WebKit Bugzilla
New
Browse
Search+
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
81070
Avoid StringImpl::getData16SlowCase() when sorting array
https://bugs.webkit.org/show_bug.cgi?id=81070
Summary
Avoid StringImpl::getData16SlowCase() when sorting array
Benjamin Poulain
Reported
2012-03-13 19:19:30 PDT
When sorting a JSArray, one of the bottleneck is the conversion from StringImpl::getData16SlowCase()
Attachments
Patch
(4.59 KB, patch)
2012-03-13 19:27 PDT
,
Benjamin Poulain
ggaren
: review+
ggaren
: commit-queue-
Details
Formatted Diff
Diff
View All
Add attachment
proposed patch, testcase, etc.
Benjamin Poulain
Comment 1
2012-03-13 19:27:27 PDT
Created
attachment 131773
[details]
Patch
Geoffrey Garen
Comment 2
2012-03-14 12:37:05 PDT
Comment on
attachment 131773
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=131773&action=review
> Source/JavaScriptCore/wtf/text/StringImpl.h:782 > + if (string1 && string2) {
It would be better to NULL check string1 and string2 at the head of the function, and return early if NULL, rather than NULL checking more than once in the body of the function, and indenting so much of the code. I think this would work: if (!string1) return -1; if (!string2) return string1->length(); ....
Geoffrey Garen
Comment 3
2012-03-14 12:38:11 PDT
> if (!string2) > return string1->length();
Oops! if (!string2) return string1->length() ? 1 : -1;
Benjamin Poulain
Comment 4
2012-03-14 12:44:24 PDT
> It would be better to NULL check string1 and string2 at the head of the function, and return early if NULL, rather than NULL checking more than once in the body of the function, and indenting so much of the code. I think this would work:
Good point! I'll update that Thanks for the review.
Benjamin Poulain
Comment 5
2012-03-14 22:11:56 PDT
Committed
r110822
: <
http://trac.webkit.org/changeset/110822
>
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug