git: Fix submodule fetch, rewrite logic #104
Merged
Conversation
This file contains 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
I don't tell you how to live your life CI |
ReillyBrogan
force-pushed
the
remove-submodules-command
branch
2 times, most recently
from
80093d4
to
d5ec53d
Compare
GZGavinZhao
reviewed
Jun 14, 2024
GZGavinZhao
previously approved these changes
Jun 14, 2024
The logic for this was quite broken and had several failure cases due to that. This commit re-writes that to be saner. Changes: - Individual exec commands are all split out to separate functions so that they can be more easily composed. No function does more than a single command - Control flow now happens entirely in the Fetch() function - Before there were several failure states around submodules due to the `git switch` command happening immediately after the `git fetch` if the repo was already cloned. This is fixed by moving to a unified control flow. - `Fetch()` will now run either the initial clone OR update refs before moving onto further steps - `git clone` no longer attempts to initialize submodules. Several features from `git submodules` aren't yet implemented by `git clone` and so submodule initialization is moved entirely into a `git submodule` command - `git switch` now happens after the clone/fetch but before `git submodule` - `git switch` now no longer runs with any submodule-related flags. Like `git clone` the features we need are just not implemented yet and as far as I can tell there are subtle behavioral differences that were causing issues. - `git submodule` will now use `--filter=blob:none` to specify that submodule checkouts should themselves be blobless when possible. As far as I can tell this works correctly with `--recursive` and recursive submodules. - Because we are now checking out submodules with blobless clones when possible the `fixPermissions()` logic has been updated to account for promisor files that could be in submodule `.git` directories
ReillyBrogan
force-pushed
the
remove-submodules-command
branch
from
d5ec53d
to
a6c2b80
Compare
GZGavinZhao
approved these changes
Jun 14, 2024
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.
The logic for this was quite broken and had several failure cases due to that. This commit re-writes that to be saner.
Changes:
git switchcommand happening immediately after thegit fetchif the repo was already cloned. This is fixed by moving to a unified control flow.Fetch()will now run either the initial clone OR update refs before moving onto further stepsgit cloneno longer attempts to initialize submodules. Several features fromgit submodulesaren't yet implemented bygit cloneand so submodule initialization is moved entirely into agit submodulecommandgit switchnow happens after the clone/fetch but beforegit submodulegit switchnow no longer runs with any submodule-related flags. Likegit clonethe features we need are just not implemented yet and as far as I can tell there are subtle behavioral differences that were causing issues.git submodulewill now use--filter=blob:noneto specify that submodule checkouts should themselves be blobless when possible. As far as I can tell this works correctly with--recursiveand recursive submodules.fixPermissions()logic has been updated to account for promisor files that could be in submodule.gitdirectories