[internal] Refactor how call sites interact with Python resolves #14287
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Rust tests and lints will be skipped. Delete if not intended. [ci skip-rust] # Building wheels and fs_util will be skipped. Delete if not intended. [ci skip-build-wheels]
# Rust tests and lints will be skipped. Delete if not intended. [ci skip-rust] # Building wheels and fs_util will be skipped. Delete if not intended. [ci skip-build-wheels]
stuhood
approved these changes
Jan 28, 2022
Comment on lines
+449
to
+450
| # TODO: compute the resolve based on request.addresses, and validate that the | ||
| # transitive closure is compatible. See coursier_fetch.py for inspiration. |
There was a problem hiding this comment.
Should this point to a particular followup issue?
There was a problem hiding this comment.
Normally, yeah, but I plan to land this change tomorrow :) I wanted to simplify things first, including an upcoming PR to kill generate-user-lockfiles
Eric-Arellano
added a commit
that referenced
this pull request
Jan 28, 2022
…le` in favor of `[python].experimental_resolves` (#14288) Turns out that Pants only needs a single resolve. Our testprojects resolve wasn't actually used by anything. (We could always add this back for testing purposes.) So, thanks to #14287, we can use `[python].experimental_resolves` just like `[python].experimental_lockfile` was being used: the feature currently supports a single resolve. Benefits to this change: 1. Lockfile is now generated via `generate-lockfiles` rather than `generate-user-lockfiles` 2. Less confusing code for us to deal with. Now we just have constraints files vs. resolves [ci skip-rust]
Merged
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The first change bundles
ResolveField.validate()and.value_or_default(), which makes call sites much less clunky.Note that this contrasts with JVM where we use
JvmSubsystem.resolves_for_target()- I stuck withField-based methods because it gives us better MyPy typing betweenResolveFieldvsCompatibleResolveFields.--
The second change prepares for how we will calculate the resolve to use. Before, we had call sites like
pytest_runner.pysetting the resolve. Instead,pex_from_targets.pyshould choose the resolve based on the direct input addresses to the request. Then it validates that the transitive closure is correct. This implies:test,package, andrunare chosen by simply inspecting the singlepython_testorpex_binarytarget, but validating all depsPexFromTargetsare all compatiblerepl, we can't do any partitioning, only validationFor now, we simply use the default resolve. A followup will implement this algorithm, similar to our JVM code in
coursier_fetch.py.Note that this means
--enable-resolvesnow gives you roughly the same behavior as[python].experimental_lockfile! That is, for now, we only support a single resolve, whatever is set by[python].default_resolve. This means that we can remove[python].experimental_lockfileandgenerate-user-lockfileto simplify this code.