuple (alongside
the existing baseline and env_protection variants) so the
parameterized_cases(...) includes that variant and exercises the special setup
branch in TestSeAStat.
---
Duplicate comments:
In `@source/tests/consistent/descriptor/test_dpa2.py`:
- Around line 269-297: The tuple unpack in the test methods unpacks many values
from self.param that are never used (e.g., repinit_tebd_input_mode,
repinit_set_davg_zero, …, use_tebd_bias); update each destructuring to only
extract the variables actually used in the test (for example extract just
precision or precision and any used flags), or replace unused names with
underscores (_) to avoid unused-variable lint errors; look for the same pattern
in the other test blocks that unpack self.param (the blocks that currently list
the long tuple) and narrow each unpack to the minimal set of names the test
reads (or use _ placeholders) so Ruff stops flagging unused variables.
---
Nitpick comments:
In `@source/tests/consistent/descriptor/test_dpa2.py`:
- Around line 66-125: The test includes repformer_set_davg_zero in
DPA2_CASE_FIELDS but the test instantiates RepformerArgs with set_davg_zero=True
(hardcoded) at the RepformerArgs(...) sites, so toggling the case field doesn't
affect the created config; update the two instantiations where
RepformerArgs(set_davg_zero=True) is used (the locations flagged around the
current RepformerArgs constructors) to read the value from the case dict (use
case["repformer_set_davg_zero"]) instead of hardcoding True so the generated
test config follows DPA2_BASELINE_CASE / DPA2_CASE_FIELDS.
🪄 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: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: de4628eb-857e-451f-a2e1-629eb1d357f0
📒 Files selected for processing (3)
source/tests/consistent/descriptor/test_dpa1.pysource/tests/consistent/descriptor/test_dpa2.pysource/tests/consistent/descriptor/test_se_e2_a.py
🚧 Files skipped from review as they are similar to previous changes (1)
- source/tests/consistent/descriptor/test_dpa1.py
Add the missing curated SeA stats variant with type_one_side=False so the compute_input_stats path also exercises the alternate branch called out in review. Authored by OpenClaw (model: gpt-5.4)
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@source/tests/consistent/common.py`:
- Around line 761-777: parameterized_cases currently accepts zero args and
returns _parameterized_with_cases(list(cases)) which yields an empty subclass
replacement and silently drops tests; update parameterized_cases to validate the
input and raise a clear error when no case tuples are provided (e.g., if not
cases: raise ValueError("parameterized_cases requires at least one case
tuple")), referencing parameterized_cases and the helper
_parameterized_with_cases so callers get a descriptive failure instead of
silently replacing the base class with object.
🪄 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: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: e6dd42ea-bcb3-4911-a4ed-821cfebe31a4
📒 Files selected for processing (1)
source/tests/consistent/common.py
Contributor
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Set generated parameterized test-class metadata to the exported module name, thread repformer_set_davg_zero through DPA2 curated cases, and update the SeA stats coverage comment to match the curated variants. Authored by OpenClaw (model: gpt-5.4)
Set generated parameterized class metadata for clearer unittest output, reject empty parameterized_cases inputs explicitly, and keep the curated descriptor cases aligned with review feedback. Authored by OpenClaw (model: gpt-5.4)
wanghan-iapcm
approved these changes
Apr 8, 2026
iProzd
approved these changes
Apr 8, 2026
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.
Problem
test_se_e2_a.py, 64 -> 9 fortest_dpa1.py, and 512 -> 12 fortest_dpa2.py.Change
parameterized_cases(*cases)insource/tests/consistent/common.pyfor explicit curated matrices without changing existingparameterized(*attrs)semantics.source/tests/consistent/descriptor/test_se_e2_a.pysource/tests/consistent/descriptor/test_dpa1.pysource/tests/consistent/descriptor/test_dpa2.pyValidation
python -m pytest --collect-only source/tests/consistent/descriptor/test_se_e2_a.py source/tests/consistent/descriptor/test_dpa1.py source/tests/consistent/descriptor/test_dpa2.pypython -m pytest source/tests/consistent/descriptor/test_se_e2_a.py source/tests/consistent/descriptor/test_dpa1.py source/tests/consistent/descriptor/test_dpa2.py -q91 passed, 397 skippedpython3 -m py_compileon the touched filesRelated
Authored by OpenClaw (model: gpt-5.4)
Summary by CodeRabbit
New Features
Tests