Remove `py3-breezy32` tox environment, as `breezy >= 3.3.1` is now in
`requirements.txt`.

Jelmer Vernooij authored 1 year ago and Nicolas Dandrimont committed 1 year ago

 * Primarily interact with the branch object, including locking
   locking that (locking the repository implicitly)

 * Open the specified location directly rather than opening the
   containing URL and then checking there are no siblings

Jelmer Vernooij authored 1 year ago and Nicolas Dandrimont committed 1 year ago

Locking the branch will lock the underlying repository anyway.

This causes Bazaar to cache, which makes the memoization in the
head_revision_id property unnecessary.

3.3.0 is very old and won't build on newer Debian versions; it excludes
various bug fixes and typing.

Previously the loader was raising an error when encountering a nested
tree (inventory entry of kind 'tree-reference').

Those nested trees are typically created from git submodules when importing
a bazaar repository from a git one.

By default bazaar does not recurse on a nested tree content when exporting
a revision and simply creates an empty directory instead.

So adopt the same behavior in the loader as it exists cases where a nested
tree cannot be resolved.

Related to #4583.

Instead of having example repositories used as tests data stored in
tarballs, prefer to build them in tests implementation instead using
the breezy API.

It helps reading the tests and it gives more flexibility to update
them if required.

Add check_revision int parameter to the BzrLoader class.

Similarly to the subversion loader, every check_revision computed revisions
the filesystem associated to a bazaar revision is exported to a temporary
folder, the hash of its root directory computed and then compared to the SWH
directory id associated to the SWH revision. If they differ, the loading is
stopped by raising an exception.

Use this new parameter in tests to check sample repositories are correctly
loaded into the archive.

Ensure that when an empty directory is removed in a bazaar revision
it is also removed from the SWH directory data model.

Previously, an empty directory added in a bazaar revision was incorrectly
mapped to a content object with directory permission in the SWH data model.

So ensure it is properly mapped to a directory object in the SWH data model.

In contrary to git, bazaar preserves empty directories in revision trees
so those should not be automatically removed by the BzrDirectory class
when a path is removed.

So fix that issue to ensure computed SWHIDs for directories will match
those computed from a bazaar export.

Previously the loader was replacing symlinks by empty files so fix their
archival by setting target paths as file contents.

Related to #4584.

Refs. swh/infra/sysadm-environment#4906

It fixes a regression related to empty committer name introduced in 182754c3.

It also ensures that null committer is also handled, in that case committer
date must also be set to None.

Fixes #4582

This reverts commit 5062e4a7.

First, this never should have been merged in this repository as this is deployment
related. Second, this creates side-effects on the main workers [1]. So simply fix it by
reverting. That will create less headable to everyone.

[1] preventing restarts of services due to the missing distribution.

Refs. swh/infra/sysadm-environment#4869
Refs. swh/infra/sysadm-environment#4868

Refs. swh/infra/sysadm-environment#4524

GitLab will display the content of the README file when browsing the
repository. But in case the file is a symlink, it will display the
path pointed by the symlink. There is a 6 year old issue about this:
https://gitlab.com/gitlab-org/gitlab/-/issues/15093

We can workaround the issue by having the content at the root of the
repository and a symlink to this file in the `docs/` directory.

Tested in swh/devel/swh-py-template!27

Related to swh/meta#4960

Related to swh/meta#4959

This fixes python 3.7 support due to poetry, a dependency of isort, that
removed support for that Python version in a recent release.

In order to remove warnings about /apidoc/*.rst files being included
multiple times in toc when building full swh documentation, prefer to
include module indices only when building standalone package documentation.

Related to T4496

- pre-commit from 4.1.0 to 4.3.0,
- codespell from 2.2.1 to 2.2.2,
- black from 22.3.0 to 22.10.0 and
- flake8 from 4.0.1 to 5.0.4. Also freeze flake8 dependencies.

Also change flake8's repo config to github (the gitlab mirror
being outdated).

Use helper fixture loading_task_creation_for_listed_origin_test from
swh-loader-core and remove redundant test.

1. version 3.3.0 introduces Rust code, which requires rustc versions newer
   than available in Debian 11
2. For some reason, pip does not even try to install setuptools-rust on
   the CI, even though it's a build dependency of Breezy

All removed items have either been already addressed or were found to
not matter.

This is adapted from the email about public bzr loader review.

Recent changes in swh-scheduler add new parameters to the celery tasks
produced from swh.scheduler.model.ListedOrigin instances.

So ensure to handle any new parameters by not hardcoding the expected
ones in task signatures.

Remove unsafe use of unnamed task parameters.

Add new tests checking task parameters produced from ListedOrigin
instances do no raise error when attempting to create a bzr loader.

Related to T4187

That hook can be frustrating as it can discard a long commit message
if it finds a typo in it so better removing it.

Related to T3922

black is considered stable since release 22.1.0 and the version
we are currently using is quite outdated and not compatible with
click 8.1.0, so it is time to bump it to its latest stable release.

Please note that E501 pycodestyle warning related to line length
is replaced by B950 one from flake8-bugbear as recommended by black.
https://black.readthedocs.io/en/stable/the_black_code_style/current_style.html#line-length

Related to T3922

pytest-postgresql 3.1.3 and pytest-redis 2.4.0 added support for
pytest >= 7 so we can now drop the pytest pinning.

Fields offset and negative_utc of TimestampWithTimezone model were
deprecated and got removed.