Lazily convert Prism/RBS locations to CodeRange objects#436
Open
sinsoku wants to merge 1 commit intoruby:masterfrom
Open
Lazily convert Prism/RBS locations to CodeRange objects#436sinsoku wants to merge 1 commit intoruby:masterfrom
sinsoku wants to merge 1 commit intoruby:masterfrom
Conversation
Member
|
I'm hoping that switching to UTF-8 in #442 will significantly cut down the time spent in |
AST node initializers were eagerly calling `code_range_from_node` to convert `Prism::Location`/`RBS::Location` into `CodeRange`, triggering expensive UTF-16LE encoding via `String#encode` (9.0% of CPU samples in `tool/dog_bench.rb`). Store the raw location in `@mid_code_range_loc` / `@cname_code_range_loc` and convert lazily on first access via `||=`. In batch mode most values are never accessed; in LSP mode the cost is deferred from file loading to first interactive use (hover, go-to-definition, etc.). Also remove `mid_code_range` from `attrs` hashes to avoid comparing unconverted locations in `Node#diff`. Measured with `tool/dog_bench.rb` (stackprof): - `String#encode`: 89 samples (9.0%) -> 19 samples (1.9%)
sinsoku
force-pushed
the
lazy-convert-code-range
branch
from
08c15e6 to
61d3805
Compare
Collaborator
Author
|
Thanks for the review! I've rebased the branch onto master and force-pushed. The default I ran
Two points:
So I think lazy conversion is still worth it, even after #442. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
AST node initializers were eagerly calling
code_range_from_nodetoconvert
Prism::Location/RBS::LocationintoCodeRange, triggeringexpensive UTF-16LE encoding via
String#encode(9.0% of CPU samplesin
tool/dog_bench.rb).Store the raw location in
@mid_code_range_loc/@cname_code_range_locand convert lazily on first access via
||=. In batch mode most valuesare never accessed; in LSP mode the cost is deferred from file loading
to first interactive use (hover, go-to-definition, etc.).
Also remove
mid_code_rangefromattrshashes to avoid comparingunconverted locations in
Node#diff.Measured with
tool/dog_bench.rb(stackprof):String#encode: 89 samples (9.0%) -> 19 samples (1.9%)