Repository navigation
Commit fc4399d
docs: record the expanded GitPython dependent audit
The five automated downstream profiles do not describe compatibility across
GitPython's wider runtime, development and documentation consumers. Record the
completed audit of 31 projects, selected by inspecting 4,060 PyPI distributions
against September 2026 downloads and adding known Git-focused integrations.
Add a report and JSON manifest with tested releases, source commit/archive
provenance, commands, selected upstream versus supplemental counts and coverage
limits. Distinguish existing GitPython mocks, direct downstream Git subprocesses,
unused declarations and indirect documentation dependencies. Link the audit from
`test/downstream/README.md` and update MLflow/DataHub's tested releases without
claiming the expanded local selections are additional automated CI profiles.
Downloaded sources, isolated environments, logs and audit-specific helper scripts
remain retained locally under `.cache/downstream-audit/`; the report explains
what is needed to reproduce those commands from a fresh checkout.
Document the three preceding compatibility fixes: immutable `IndexFile.entries`
read snapshots for dlt/Renku, global `-C` command selection for Mistral Vibe
worktrees, and discovery context for separate Git directories. Complete final
Mistral Vibe runs pass 1,358 of 1,359 cases on each backend; retain its remaining
GC-timing assertion failure. A Diffusers helper's independent `UnboundLocalError`
also reproduces with released GitPython 3.2.1 and does not justify a library fix.
Record services, credentials, missing caller tests and intentionally uncovered
paths without counting them as passing or claiming purely native Gix execution.
All 31 selected projects were processed by individual dependent assignments.
The final Jupytext pre-commit selection passes all 30 upstream cases per backend
after providing real kernel and Pandoc tooling. After the production fixes, the
full GitPython suites passed 1,805 tests on CLI and 1,907 on Gix, each with 59
passing subtests; existing skips and xfail are recorded in the report. Manifest
checks verify the exact selected project set, evidence files, counts, source
provenance, commands, table rows and local Markdown links. `git diff --check`
passes. No production code changed in this documentation commit.1 parent 46c4bed commit fc4399d
3 files changed
Lines changed: 2810 additions & 4 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
@@ -1,5 +1,10 @@ | |||
| 1 | 1 | # Downstream compatibility | |
| 2 | 2 | ||
| 3 | + The [2026-10-22 audit](audit-2026-10-22.md) covers 31 projects, including major | ||
| 4 | + runtime, development and documentation users. Its [manifest](audit-2026-10-22.json) | ||
| 5 | + records releases, source hashes, commands, results and coverage limits. The runner | ||
| 6 | + below automates five CI profiles; it is not the complete audit inventory. | ||
| 7 | + | ||
| 3 | 8 | Run released projects' GitPython-related tests against this checkout with | |
| 4 | 9 | [uv](https://docs.astral.sh/uv/) and Git 2.52 or newer: | |
| 5 | 10 | ||
@@ -35,19 +40,22 @@ settings, pytest options/plugins, and Python import paths are cleared before | |||
| 35 | 40 | setup and testing. Select a different Git executable by putting its directory | |
| 36 | 41 | first on `PATH`. | |
| 37 | 42 | ||
| 38 | - ## Selection and coverage | ||
| 43 | + ## Automated CI profiles | ||
| 39 | 44 | ||
| 40 | - The selection uses current runtime users, including optional end-user features, | ||
| 45 | + These five profiles use current runtime users, including optional end-user features, | ||
| 41 | 46 | ranked by September 2026 PyPI distribution downloads. Development, documentation, | |
| 42 | 47 | and test-only dependencies are excluded. Downloads are not unique installations. | |
| 43 | 48 | ||
| 44 | 49 | | Project | Distribution downloads | Last tested release | Selected coverage | | |
| 45 | 50 | | --- | ---: | --- | --- | | |
| 46 | 51 | | LangChain Community | 27,882,881 | 0.4.2 | 2 upstream GitLoader tests: real clones, commits, checkout, tree traversal, ignored paths, and remote validation | | |
| 47 | - | MLflow (`mlflow-skinny`) | 25,850,354 | 3.17.0 | 47 upstream tests: 31 repository/project/model-versioning cases plus 16 Git context and credential-redaction contract cases | | ||
| 52 | + | MLflow (`mlflow-skinny`) | 25,850,354 | 3.17.2 | 53 unique upstream repository/project/model-versioning, Git context and credential-redaction cases in the expanded audit | | ||
| 48 | 53 | | Bandit | 24,935,372 | 1.9.4 | 12 upstream baseline CLI tests: real repository creation, commits, branches, resets, discovery, and dirty state | | |
| 49 | 54 | | SWE-bench | 22,942,741 | 5.0.2 | 4 supplemental integration cases for `AutoContextManager`; no upstream tests cover its GitPython callers | | |
| 50 | - | DataHub (`acryl-datahub`) | 5,019,402 | 1.7.0.14 | 7 upstream tests passed, 1 credential-dependent skip: public clone/checkout, SSH timeout, exception/redaction contracts, and configuration | | ||
| 55 | + | DataHub (`acryl-datahub`) | 5,019,402 | 1.7.0.16 | 8 upstream tests passed, 1 credential-dependent skip: clone/checkout, SSH timeout, exception/redaction contracts, configuration and ODCS sourcing | | ||
| 56 | + | ||
| 57 | + The runner automates the original selections. The linked audit broadens MLflow | ||
| 58 | + with `tests/projects/test_utils.py` and DataHub with its ODCS Git sourcing test. | ||
| 51 | 59 | ||
| 52 | 60 | Source: [top-pypi-packages](https://hugovk.github.io/top-pypi-packages/top-pypi-packages.min.json), | |
| 53 | 61 | snapshot updated **2026-10-01 12:40:51 UTC**. Its | |
0 commit comments