Remove GIT_INDEX_FILE from env before running hooks#1815
Remove GIT_INDEX_FILE from env before running hooks#1815souliane wants to merge 3 commits intoj178:masterfrom
GIT_INDEX_FILE from env before running hooks#1815Conversation
When `git commit -a` invokes the pre-commit hook, GIT_INDEX_FILE points to .git/index.lock. If a hook subprocess (e.g., a test suite) runs git commands, those commands read/write the lock file instead of the normal index, corrupting it. Prek's subsequent `git diff` then fails with "fatal: unable to read <hash>". Strip GIT_INDEX_FILE from the process environment after prek's own git operations (write-tree, diff-index, collect-files) but before hook execution. This way hooks and their subprocesses see the normal .git/index. Fixes j178#1786
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #1815 +/- ##
=======================================
Coverage 91.72% 91.72%
=======================================
Files 98 98
Lines 20055 20055
=======================================
Hits 18395 18395
Misses 1660 1660 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
📦 Cargo Bloat ComparisonBinary size change: +0.00% (24.8 MiB → 24.8 MiB) Expand for cargo-bloat outputHead Branch ResultsBase Branch Results |
⚡️ Hyperfine BenchmarksSummary: 0 regressions, 0 improvements above the 10% threshold. Environment
CLI CommandsBenchmarking basic commands in the main repo:
|
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base --version |
2.5 ± 0.2 | 2.3 | 3.6 | 1.01 ± 0.08 |
prek-head --version |
2.4 ± 0.1 | 2.3 | 2.6 | 1.00 |
prek list
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base list |
9.5 ± 0.2 | 9.1 | 10.3 | 1.00 ± 0.03 |
prek-head list |
9.4 ± 0.2 | 9.1 | 9.8 | 1.00 |
prek validate-config .pre-commit-config.yaml
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base validate-config .pre-commit-config.yaml |
3.2 ± 0.1 | 3.1 | 3.4 | 1.00 |
prek-head validate-config .pre-commit-config.yaml |
3.3 ± 0.1 | 3.2 | 3.6 | 1.02 ± 0.03 |
prek sample-config
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base sample-config |
2.8 ± 0.1 | 2.6 | 2.9 | 1.00 |
prek-head sample-config |
2.8 ± 0.1 | 2.6 | 2.9 | 1.00 ± 0.03 |
Cold vs Warm Runs
Comparing first run (cold) vs subsequent runs (warm cache):
prek run --all-files (cold - no cache)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run --all-files |
159.6 ± 1.6 | 156.3 | 161.4 | 1.00 |
prek-head run --all-files |
161.2 ± 2.2 | 158.6 | 165.5 | 1.01 ± 0.02 |
prek run --all-files (warm - with cache)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run --all-files |
164.7 ± 3.3 | 160.0 | 171.2 | 1.02 ± 0.03 |
prek-head run --all-files |
161.6 ± 3.5 | 156.8 | 170.3 | 1.00 |
Full Hook Suite
Running the builtin hook suite on the benchmark workspace:
prek run --all-files (full builtin hook suite)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run --all-files |
162.9 ± 3.7 | 156.9 | 173.5 | 1.00 |
prek-head run --all-files |
165.7 ± 25.2 | 156.4 | 338.2 | 1.02 ± 0.16 |
Individual Hook Performance
Benchmarking each hook individually on the test repo:
prek run trailing-whitespace --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run trailing-whitespace --all-files |
23.1 ± 0.5 | 22.1 | 24.2 | 1.03 ± 0.03 |
prek-head run trailing-whitespace --all-files |
22.5 ± 0.3 | 21.8 | 23.0 | 1.00 |
prek run end-of-file-fixer --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run end-of-file-fixer --all-files |
30.5 ± 2.3 | 26.4 | 35.3 | 1.04 ± 0.10 |
prek-head run end-of-file-fixer --all-files |
29.4 ± 1.7 | 27.0 | 33.8 | 1.00 |
prek run check-json --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run check-json --all-files |
13.5 ± 0.4 | 12.3 | 14.2 | 1.05 ± 0.05 |
prek-head run check-json --all-files |
12.8 ± 0.4 | 12.1 | 13.5 | 1.00 |
prek run check-yaml --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run check-yaml --all-files |
12.5 ± 0.3 | 12.0 | 13.3 | 1.03 ± 0.03 |
prek-head run check-yaml --all-files |
12.2 ± 0.3 | 11.7 | 12.9 | 1.00 |
prek run check-toml --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run check-toml --all-files |
12.3 ± 0.3 | 11.9 | 13.0 | 1.00 ± 0.03 |
prek-head run check-toml --all-files |
12.2 ± 0.3 | 11.7 | 12.7 | 1.00 |
prek run check-xml --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run check-xml --all-files |
12.5 ± 0.3 | 12.1 | 13.7 | 1.03 ± 0.03 |
prek-head run check-xml --all-files |
12.2 ± 0.2 | 11.8 | 12.6 | 1.00 |
prek run detect-private-key --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run detect-private-key --all-files |
19.6 ± 1.6 | 17.4 | 23.0 | 1.03 ± 0.12 |
prek-head run detect-private-key --all-files |
19.0 ± 1.5 | 16.8 | 22.0 | 1.00 |
prek run fix-byte-order-marker --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run fix-byte-order-marker --all-files |
25.1 ± 1.5 | 21.6 | 29.2 | 1.00 |
prek-head run fix-byte-order-marker --all-files |
25.5 ± 1.7 | 22.0 | 29.2 | 1.02 ± 0.09 |
Installation Performance
Benchmarking hook installation (fast path hooks skip Python setup):
prek install-hooks (cold - no cache)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base install-hooks |
5.0 ± 0.1 | 4.9 | 5.1 | 1.01 ± 0.03 |
prek-head install-hooks |
4.9 ± 0.1 | 4.8 | 5.1 | 1.00 |
prek install-hooks (warm - with cache)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base install-hooks |
5.1 ± 0.1 | 5.0 | 5.2 | 1.01 ± 0.02 |
prek-head install-hooks |
5.0 ± 0.1 | 5.0 | 5.1 | 1.00 |
File Filtering/Scoping Performance
Testing different file selection modes:
prek run (staged files only)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run |
54.5 ± 1.3 | 52.1 | 57.0 | 1.00 ± 0.04 |
prek-head run |
54.4 ± 1.4 | 52.8 | 58.1 | 1.00 |
prek run --files '*.json' (specific file type)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run --files '*.json' |
9.5 ± 0.2 | 9.3 | 9.7 | 1.00 ± 0.03 |
prek-head run --files '*.json' |
9.5 ± 0.2 | 9.2 | 10.1 | 1.00 |
Workspace Discovery & Initialization
Benchmarking hook discovery and initialization overhead:
prek run --dry-run --all-files (measures init overhead)
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run --dry-run --all-files |
14.5 ± 0.7 | 14.0 | 17.4 | 1.01 ± 0.05 |
prek-head run --dry-run --all-files |
14.4 ± 0.2 | 14.0 | 14.9 | 1.00 |
Meta Hooks Performance
Benchmarking meta hooks separately:
prek run check-hooks-apply --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run check-hooks-apply --all-files |
13.0 ± 0.2 | 12.8 | 13.5 | 1.00 |
prek-head run check-hooks-apply --all-files |
13.1 ± 0.2 | 12.8 | 13.4 | 1.00 ± 0.02 |
prek run check-useless-excludes --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run check-useless-excludes --all-files |
12.9 ± 0.1 | 12.7 | 13.1 | 1.00 |
prek-head run check-useless-excludes --all-files |
12.9 ± 0.1 | 12.7 | 13.0 | 1.00 ± 0.01 |
prek run identity --all-files
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
prek-base run identity --all-files |
11.4 ± 0.1 | 11.1 | 11.6 | 1.00 |
prek-head run identity --all-files |
11.4 ± 0.1 | 11.2 | 11.6 | 1.00 ± 0.01 |
GIT_INDEX_FILE from env before running hooks
37bf856 to
089f9d2
Compare
Summary
Fixes #1786
When
git commit -ainvokes the pre-commit hook,GIT_INDEX_FILEpoints to.git/index.lock. If a hook subprocess (e.g., a test suite like pytest) runs git commands internally, those commands read/write the lock file instead of the normal.git/index, corrupting the lock. Prek's subsequentgit diffthen fails withfatal: unable to read <hash>.This PR strips
GIT_INDEX_FILEfrom the process environment after prek's own git operations (write-tree,diff-index,collect-files) but before hook execution.Reproduction
The repro repo has a single hook that runs
git init+git addin a temp directory (simulating what happens when a test suite runs tests that invoke git commands). The inheritedGIT_INDEX_FILEcauses the temp dir'sgit addto write to the parent repo's.git/index.lock.Verification
After this fix,
git commit -a -m "test"on the repro repo succeeds.Note: I don't have a Rust toolchain installed locally, so I couldn't run the test suite. CI should validate the build. If you'd like me to add a test case, let me know the preferred pattern.