Bug 42755 - Web Inspector: Almost all inspector tests are failing if comma is used as decimal separator.
: Web Inspector: Almost all inspector tests are failing if comma is used as dec...
Status: RESOLVED FIXED
: WebKit
Web Inspector (Deprecated)
: 528+ (Nightly build)
: All All
: P2 Normal
Assigned To:
:
:
:
:
  Show dependency treegraph
 
Reported: 2010-07-21 08:10 PST by
Modified: 2010-08-11 00:11 PST (History)


Attachments
[patch] initial version. (1.15 KB, patch)
2010-07-21 08:18 PST, Ilya Tikhonovsky
pfeldman: review+
Review Patch | Details | Formatted Diff | Diff


Note

You need to log in before you can comment on or make changes to this bug.


Description From 2010-07-21 08:10:22 PST
%subj%
------- Comment #1 From 2010-07-21 08:11:28 PST -------
Windows with Russian locale.
------- Comment #2 From 2010-07-21 08:18:19 PST -------
Created an attachment (id=62183) [details]
[patch] initial version.
------- Comment #3 From 2010-07-21 08:19:36 PST -------
(From update of attachment 62183 [details])
This is a workaround, and not the real fix. The real fix is to change to a String formatting function that is locale independent.
------- Comment #4 From 2010-07-21 08:23:34 PST -------
(In reply to comment #3)
> (From update of attachment 62183 [details] [details])
> This is a workaround, and not the real fix. The real fix is to change to a String formatting function that is locale independent.

It is not clear whether String formatting should be changed as a whole because of this use case. The code being fixed is supposed to produce strings compatible with JSON and it is JSON spec that requires numbers to have "." separators. Anyways, the fix is incomplete.
------- Comment #5 From 2010-07-21 08:38:03 PST -------
(From update of attachment 62183 [details])
Ok, given that 0.5 will be formatted Ok we should land this.
------- Comment #6 From 2010-07-21 10:04:24 PST -------
As Darin said, String formatting needs to be fixed - see also bug 18994.

This is a rather poor workaround - it fixes Russian locale, but a locale can use almost arbitrary formatting for numbers. In particular, even proper Russian formatting "(123.456.789,00) will remain broken.
------- Comment #7 From 2010-07-21 12:20:57 PST -------
(In reply to comment #6)
> As Darin said, String formatting needs to be fixed - see also bug 18994.
> 
> This is a rather poor workaround - it fixes Russian locale, but a locale can use almost arbitrary formatting for numbers. In particular, even proper Russian formatting "(123.456.789,00) will remain broken.

We should document all occurences of work arounds for the string formating problems. So we can fix them all at once. Otherwise, adding all this br's on bug 18994 is also some kind of documentation.
------- Comment #8 From 2010-07-21 12:28:11 PST -------
(In reply to comment #6)
> As Darin said, String formatting needs to be fixed - see also bug 18994.
> 
> This is a rather poor workaround - it fixes Russian locale, but a locale can use almost arbitrary formatting for numbers. In particular, even proper Russian formatting "(123.456.789,00) will remain broken.

According to spec %f doesn't uses thousands separator. 
It will be the problem if some locale uses something different as decimal separator.
------- Comment #9 From 2010-08-10 22:45:30 PST -------
already landed
------- Comment #10 From 2010-08-10 23:37:24 PST -------
(In reply to comment #8)
> (In reply to comment #6)
> > As Darin said, String formatting needs to be fixed - see also bug 18994.
> > 
> > This is a rather poor workaround - it fixes Russian locale, but a locale can use almost arbitrary formatting for numbers. In particular, even proper Russian formatting "(123.456.789,00) will remain broken.
> 
> According to spec %f doesn't uses thousands separator. 
> It will be the problem if some locale uses something different as decimal separator.

I guess this is what ap tried to mention. :-P We have many localas with this schema (points and commas in a float). This fix is a hack, that works for some locals, but many still fail, or will fail now. It's sad that there was no FIXME added in the patch. Normaly this patch shouldn't be in trunk. :-(
------- Comment #11 From 2010-08-11 00:11:13 PST -------
yep. it should be fixed another way.

https://bugs.webkit.org/show_bug.cgi?id=43832