Closed Titou325 closed 7 months ago
Thanks @Titou325
@varungandhi-src go any feedback?
Hi @varungandhi-src,
Thanks for the feedback. I have updated the snapshot format to match your expectations, with the difference of the start enclosing range being a down-pointing caret. It seems to me that this would be most natural given that the start range information is given above the first line (due to possible decorators presence), while the end range is below the last line of the range.
This PR is ready for review but will require subsequent modifications to merge into main, although they are currently blocked by #91.
This PR is ready for review but will require subsequent modifications to merge into main, although they are currently blocked by https://github.com/sourcegraph/scip-python/issues/91
We don't need to be blocked by that issue. Either I or you can run the snapshots on Ubuntu and push that once the code changes are done.
Check for possible regressions due to updating scip.proto
scip.proto
evolves in a backwards compatible way, so it shouldn't affect indexer output.
Hi!
Updated the PR regarding all points above mentioned. I don't have a Linux runner easily accessible at this time, would be glad if you could update the snapshots @varungandhi-src. N.B. the showDocs
option will probably be required on the other files.
Thanks a lot for the feedback!
@Titou325 I've updated the snapshots here using a Linux runner: https://github.com/codemuse-app/scip-python/pull/1
Hi @varungandhi-src, I have merged and updated the PR, fixing the issue with duplicate class declarations. This is an org owned fork, so I am unfortunately unable to give blanket permissions but I have given you write access.
The snapshots should be updated on Linux runners and then I believe we are good to go. Thanks!
@Titou325 if you'd like, I think you should be able to remove my maintain access on your fork, and use this option for future PRs.
Perfect, thanks! Can we bump versions to trigger a new release?
Thanks @Titou325 awesome work
This is a first attempt at implementing enclosing ranges for function and class definitions, closing #86.
I have attempted to keep it as simple as possible, passing a parent node if needed to
pushNewOccurrence
.This does not implement enclosing ranges for reference occurences.
These are the pending items:
scip.proto
# format-options: showDocs
to other test snapshot inputs and regenerate expected outputs (currently blocked by #91)Any feedback is very welcome. Until we can update the remaining snapshots at the very least, this PR will stay in draft. Thanks!