[ty] Attach subdiagnostics to unresolved-import errors for relative imports as well as absolute imports
#21554
Conversation
This file contains hidden or 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
AlexWaygood
requested review from
carljm,
dcreager and
sharkdp
as code owners
AlexWaygood
added
ty
Diagnostic diff on typing conformance testsNo changes detected when running ty on typing conformance tests ✅ |
Member
Lies! 😆 |
MichaReiser
approved these changes
Nov 21, 2025
|
… imports as well as absolute imports
AlexWaygood
force-pushed
the
alex/import-diagnostic-hints
branch
from
cbde2f4 to
a4df857
Compare
dcreager
added a commit
that referenced
this pull request
Nov 24, 2025
…d-typevar * origin/main: (24 commits) [ty] Remove brittle constraint set reveal tests (#21568) [`ruff`] Catch more dummy variable uses (`RUF052`) (#19799) [ty] Use the same snapshot handling as other tests (#21564) [ty] suppress autocomplete suggestions during variable binding (#21549) Set severity for non-rule diagnostics (#21559) [ty] Add `with_type` convenience to display code (#21563) [ty] Implement docstring rendering to markdown (#21550) [ty] Reduce indentation of `TypeInferenceBuilder::infer_attribute_load` (#21560) Bump 0.14.6 (#21558) [ty] Improve debug messages when imports fail (#21555) [ty] Add support for relative import completions [ty] Refactor detection of import statements for completions [ty] Use dedicated collector for completions [ty] Attach subdiagnostics to `unresolved-import` errors for relative imports as well as absolute imports (#21554) [ty] support PEP 613 type aliases (#21394) [ty] More low-hanging fruit for inlay hint goto-definition (#21548) [ty] implement `TypedDict` structural assignment (#21467) [ty] Add more random TypeDetails and tests (#21546) [ty] Add goto for `Unknown` when it appears in an inlay hint (#21545) [ty] Add type definitions for `Type::SpecialForm`s (#21544) ...
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.
Summary
If we're not able to resolve an import, we attach subdiagnostics to the error saying which search paths we looked for the import in, and linking to our settings page. However, we currently only attach these subdiagnostics if
level == 0-- i.e., if it was an absolute import that was unresolved. This doesn't make much sense; if anything, relative imports are more sensitive to the details of exactly which search paths the user has or hasn't got configured. It looks to me like this is just an accident; this PR changes things so that we attach the subdiagnostic to allunresolved-importdiagnostics, not just absolute ones.The diff here is easiest to review with GitHub's "hide whitespace changes" feature enabled 😄
Test Plan
Snapshot updated