Store line numbers of imports in the cache metadata#4533
Merged
Conversation
This allows incremental modes to generate error messages that are both better and more consistent with non-incremental modes. Storing the data in a parallel array is kind of ugly and was done for consistency with dep_prios. It might be worth doing a refactor, though any other way will make the json bigger, which is presumbably why dep_prios was done with an array.
msullivan
force-pushed
the
incremental_dep_linenos
branch
from
706e19e to
dbd9282
Compare
carljm
added a commit
to carljm/mypy
that referenced
this pull request
Feb 14, 2018
* master: (32 commits) Fix some fine-grained cache/fswatcher problems (python#4560) Sync typeshed (python#4559) Add _cached suffix to test cases in fine-grained tests with cache (python#4558) Add back support for simplified fine-grained logging (python#4557) Type checking of class decorators (python#4544) Sync typeshed (python#4556) When loading from a fine-grained cache, use the real path, not the cached (python#4555) Switch all of the fine-grained debug logging to use manager.log (python#4550) Caching for fine-grained incremental mode (python#4483) Fix --warn-return-any for NotImplemented (python#4545) Remove myunit (python#4369) Store line numbers of imports in the cache metadata (python#4533) README.md: Fix a typo (python#4529) Enable generation and caching of fine-grained dependencies from normal runs (python#4526) Move argument parsing for the fine-grained flag into the main arg parsing code (python#4524) Don't warn about unrecognized options starting with 'x_' (python#4522) stubgen: don't append star arg when args list already has varargs appended (python#4518) Handle TypedDict in diff and deps (python#4510) Fix Options.__repr__ to not infinite recurse (python#4514) Fix some fine-grained incremental bugs with newly imported files (python#4502) ...
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
This allows incremental modes to generate error messages that are
both better and more consistent with non-incremental modes.
Storing the data in a parallel array is kind of ugly and was done for
consistency with dep_prios. It might be worth doing a refactor, though
any other way will make the json bigger, which is presumbably why
dep_prios was done with an array.
This is a big yak-shave: this import line number thing was the root cause of differing error message behavior when using fine-grained cache info.