[webkit-reviews] review granted: [Bug 239500] Replace String::replaceWithLiteral() with a String::replace() overload that takes in an ASCIILiteral : [Attachment 457897] Patch

[bugzilla-daemon at webkit.org]

Darin Adler <darin at apple.com> has granted Chris Dumez <cdumez at apple.com>'s
request for review:
Bug 239500: Replace String::replaceWithLiteral() with a String::replace()
overload that takes in an ASCIILiteral
https://bugs.webkit.org/show_bug.cgi?id=239500

Attachment 457897: Patch

https://bugs.webkit.org/attachment.cgi?id=457897&action=review




--- Comment #2 from Darin Adler <darin at apple.com> ---
Comment on attachment 457897
  --> https://bugs.webkit.org/attachment.cgi?id=457897
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=457897&action=review

> Source/WTF/wtf/text/WTFString.h:198
> +    String& replace(UChar target, ASCIILiteral);

I would sort this up with the UChar/StringView overload above it.

> Source/WebCore/editing/markup.cpp:423
> +    return result.toString().replace('\0', ""_s);

I wish it was clearer when to use ""_s and when to use emptyString(). I suppose
that ideally the two would always generate identical code, the more optimal of
the two.

Another possibility would be to add a function named something like removeAll.
Makes me realize that replace should maybe be named replaceAll.

> Tools/TestWebKitAPI/Tests/WTF/WTFString.cpp:247
> +    testString.replace('2', "3"_s);

If there were a lot of callers like this we’d probably want a replace(UChar,
UChar).