Remap $PWD to empty string instead of "."

[goffrie]

This fixes an inconsistency in file paths (included in e.g. panic info), wherein some paths would have a leading ./ and some would not.

I believe specifically generic code that is instantiated across a crate boundary would end up with an absolute path, then have it remapped to ., whereas other code would use the literal path passed on the command line to rustc, which is generally a relative path with no ./ prefix.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[krasimirgg]

Could this yield rewritten paths that look like absolute paths, e.g., if $pwd is /path/to/dir and a relative path in a message is to subdir/lib.rs, replacing $pwd with the empty string could yield messages with a path like /subdir/lib.rs, which looks like an absolute path and is confusing IMO.

[goffrie]

In my testing, rustc is smart and does not do that (it strips the leading slash).

…

On Mon., Sep. 26, 2022, 01:32 Krasimir Georgiev, ***@***.***> wrote: Could this yield rewritten paths that look like absolute paths, e.g., if $pwd is /path/to/dir and a relative path in a message is to subdir/lib.rs, replacing $pwd with the empty string could yield messages with a path like /subdir/lib.rs, which looks like an absolute path and is confusing IMO. — Reply to this email directly, view it on GitHub <#1563 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAJO2TQI434TYWFIEORAA3LWAFNRVANCNFSM6AAAAAAQTR7IFI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[krasimirgg]

Looks good then! @goffrie could you address this PR's google-cla check?

[goffrie]

Done.

[UebelAndre]

In my testing, rustc is smart and does not do that (it strips the leading slash).
…

Could this be demonstrated by a test?

[goffrie]

I added a test and manually verified that it fails without this change (the path has a leading ./).

[krasimirgg]

Thank you!