Reduce code size of testing tokens if they're a number #5181
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
epage
reviewed
Oct 23, 2023
19k is a decent amount of savings so it seems worth it. |
alexcrichton
force-pushed
the
smaller-is-number
branch
from
9268629
to
f7176d5
Compare
epage
reviewed
Oct 24, 2023
epage
reviewed
Oct 24, 2023
alexcrichton
force-pushed
the
smaller-is-number
branch
from
f7176d5
to
a17ce61
Compare
This commit is a tiny win in compiled code size of a final binary including `clap` which shaves off 19k of compiled code locally. Previously tokens were checked if they were a number by using `.parse::<f64>().is_ok()`, but parsing floats is relatively heavyweight in terms of code size. This replaces the check with a more naive "does this string have lots of ascii digits" check where the compiled size of this check should be much smaller.
alexcrichton
force-pushed
the
smaller-is-number
branch
from
a17ce61
to
9a9aabc
Compare
epage
approved these changes
Oct 24, 2023
|
Thanks! |
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Outdated suggestions cannot be applied.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 commit is a tiny win in compiled code size of a final binary including
clapwhich shaves off 19k of compiled code locally. Previously tokens were checked if they were a number by using.parse::<f64>().is_ok(), but parsing floats is relatively heavyweight in terms of code size. This replaces the check with a more naive "does this string have lots of ascii digits" check where the compiled size of this check should be much smaller.I'll note that this is not a critical optimization at all and is one I'm mostly curious if it's of interest. There's other various edge cases that were previously classified as numbers which will no longer be classified as numbers such as
nan,inf,+1,1e10, etc. That would mean that this is technically a breaking change, and if that's not acceptable then this is most definitely not worth it.