tooling: Use upstream checkers#18087
Conversation
a15b9a1 to
599d874
Compare
Signed-off-by: Ryan Northey <ryan@synca.io>
599d874 to
1d8ce2c
Compare
Signed-off-by: Ryan Northey <ryan@synca.io>
1d8ce2c to
7aeadc4
Compare
|
|
||
| def main(*args) -> int: | ||
| return PipChecker(*args).run() | ||
| return EnvoyPipChecker(*args).run() |
There was a problem hiding this comment.
Are these wrappers just for repository path magic? Is there a way to factor this out? I notice the path logic appears multiple times. Otherwise LGTM.
There was a problem hiding this comment.
i think we want this
if someone has some issue with eg the PipChecker they are going to look at this file and see that the only thing coming locally is the path, and that the rest of the logic is coming from the upstream package
with others it will not just be the path - but eg extension_categories or repository_locations etc etc
the object model here is designed to make these data sources explicit while shifting the actual implementation code upstream
|
/wait for docstring/comment updates |
Signed-off-by: Ryan Northey <ryan@synca.io>
Signed-off-by: Ryan Northey <ryan@synca.io>
|
|
||
| def _strip_lines(self, lines: Iterable[str]) -> List[str]: | ||
| return [self._strip_line(line) for line in lines if line] | ||
| def path(self) -> pathlib.Path: |
There was a problem hiding this comment.
As discussed offline, would be nice to get rid of this boilerplate. But this can be a distinct PR.
|
/lgtm deps |
Commit Message: tooling: Use upstream pip check
Additional Description:
Risk Level:
Testing:
Docs Changes:
Release Notes:
Platform Specific Features:
[Optional Runtime guard:]
[Optional Fixes #Issue]
[Optional Deprecated:]
[Optional API Considerations:]