Repository navigation
chore: remove python-pydantic and python-pydantic-core from the factory - #244
Merged
Merged
Conversation
Neither is in Utah's image closure: nothing in projectbluefin/utah packages/utah.toml or packages/bluefin.toml, config/bluefin-packages.toml, or any factory recipe requires python3-pydantic, and in Fedora the only consumer of python3-pydantic-core is python3-pydantic itself. They came in with the first closure wave (48bb840) and have never built here: - python-pydantic-core is a Rust extension whose BuildRequires are crate(ahash/default) and friends. Fedora 44 and Hummingbird ship no Rust crate RPMs (the same reason Utah documents for fish), so builddep cannot be satisfied (run 35966980787). - python-pydantic then fails builddep for want of pydantic-core (runs 35966980787, 36032253217). Because publication is atomic, two packages Utah does not use were holding back every package it does. Removed per docs/contributing.md "Removing a package": config/upstream-sources.json, .packit.yaml, packages/<name>/, and the count assertions (373 -> 371). The contributing guide now names the four test files that carry a count. python-typing-inspection and python-annotated-types were imported for pydantic too, but they build cleanly and are left alone. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QMTLsnqW2jtz5QsNC3W3qv
hanthor
added a commit
that referenced
this pull request
Sep 25, 2026
…ates (#248) * Revert "chore: remove python-pydantic and python-pydantic-core from the factory (#244)" This reverts commit 41f2e88. * fix: restore python-pydantic and build pydantic-core from vendored crates Reverts #244. input-remapper is in the consumer contract (config/bluefin-packages.toml) and requires python3dist(pydantic) at runtime, so publish in run 36073571248 failed on nothing provides python3.14dist(pydantic) needed by input-remapper #244 had checked only BuildRequires and Utah's manifests. pydantic-core never built because Fedora's recipe takes every Rust crate as a system RPM through %cargo_generate_buildrequires, and Fedora 44 and Hummingbird ship none of them. Build it vendored instead: - tools/generated_sources.py gains a python-pydantic-core generator: the sha512-pinned PyPI sdist plus vendor/<crate>-<version> for every crate its Cargo.lock pins, each download checked against the Cargo.lock SHA-256, with .cargo-checksum.json as cargo vendor writes it. Stdlib only, no cargo needed; two runs give identical bytes (sha512 b20f43b5...). - The lock entry becomes a generate entry; `sources` is repinned. - The spec uses %cargo_prep -v vendor, drops %cargo_generate_buildrequires, writes %cargo_vendor_manifest, and drops Fedora's two Cargo.toml patches (jiter range, PyO3 0.29) that only exist to match Fedora's crate RPMs; the lockfile's PyO3 0.28.3 supports Python 3.14. Verified in fedora:44 + Hummingbird (cargo 1.98.1, Python 3.14.7): `cargo check --offline --locked --release` against the vendored tree succeeds, and tools/source_pipeline.py accepts the generated source. docs/contributing.md now says to check the runtime closure of the consumer contract before removing a package as unneeded. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QMTLsnqW2jtz5QsNC3W3qv --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
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.
Why
Neither package is in Utah's image closure. Nothing requires
python3-pydanticin projectbluefin/utahpackages/utah.toml/packages/bluefin.toml, inconfig/bluefin-packages.toml, or in any factory recipe. In Fedora, the only consumer ofpython3-pydantic-coreispython3-pydantic. Both came in with the first closure wave (48bb840), and neither has ever built here:crate(ahash/default)and other crates. Fedora 44 and Hummingbird ship no Rust crate RPMs (Utah documents the same limit for fish), so builddep cannot be satisfied (job 107527896049, run 35966980787).Publication is atomic, so these two packages that Utah does not use were holding back every package it does use (wpa_supplicant, fprintd, cups-filters, ...).
What
Removed per docs/contributing.md "Removing a package":
config/upstream-sources.jsonentries.packit.yamlblockspackages/python-pydantic/,packages/python-pydantic-core/test_source_inventory.py, which excludes Hummingbird-supplied mesa), stage count 72 -> 70, and the counts indocs/architecture.md.The "Removing a package" section now names the four test files that hold a count, because
test_source_inventory.pywas easy to miss.python-typing-inspectionandpython-annotated-typeswere also imported for pydantic. They build cleanly, so they are left alone.Coordination: #239 bumps the same counts 373 -> 374. Whichever PR lands second gets rebased.
🤖 Generated with Claude Code
https://claude.ai/code/session_01QMTLsnqW2jtz5QsNC3W3qv