Followup from #10040 - Support uv compiled requirements files #10688
Conversation
Co-authored-by: Edgar Ramírez Mondragón <16805946+edgarrmondragon@users.noreply.github.com>
Co-authored-by: skshetry <18718008+skshetry@users.noreply.github.com>
…ger uv pip compile test coverage
Co-authored-by: Edgar Ramírez Mondragón <16805946+edgarrmondragon@users.noreply.github.com>
…ater.rb Co-authored-by: beagold <86345081+beagold@users.noreply.github.com>
…ater.rb Co-authored-by: beagold <86345081+beagold@users.noreply.github.com>
| def uv_pip_compile_options_from_compiled_file(requirements_file) | ||
| options = ["--output-file=#{requirements_file.name}"] | ||
|
|
||
| options << "--no-emit-index-url" if requirements_file.content.include?("index-url http") |
There was a problem hiding this comment.
This would cause dependabot to strip an existing index-url.
see #10040 (comment)
| options << "--no-emit-index-url" if requirements_file.content.include?("index-url http") | |
| options << "--emit-index-url" if requirements_file.content.include?("index-url http") |
probably needs a test generated with --emit-index-url to make sure dependabot doesn't remove it
| @@ -277,6 +287,32 @@ def run_pip_compile_command(command, fingerprint:) | |||
| run_command(command, fingerprint: fingerprint) | |||
| end | |||
|
|
|||
| def uv_pip_compile_options_from_compiled_file(requirements_file) | |||
There was a problem hiding this comment.
this implementation needs to be updated to match the one in pip_compile_file_updater.rb
|
OK, quick update- this isn't going to work as is. There are other bugs in it also, and now taking a deeper look at the way pdm/uv/pip-compile/etc are supported (or not) here, that's showing up a bit more to do. will come back to it shortly. @jonjanego - can you reach out to me please? thanks! |
can you email me at jonjanego@github.com - i do not see your email on your profile |
Skimming the template to make this a bit easier to understand for maintainers.
This is a replay of PR #10040.
to help out @avilaton, i have....
I took note of the refactoring note given some of the duplication between the file and update methods. It's not a bad idea in general to abstract some of that out, however the level of effort there might not be worth it just now -- so I left it as is.
anything else needed to get this landed, @jonjanego / @sachin-sandhu ?
thanks all!