Fix codspeed - #620
Conversation
|
Warning Rate limit exceeded
⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (5)
WalkthroughAdds CI test-data priming: a new priming script, refactors the cache-test-data composite action to compute OS-scoped DATA_CACHE_KEY/DATA_CACHE_PATH with mode/outputs, and integrates a prime_test_data_cache job into multiple workflows so caches are restored/primed before tests and benchmarks. (50 words) Changes
Possibly related PRs
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 inconclusive)
✅ Passed checks (2 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 |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b5e05fa97d
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| prime_test_data_cache: | ||
| needs: setup |
There was a problem hiding this comment.
The new prime_test_data_cache job runs unconditionally on pull requests, but test_code is still guarded by if: ... !contains(..., 'no_ci'). On PRs labeled no_ci, this now runs a full 3-OS cache priming matrix anyway, and the workflow can still fail there even though tests were intentionally skipped. This regresses the existing “turn off CI” behavior and can block PRs or consume significant CI time unexpectedly.
Useful? React with 👍 / 👎.
| prime_test_data_cache: | ||
| needs: setup |
There was a problem hiding this comment.
Like the main test workflow, the added prime_test_data_cache job here is unconditional while test_code_min_deps is still skipped for no_ci PRs. That means PRs marked no_ci still execute the expensive cache-priming matrix (and can fail in it), which defeats the intended skip mechanism and introduces avoidable CI failures/cost.
Useful? React with 👍 / 👎.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
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)
.github/actions/cache-test-data/action.yml (1)
4-7:⚠️ Potential issue | 🟠 MajorDon't let
restore-keysbypass the manual cache reset.Line 5 says
cache-numberis the reset knob, but Lines 62-63 still fall back to anydata-${{ runner.os }}-${{ env.DATA_REGISTRY_HASH }}-*archive. Bumping the number will therefore copy the old cache forward instead of forcing a clean rebuild when you're trying to recover from stale or corrupted contents.Suggested fix
with: enableCrossOsArchive: true path: ${{ env.DATA_CACHE_PATH }} key: ${{ env.DATA_CACHE_KEY }} - restore-keys: | - data-${{ runner.os }}-${{ env.DATA_REGISTRY_HASH }}-Also applies to: 62-63
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/actions/cache-test-data/action.yml around lines 4 - 7, The restore behavior currently allows the wildcard fallback key data-${{ runner.os }}-${{ env.DATA_REGISTRY_HASH }}-* to be used even after bumping cache-number, so bumping cache-number doesn't force a clean rebuild; update the cache action configuration so the cache key includes the cache-number knob (cache-number) and remove or adjust the wildcard restore-keys fallback so it also includes cache-number (or omit the wildcard entirely) — ensure the primary key and any restore-keys reference the cache-number symbol (and not only the DATA_REGISTRY_HASH) so changing cache-number reliably invalidates prior caches.
🧹 Nitpick comments (1)
.github/workflows/prime_test_data_cache.yml (1)
3-63: Consider exposing this as a reusable workflow.The same checkout/setup/install/restore/download/save sequence now exists in
get_coverage.yml,runtests.yml,run_min_dep_tests.yml,test_doc_build.yml, andprofile.yml, including a mix of hardcoded and env-drivencache-numbervalues. Pulling this behindworkflow_callwould keep cache-key changes and future cache resets in one place.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/prime_test_data_cache.yml around lines 3 - 63, Extract the repeated checkout/setup/install/restore/download/save sequence into a reusable workflow by converting the current job "prime_test_data_cache" into a workflow callable via workflow_call; create inputs for PYTHON_DEFAULT, cache-number (and optional matrix.os or allow the caller to supply a matrix), and a boolean/enum for whether to run the download/save steps, then replace the duplicated blocks in get_coverage.yml, runtests.yml, run_min_dep_tests.yml, test_doc_build.yml, and profile.yml to call the new reusable workflow (uses: ./.github/workflows/prime_test_data_cache.yml) and pass the required inputs; ensure the job still references the same steps (actions/checkout@v4, ./.github/actions/load-shared-vars, actions/setup-python@v6, pip install -e ., ./.github/actions/cache-test-data, and .github/scripts/cache_test_data.py) so cache-key logic and cache-number remain centralized and consistent.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/workflows/profile.yml:
- Around line 65-70: The action metadata in
.github/actions/mamba-install-dascore/action.yml contains an unsupported field
for the input named install-package; open that action.yml, locate the inputs
section and remove the invalid "type: boolean" entry for install-package (and
any other "type" keys on inputs), leaving only supported keys like description,
required, default, or deprecationMessage so the workflow step using
./.github/actions/mamba-install-dascore validates correctly.
---
Outside diff comments:
In @.github/actions/cache-test-data/action.yml:
- Around line 4-7: The restore behavior currently allows the wildcard fallback
key data-${{ runner.os }}-${{ env.DATA_REGISTRY_HASH }}-* to be used even after
bumping cache-number, so bumping cache-number doesn't force a clean rebuild;
update the cache action configuration so the cache key includes the cache-number
knob (cache-number) and remove or adjust the wildcard restore-keys fallback so
it also includes cache-number (or omit the wildcard entirely) — ensure the
primary key and any restore-keys reference the cache-number symbol (and not only
the DATA_REGISTRY_HASH) so changing cache-number reliably invalidates prior
caches.
---
Nitpick comments:
In @.github/workflows/prime_test_data_cache.yml:
- Around line 3-63: Extract the repeated
checkout/setup/install/restore/download/save sequence into a reusable workflow
by converting the current job "prime_test_data_cache" into a workflow callable
via workflow_call; create inputs for PYTHON_DEFAULT, cache-number (and optional
matrix.os or allow the caller to supply a matrix), and a boolean/enum for
whether to run the download/save steps, then replace the duplicated blocks in
get_coverage.yml, runtests.yml, run_min_dep_tests.yml, test_doc_build.yml, and
profile.yml to call the new reusable workflow (uses:
./.github/workflows/prime_test_data_cache.yml) and pass the required inputs;
ensure the job still references the same steps (actions/checkout@v4,
./.github/actions/load-shared-vars, actions/setup-python@v6, pip install -e .,
./.github/actions/cache-test-data, and .github/scripts/cache_test_data.py) so
cache-key logic and cache-number remain centralized and consistent.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 2f406846-b9b0-4a53-8659-5a14c8051e61
📒 Files selected for processing (8)
.github/actions/cache-test-data/action.yml.github/scripts/cache_test_data.py.github/workflows/get_coverage.yml.github/workflows/prime_test_data_cache.yml.github/workflows/profile.yml.github/workflows/run_min_dep_tests.yml.github/workflows/runtests.yml.github/workflows/test_doc_build.yml
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #620 +/- ##
=======================================
Coverage 99.48% 99.48%
=======================================
Files 133 133
Lines 11494 11494
=======================================
Hits 11435 11435
Misses 59 59
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
.github/actions/cache-test-data/action.yml (1)
69-87: Consider the implications ofenableCrossOsArchivewith OS-specific paths.The
enableCrossOsArchive: truesetting allows cache restoration across different OS types. However, sinceDATA_CACHE_PATHvaries by OS (e.g.,~/.cache/dascoreon Linux vs~/Library/Caches/dascoreon macOS), a cache created on one OS would be restored to a different path on another OS. This is likely fine if each OS job primes its own cache, but verify this is the intended behavior.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/actions/cache-test-data/action.yml around lines 69 - 87, The workflow enables cross-OS caching via enableCrossOsArchive: true while DATA_CACHE_PATH is OS-specific, which can cause caches created on one OS to be restored to a different path; update the restore-test-data and cache-test-data steps to either disable enableCrossOsArchive (set it to false) or make the cache paths and keys OS-aware (e.g., include runner.os in DATA_CACHE_PATH or in DATA_CACHE_KEY/restore-keys) so saved caches map to the correct platform; ensure the change is applied for both uses of enableCrossOsArchive and references to DATA_CACHE_PATH/DATA_CACHE_KEY.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/workflows/profile.yml:
- Around line 53-71: The benchmarks job depends on prime_test_data_cache but
never restores the cached test data; add a step named like "Restore test data
cache" at the start of the benchmarks job (before running setup/install steps
such as ./.github/actions/mamba-install-dascore) that uses actions/cache@v3 to
restore the same cache key and paths used by the prime_test_data_cache job
(match the key/paths/restore-keys used there so the cache is actually restored
into the test data directory or env var like TEST_DATA_CACHE_PATH).
---
Nitpick comments:
In @.github/actions/cache-test-data/action.yml:
- Around line 69-87: The workflow enables cross-OS caching via
enableCrossOsArchive: true while DATA_CACHE_PATH is OS-specific, which can cause
caches created on one OS to be restored to a different path; update the
restore-test-data and cache-test-data steps to either disable
enableCrossOsArchive (set it to false) or make the cache paths and keys OS-aware
(e.g., include runner.os in DATA_CACHE_PATH or in DATA_CACHE_KEY/restore-keys)
so saved caches map to the correct platform; ensure the change is applied for
both uses of enableCrossOsArchive and references to
DATA_CACHE_PATH/DATA_CACHE_KEY.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 830feadf-b827-4f99-9a3f-005c26cfe90d
📒 Files selected for processing (2)
.github/actions/cache-test-data/action.yml.github/workflows/profile.yml
| benchmarks: | ||
| name: Run benchmarks | ||
| needs: prime_test_data_cache | ||
| runs-on: ubuntu-latest | ||
| steps: | ||
| - uses: actions/checkout@v4 | ||
| with: | ||
| fetch-tags: "true" | ||
| fetch-depth: '0' | ||
|
|
||
| - uses: ./.github/actions/load-shared-vars | ||
|
|
||
| - uses: ./.github/actions/mamba-install-dascore | ||
| with: | ||
| # Defined in .github/actions/load-shared-vars/action.yml | ||
| python-version: ${{ env.PYTHON_DEFAULT }} | ||
| install-group-str: "[profile]" | ||
| cache-number: 1 | ||
|
|
There was a problem hiding this comment.
Missing test data cache restore in benchmarks job.
The benchmarks job depends on prime_test_data_cache but doesn't actually restore the cached test data. GitHub Actions caches must be explicitly restored in each job that needs them—the needs directive only ensures job ordering, not cache sharing.
Add a step to restore the test data cache before running benchmarks:
🔧 Proposed fix to add cache restore
- uses: ./.github/actions/load-shared-vars
+ - uses: ./.github/actions/cache-test-data
+ with:
+ mode: restore
+ cache-number: 1
+
- uses: ./.github/actions/mamba-install-dascore
with:🧰 Tools
🪛 actionlint (1.7.11)
[error] 65-65: could not parse action metadata in "/home/jailuser/git/.github/actions/mamba-install-dascore": line 4: unexpected key "type" for definition of input "install-package"
(action)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.github/workflows/profile.yml around lines 53 - 71, The benchmarks job
depends on prime_test_data_cache but never restores the cached test data; add a
step named like "Restore test data cache" at the start of the benchmarks job
(before running setup/install steps such as
./.github/actions/mamba-install-dascore) that uses actions/cache@v3 to restore
the same cache key and paths used by the prime_test_data_cache job (match the
key/paths/restore-keys used there so the cache is actually restored into the
test data directory or env var like TEST_DATA_CACHE_PATH).
Merging this PR will improve performance by 24.59%
|
| Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|
| ⚡ | test_wiggle |
303.3 ms | 258.4 ms | +17.39% |
| ⚡ | test_transpose |
4.2 ms | 3.4 ms | +23.06% |
| ⚡ | test_contiguous_merge |
220.8 ms | 195 ms | +13.22% |
| ⚡ | test_scan_df |
478.7 ms | 407.1 ms | +17.58% |
| ⚡ | test_taper |
12.3 ms | 11.1 ms | +10.86% |
| ⚡ | test_last |
6.3 ms | 5.2 ms | +21.5% |
| ⚡ | test_max |
14.9 ms | 12.6 ms | +17.52% |
| ⚡ | test_differentiate |
33.8 ms | 27.1 ms | +24.49% |
| ⚡ | test_no_overlap_merge |
265.7 ms | 217.4 ms | +22.2% |
| ⚡ | test_scan |
347.8 ms | 299.7 ms | +16.06% |
| ⚡ | test_select_half_range |
49.7 ms | 40.3 ms | +23.07% |
| ⚡ | test_diverse_merge |
697.1 ms | 607.6 ms | +14.73% |
| ⚡ | test_select_full_range |
69.9 ms | 56.6 ms | +23.6% |
| ⚡ | test_min |
14.9 ms | 12.6 ms | +17.53% |
| ⚡ | test_1second_chunk |
1,219 ms | 979.1 ms | +24.51% |
| ⚡ | test_definite_integrate |
34 ms | 27.9 ms | +21.78% |
| ⚡ | test_select_string_match |
25.5 ms | 21 ms | +21.29% |
| ⚡ | test_select |
26.6 ms | 21.4 ms | +24.13% |
| ⚡ | test_half_second_chunk |
2.4 s | 1.9 s | +24.59% |
| ⚡ | test_align_1d_shift_valid |
22.9 ms | 20.2 ms | +13.5% |
| ... | ... | ... | ... | ... |
ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.
Comparing fix_codspeed (8c4ac64) with master (5de00c6)
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
.github/workflows/runtests.yml (1)
85-116:⚠️ Potential issue | 🟠 MajorMissing test data cache restore in test_code job.
The
test_codejob depends onprime_test_data_cachebut doesn't restore the cached test data. GitHub Actions caches must be explicitly restored in each job that needs them—theneedsdirective only ensures job ordering, not cache sharing.Add a step to restore the test data cache after checkout:
🔧 Proposed fix to add cache restore
- uses: actions/checkout@v4 with: fetch-tags: 'true' fetch-depth: '0' + - uses: ./.github/actions/cache-test-data + with: + mode: restore + cache-number: ${{ env.CACHE_NUMBER }} + - uses: ./.github/actions/mamba-install-dascore with: python-version: ${{ matrix.python-version }}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/runtests.yml around lines 85 - 116, The test_code job depends on prime_test_data_cache but never restores the cached test data; add a cache-restore step immediately after the actions/checkout step in the test_code job to restore the same cache key used by prime_test_data_cache so test data is available when running ./.github/test_code.sh; reference the job name test_code, the dependency prime_test_data_cache, and the existing checkout step to locate where to insert the restore step..github/workflows/run_min_dep_tests.yml (1)
81-109:⚠️ Potential issue | 🟠 MajorMissing test data cache restore in test_code_min_deps job.
Same issue as
runtests.yml—the job depends onprime_test_data_cachebut doesn't restore the cached test data. Add a cache restore step after checkout.🔧 Proposed fix to add cache restore
- uses: actions/checkout@v4 with: fetch-tags: "true" fetch-depth: '0' + - uses: ./.github/actions/cache-test-data + with: + mode: restore + cache-number: 1 + - uses: ./.github/actions/mamba-install-dascore with: python-version: ${{ matrix.python-version }}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/run_min_dep_tests.yml around lines 81 - 109, The test_code_min_deps job is missing a restore of the primed test data cache; add a cache-restore step immediately after the actions/checkout@v4 step in the test_code_min_deps job to restore the artifacts created by the prime_test_data_cache job (use the same cache key/restore logic used in runtests.yml or the prime_test_data_cache output), so the job actually gets the prepared test data before running the mamba install and "Install dascore (min deps)" steps.
♻️ Duplicate comments (2)
.github/workflows/profile.yml (2)
62-91:⚠️ Potential issue | 🟠 MajorMissing test data cache restore in benchmarks job.
The
benchmarksjob depends onprime_test_data_cachebut doesn't restore the cached test data. Add a cache restore step after checkout and before running benchmarks.🔧 Proposed fix to add cache restore
- uses: ./.github/actions/load-shared-vars + - uses: ./.github/actions/cache-test-data + with: + mode: restore + cache-number: 1 + - uses: ./.github/actions/mamba-install-dascore with:🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/profile.yml around lines 62 - 91, The benchmarks job declares a dependency on prime_test_data_cache but never restores that cache; add a cache-restore step immediately after the actions/checkout step (and before running the benchmarks) that uses the same cache key/paths produced by the prime_test_data_cache job (use actions/cache@v3 or your existing cache action), so the test data populated by prime_test_data_cache is available to the Run benchmarks step (refer to the benchmarks job, the prime_test_data_cache dependency, and the Run benchmarks step/micromamba run command to place the restore).
77-82:⚠️ Potential issue | 🟠 MajorInvalid metadata in mamba-install-dascore action blocks workflow parsing.
The static analysis tool reports that
.github/actions/mamba-install-dascore/action.ymlcontains an unsupportedtype: booleanfield for theinstall-packageinput. GitHub Actions action metadata only supportsdescription,required,default, anddeprecationMessagefor inputs. This needs to be fixed in the action definition file (not included in this PR).🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/profile.yml around lines 77 - 82, The action metadata for the mamba-install-dascore action has an unsupported "type: boolean" on the install-package input; remove the "type" key from the install-package input in the action metadata and instead represent a boolean default as a string default (e.g., default: "true"/"false") or omit it and use required/default only, making sure only the allowed input keys (description, required, default, deprecationMessage) are present; update the install-package input entry in the mamba-install-dascore action metadata accordingly so the workflow parser accepts it.
🧹 Nitpick comments (1)
.github/workflows/run_min_dep_tests.yml (1)
16-19: Consider adding CACHE_NUMBER env for consistency.This workflow uses hardcoded
cache-number: 1(lines 67, 79) whileruntests.ymldefines aCACHE_NUMBERenv variable. Consider adding the same env variable for easier cache invalidation across workflows.♻️ Proposed fix for consistency
env: + # used to manually trigger cache reset. Just increment if needed. + CACHE_NUMBER: 1 # Ensure matplotlib doesn't try to show figures in CI MPLBACKEND: AggThen update lines 67 and 79 to use
${{ env.CACHE_NUMBER }}.Also applies to: 67-67, 79-79
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/run_min_dep_tests.yml around lines 16 - 19, Add a CACHE_NUMBER environment variable to the workflow env block (next to MPLBACKEND and QT_QPA_PLATFORM) and set it to the same default used in runtests.yml (e.g., 1); then replace the hardcoded cache-number: 1 occurrences with cache-number: ${{ env.CACHE_NUMBER }} so the workflow uses the shared CACHE_NUMBER value (search for CACHE_NUMBER, MPLBACKEND, QT_QPA_PLATFORM, and the two cache-number entries to locate the spots to update).
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Outside diff comments:
In @.github/workflows/run_min_dep_tests.yml:
- Around line 81-109: The test_code_min_deps job is missing a restore of the
primed test data cache; add a cache-restore step immediately after the
actions/checkout@v4 step in the test_code_min_deps job to restore the artifacts
created by the prime_test_data_cache job (use the same cache key/restore logic
used in runtests.yml or the prime_test_data_cache output), so the job actually
gets the prepared test data before running the mamba install and "Install
dascore (min deps)" steps.
In @.github/workflows/runtests.yml:
- Around line 85-116: The test_code job depends on prime_test_data_cache but
never restores the cached test data; add a cache-restore step immediately after
the actions/checkout step in the test_code job to restore the same cache key
used by prime_test_data_cache so test data is available when running
./.github/test_code.sh; reference the job name test_code, the dependency
prime_test_data_cache, and the existing checkout step to locate where to insert
the restore step.
---
Duplicate comments:
In @.github/workflows/profile.yml:
- Around line 62-91: The benchmarks job declares a dependency on
prime_test_data_cache but never restores that cache; add a cache-restore step
immediately after the actions/checkout step (and before running the benchmarks)
that uses the same cache key/paths produced by the prime_test_data_cache job
(use actions/cache@v3 or your existing cache action), so the test data populated
by prime_test_data_cache is available to the Run benchmarks step (refer to the
benchmarks job, the prime_test_data_cache dependency, and the Run benchmarks
step/micromamba run command to place the restore).
- Around line 77-82: The action metadata for the mamba-install-dascore action
has an unsupported "type: boolean" on the install-package input; remove the
"type" key from the install-package input in the action metadata and instead
represent a boolean default as a string default (e.g., default: "true"/"false")
or omit it and use required/default only, making sure only the allowed input
keys (description, required, default, deprecationMessage) are present; update
the install-package input entry in the mamba-install-dascore action metadata
accordingly so the workflow parser accepts it.
---
Nitpick comments:
In @.github/workflows/run_min_dep_tests.yml:
- Around line 16-19: Add a CACHE_NUMBER environment variable to the workflow env
block (next to MPLBACKEND and QT_QPA_PLATFORM) and set it to the same default
used in runtests.yml (e.g., 1); then replace the hardcoded cache-number: 1
occurrences with cache-number: ${{ env.CACHE_NUMBER }} so the workflow uses the
shared CACHE_NUMBER value (search for CACHE_NUMBER, MPLBACKEND, QT_QPA_PLATFORM,
and the two cache-number entries to locate the spots to update).
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 6f665423-7a12-4bca-b594-099c9e3ec0d1
📒 Files selected for processing (3)
.github/workflows/profile.yml.github/workflows/run_min_dep_tests.yml.github/workflows/runtests.yml
d039b99 to
abef7ed
Compare
1 participant
Description
This PR tries to fix the codspeed profiling. The problem is the new DASVader files require a newer version of hdf5 which gets installed with the micromamba env but not necessarily pip/uv.
Checklist
I have (if applicable):
Summary by CodeRabbit