tools/git-repo.git/git_command.py, branch v2.32

Fix flake8 warnings for some files

2023-01-05T18:43:12+00:00

Change-Id: If67f8660cfb0479f0e710b3566285ef401fcf077 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/355969 Tested-by: Sergiy Belozorov Reviewed-by: Mike Frysinger

Extract env building into a testable helper.

2022-11-16T18:26:49+00:00

Previously env dict building was untested and mixed with other mutative actions. Extract the dict building into a dedicated function and author tests to ensure the functionality is working as expected. BUG: b/255376186 BUG: https://crbug.com/gerrit/16247 Change-Id: I0c88e53eb285c5c3fb27f8e6b3a903aedb8e02a8 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/351874 Reviewed-by: LaMont Jones Tested-by: Sam Saccone

Do not set ALT object dirs when said path resolves to the same dir.

2022-11-16T18:26:49+00:00

Due to symlink resolution git was treating this as two different directories even if the paths were the same. This mitigates the git core bug inside of repo (while the git core fix is being worked on). Bug: b/255376186 Bug: https://crbug.com/gerrit/16247 Change-Id: I12458ee04c307be916851dddd36231997bc8839e Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/351836 Tested-by: Sam Saccone Reviewed-by: LaMont Jones

Set tracing to always on and save to .repo/TRACE_FILE.

2022-11-03T21:07:07+00:00

- add `--trace_to_stderr` option so stderr will include trace outputs and any other errors that get sent to stderr - while TRACE_FILE will only include trace outputs piggy-backing on: https://gerrit-review.googlesource.com/c/git-repo/+/349154 Change-Id: I3895a84de4b2784f17fac4325521cd5e72e645e2 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/350114 Reviewed-by: LaMont Jones Tested-by: Joanna Wang

git_command: fix input passing

2022-08-22T19:34:59+00:00

After reworking this function to use subprocess for output capturing in commit c87c1863b1df392042c8859b81475a65315c8a9d ("git_command: switch process capturing over to subprocess"), passing input via stdin write no longer works. We have to pass it via communicate(), and we have to pass it a string instead of bytes (since we always use encoding='utf-8' now). This is fine since the only user of the input= setting today is already passing in a string. Bug: https://crbug.com/gerrit/16151 Change-Id: Ic58db1e568b8f8aa840a6d62c5a157c14aa6d9bc Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/343515 Tested-by: Mike Frysinger Reviewed-by: LaMont Jones

add a few more docs to existing funcs

2022-08-18T17:31:01+00:00

Change-Id: I27317a59aba67c05ca1fd333e8f064c0edccb209 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/343185 Reviewed-by: LaMont Jones Tested-by: Mike Frysinger

project: store objects in project-objects directly

2022-01-19T17:24:51+00:00

In order to stop sharing objects/ directly between shared projects, we have to fetch the remote objects into project-objects/ manually. So instead of running git operations in the individual project dirs and relying on .git/objects being symlinked to project-objects/, tell git to store any objects it fetches in project-objects/. We do this by leveraging the GIT_OBJECT_DIRECTORY override. This has been in git forever, or at least since v1.7.2 which is what we already hard require. This tells git to save new objects to the specified path no matter where it's being run otherwise. We still otherwise run git in the project-specific dir so that it can find the right set of refs that it wants to compare against, including local refs. For that reason, we also have to leverage GIT_ALTERNATE_OBJECT_DIRECTORIES to tell git where to find objects that are not in the upstream remote. This way git doesn't blow up when it can't find objects only associated with local commits. As it stands right now, the practical result is the same: since we symlink the project objects/ dir to the project-objects/ tree, the default objects dir, the one we set $GIT_OBJECT_DIRECTORY to, and the one we set $GIT_ALTERNATE_OBJECT_DIRECTORIES to are actually all the same. So this commit by itself should be safe. But in a follow up commit, we can replace the symlink with a separate dir and git will keep working. Bug: https://crbug.com/gerrit/15553 Change-Id: Ie4e654aec3e1ee307eee925a54908a2db6a5869f Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/328100 Reviewed-by: Jack Neus Tested-by: Mike Frysinger

Add the ability to administratively enroll repo into using superproject.

2021-07-01T16:27:43+00:00

Repo will remember a choice and an expiration time of the choice, per user, about whether to use superproject by default. When not specified from command line and the choice is not expired, repo would use the user default value. When a user default value is not present and when the system wide enable default is provided in git's system configuration, repo would ask the user for a confirmation which will be valid for two weeks. git_config.py: Add support for system config. When reading system config, we would use --system to avoid hardcoding a path as the value may be different on some other distributions. git_superproject.py: Add a new subroutine, _UseSuperproject(), which returns whether superproject should be used and whether it is from a user configuration. The value is determined in the following order: 1. If the user specifies either --use-superproject or --no-use-superproject, then that choice is being used. 2. If neither is specified, we would then check the saved value (upon repo init) and use that choice when there was a choice. 3. We then check if there is a saved and unexpired value for user's choice in their ~/.gitconfig, and use the unexpired choice, if available. 4. Finally, if all the above didn't give us a decision, and if the git system configuration is providing a rollout hint, present a prompt to user for their decision and save it in ~/.gitconfig. subcmds/sync.py: Make use of the new UseSuperproject() provided by git_superproject.py. While there also silent stderr from git describe when determining the version of repo. Bug: [google internal] b/190688390 Change-Id: Iad3ee03026342ee500e5d65e2f0fa600d7637613 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/309762 Reviewed-by: Mike Frysinger Tested-by: Xin Li

ssh: rewrite proxy management for multiprocessing usage

2021-05-10T21:16:06+00:00

We changed sync to use multiprocessing for parallel work. This broke the ssh proxy code as it's all based on threads. Rewrite the logic to be multiprocessing safe. Now instead of the module acting as a stateful object, callers have to instantiate a new ProxyManager class that holds all the state, an pass that down to any users. Bug: https://crbug.com/gerrit/12389 Change-Id: I4b1af116f7306b91e825d3c56fb4274c9b033562 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/305486 Tested-by: Mike Frysinger Reviewed-by: Chris Mcdonald

ssh: move all ssh logic to a common place

2021-05-06T19:09:16+00:00

We had ssh logic sprinkled between two git modules, and neither was quite the right home for it. This largely moves the logic as-is to its new home. We'll leave major refactoring to followup commits. Bug: https://crbug.com/gerrit/12389 Change-Id: I300a8f7dba74f2bd132232a5eb1e856a8490e0e9 Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/305483 Reviewed-by: Chris Mcdonald Tested-by: Mike Frysinger