Fix nondeterministic test by using dict.fromkeys ~= orderedset
#814
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
|
This pull request was exported from Phabricator. Differential Revision: D52662624 |
stroxler
force-pushed
the
export-D52662624
branch
from
0a16a95 to
e3829a5
Compare
stroxler
added a commit
to stroxler/pyre-check
that referenced
this pull request
Jan 10, 2024
…ebook#814) Summary: For a number of weeks now I've been frustrated by github test failure notifications. I dug into the root cause for BE day, and it is that we are using `list(set(_))` to deduplicate two lists of references in the find-references logic. This is fine and correct, but also nondeterministic because `set` ordering isn't fixed. There's no ordered set in the python standard library, and it's not worth adding a dependency to get one. Instead, the standard trick to work around this is to use a dict of `None` so the keys act like a set, which solves the problem because in 3.7+ dict keys are ordered. Differential Revision: D52662624
|
This pull request was exported from Phabricator. Differential Revision: D52662624 |
…ebook#814) Summary: **Problem** For a number of weeks now I've been frustrated by github test failure notifications; many of my commits are generating tasks to fix test failures. This test is actually disabled in internal CI because it isn't deterministic; that means at least it isn't noisy internally, but we should still fix it! **Root Cause** I dug into the root cause for BE day, and it is that we are using `list(set(_))` to deduplicate two lists of references in the find-references logic. Deduping is totally correct, but also nondeterministic because `set` ordering isn't fixed. **Solution** There's no ordered set in the python standard library, and it's not worth adding a dependency to get one. Instead, the standard trick to work around this is to use a dict of `None` so the keys act like a set, which solves the problem because in 3.7+ dict keys are ordered. Differential Revision: D52662624
stroxler
force-pushed
the
export-D52662624
branch
from
e3829a5 to
03a9d4c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D52662624 |
|
This pull request has been merged in f5a1d8e. |
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:
For a number of weeks now I've been frustrated by github test failure
notifications. I dug into the root cause for BE day, and it is that we
are using
list(set(_))to deduplicate two lists of references inthe find-references logic.
This is fine and correct, but also nondeterministic because
setorderingisn't fixed. There's no ordered set in the python standard library, and
it's not worth adding a dependency to get one. Instead, the standard trick
to work around this is to use a dict of
Noneso the keys act like a set,which solves the problem because in 3.7+ dict keys are ordered.
Differential Revision: D52662624