Bug 65994
| Summary: | FindInPage (Ctrl+F) for "≠" also returns "=" | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Dan Hill-Moses <danhillmoses> |
| Component: | Text | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED WORKSFORME | ||
| Severity: | Trivial | CC: | ap, cristian.vat, jshin |
| Priority: | P4 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | All | ||
| OS: | All | ||
| URL: | http://en.wikipedia.org/wiki/List_of_Unicode_characters | ||
Dan Hill-Moses
Chromium Version : 13.0.782.107
Webkit Version : 535.1
URLs (if applicable) : any, but http://en.wikipedia.org/wiki/List_of_Unicode_characters is a good example
Other browsers tested:
Iceweasel 3.5.19: OK
Epiphany 3.0.4: FAIL
What steps will reproduce the problem?
1. Goto any website with an equals sign (the character "=", that's U+003D)
2. Search for the inequality symbol ("≠", that's U+2260)
What is the expected result?
To only return the expected character, "≠".
What happens instead?
Both "=" and "≠" are returned.
Please provide any additional information below. Attach a screenshot if
possible.
Full about:version :
Chromium 13.0.782.107 (Developer Build 94237) Built on Debian unstable, running on Debian wheezy/sid
OS Linux
WebKit 535.1 (trunk@87771)
JavaScript V8 3.3.10.22
Flash 11.0 d1
User Agent Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.107 Safari/535.1
Command Line /usr/lib/chromium/chromium --password-store=detect --flag-switches-begin --flag-switches-end
Executable Path /usr/lib/chromium/chromium
Profile Path /home/dan/.config/chromium/Default
uname -a:
Linux debian 3.0.0-1-amd64 #1 SMP Sun Jul 24 02:24:44 UTC 2011 x86_64 GNU/Linux
Can also replicate in Google Chrome 13.0.782.112 for Mac OS X 10.5.8 on x86.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Jungshik Shin
This is interesting. It's because U+260 is canonically equivalent to U+003D followed by U+0338 (combining long slash overlay).
A ticket is being filed against the CLDR to give a primary weight to U+0338 in the search collator. I'll add a reference to that ticket once it's filed.
This will affect all ports that use ICU/CLDR (Safari, Chrome, Android, etc).
We can special-case U+0338 in WebCore, but ....
Jungshik Shin
BTW, chrome bug is http://crbug.com/92280
Jungshik Shin
The CLDR bug on this issue is http://unicode.org/cldr/trac/ticket/4081
Cristian Vat
The CLDR bug is fixed and closed since 4 years ago.
Already tested the chrome bug at http://crbug.com/92280 and everything seems to be working properly there so maybe this could also be closed.
Alexey Proskuryakov
Testing Safari on OS X Mavericks, this indeed doesn't happen any more.