Use more compact cache representation for int and str#19750
Merged
ilevkivskyi merged 5 commits intopython:masterfrom Aug 29, 2025
Merged
Use more compact cache representation for int and str#19750ilevkivskyi merged 5 commits intopython:masterfrom
ilevkivskyi merged 5 commits intopython:masterfrom
Conversation
This comment has been minimized.
This comment has been minimized.
JukkaL
reviewed
Aug 28, 2025
Collaborator
JukkaL
left a comment
There was a problem hiding this comment.
Nice -- did you check how much this reduces the size of binary cache files?
| write_int(b, 2 ** 85) | ||
| write_int(b, 255) | ||
| write_int(b, -1) | ||
| write_int(b, -255) |
Collaborator
There was a problem hiding this comment.
Test also the edge cases (-11, -10, -9, 116, 117, 118). Test a few more different lengths of integers (e.g. 15 bits, 23 bits, 30 bits) with arbitrary lower bits.
Member
Author
There was a problem hiding this comment.
OK, I think it would also make sense to test something like len(data.getvalue()) == 1 etc.
Member
Author
It looks like it's ~40% smaller now, for example:
|
Contributor
|
According to mypy_primer, this change doesn't affect type check results on a corpus of open source code. ✅ |
JukkaL
approved these changes
Aug 29, 2025
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
After looking more at some real data I found that:
ints are between -10 and 117. Values are a bit arbitrary TBH, the idea is that we should include small negative values (forTypeVarIds) and still be able to fit them in 1 byte.Note there are very few integers that would fit in two bytes currently. This is because we only store line for type alias nodes, and type aliases are usually defined at the top of a module. We can add special case for two bytes later when needed.
We could probably save another byte for long strings and medium integers, but I don't want to have anything fancy that would only affect less than 0.1% cases.
Finally you may notice I add a small correctness change I noticed accidentally when working on this, it is not really related, but it is so minor that it doesn't deserve a separate PR.