Skip to content

improvements for Cargo #1323

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 11 commits into from
Mar 18, 2024
Merged

improvements for Cargo #1323

merged 11 commits into from
Mar 18, 2024

Conversation

Byron
Copy link
Member

@Byron Byron commented Mar 16, 2024

Improvements to support rust-lang/cargo#13592 .

Tasks

… anymore.

It does this by ignoring exclude patterns, preferring false-positives that way.
@Byron Byron force-pushed the improvements-for-cargo branch from dbf3be5 to 0386871 Compare March 17, 2024 06:38
Byron added 6 commits March 17, 2024 10:28
…tions in `.gitignore`.

Initially, symlinks appear like symlinks thanks to `lstat`, but to do
exclude handling correctly these need another `stat` call.

However, this is also not done in Git, but done in `libgit2` only,
so a toggle was added to act like Git by default, but allow obtaining
the same behaviour as git2 for compatibility.
The mode can't be an enum as it would degenerate information, possibly,
which is something we really have to avoid.

If an enum is desired one day, have a `.kind()` method that creates
an enum from the bitflag.
Now symlinked repositories can be traversed as well.
… flag.

That way, it can be much more responsive to interruption.
@Byron Byron force-pushed the improvements-for-cargo branch from 8229c15 to bc30683 Compare March 17, 2024 22:36
Byron added 2 commits March 17, 2024 23:44
That way, more copying happens but the usability increases tremendously as well.
It's breaking as public types moved from `repository::dirwalk` to `dirwalk`,
dissolving `repository::dirwalk` entirely.
@Byron Byron force-pushed the improvements-for-cargo branch from bc30683 to eff82eb Compare March 17, 2024 22:44
Byron added 2 commits March 18, 2024 08:32
That way this won't fail, but *just* silently degenerates information.

The idea is that during FSCK, objects should be decoded and then
re-encoded to see if they are still the same. If not, this means
some leniency kicked in.

Maybe for FSCK, there would also have to be a refactor so there
is a lenient and strict parsing mode. One problem at a time.
@Byron Byron merged commit 41cd53e into main Mar 18, 2024
18 checks passed
@Byron Byron deleted the improvements-for-cargo branch March 18, 2024 08:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

another parsing failure of malformed author/committer timestamp
1 participant