perf: greatly improve hash efficiency in computing attributes #158
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
Enter-tainer
approved these changes
Nov 25, 2024
|
It would be better if we can add benchmark related code too. |
|
Here are the benchmark results, run from my laptop: With Compared with std We can see |
|
More benches with touying. The two files are not so powerful as tablex 😃 |
QuadnucYard
force-pushed
the
perf-hashmap
branch
from
dbed0cc
to
c3225d5
Compare
|
Some bench results before this change: |
|
Running benchmark: cargo bench # run all benchmarks
cargo bench -- pretty # run benchmarks of pretty-print only |
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Outdated suggestions cannot be applied.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.
This PR resolves a critical issue that the formatter runs surprisingly slow on some large documents (e.g., the touying package), which was noticed earlier by @Myriad-Dreamin but not publicly disclosed.
After conducting an investigation with timing, I found that it is the HashMap in node attribution computation instead of Doc creating and pretty printing that hurts. In discussions with @Enter-tainer and @Myriad-Dreamin, we discovered that the previously used
SyntaxNodehash value is computed recursively, which should be avoided in our preprocessing. As an alternative, we can use the span as the unique identifier for nodes. The root node needs to be created fromSource. Otherwise, it would have no valid span attached.With the changes in this PR, the time of formatting
tablex.typ(the largest one in our test assets) dramatically drops from ~500ms to ~20ms. Additionally, unit tests can be completed much faster, although the improvement is not that significant.Benchmarks will be added later.