[Webkit-unassigned] [Bug 137967] Use constants from wtf/MathExtras.h

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=137967

Milan Crha <mcrha at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #240279|0                           |1
        is obsolete|                            |
 Attachment #240279|review?, commit-queue?      |
              Flags|                            |
 Attachment #240348|                            |review?, commit-queue?
              Flags|                            |

--- Comment #10 from Milan Crha <mcrha at redhat.com> ---
Created attachment 240348
  --> https://bugs.webkit.org/attachment.cgi?id=240348&action=review
proposed patch ]I[

Updated patch with the style fix.

(In reply to comment #8)
> Please, use the same bug title in all changelogs. See
> Tools/Scripts/prepare-ChangeLog, it's recommended to use that tools to
> generate the changelog entries, it's not only easier for you, but also
> ensures consistency.

I did use the tool, but not with all its features. I regenerated the ChangeLogs with it with used --bug now. I agree it's much simpler than hand-editing all touched ChangeLogs.

(In reply to comment #6)
> If you use the webkit-patch tool to upload the patch it should point out
> style errors before the patch hits the bug tracker. You can also use
> check-webkit-style.

I do not use the webkit-patch, I use only the prepare-ChangeLogs. Looking into its --help it states it also checks the style
>   --[no-]style    Run check-webkit-style script when done (default: style)
but I did not get any style error when the prepare-ChangeLog finished.

(In reply to comment #9)
> I agree with Alexey; I don't see much value to redefining constants with
> different names that, in general, are defined in a standard way. If it were
> up to me I would get rid of all of these duplicate macros in favor of the
> standard ones.

I do not know why mingw/msys doesn't see those constants from math.h or cmath, but it might be similar like bug #137966, some compilers were fine with isnan (probably 'using namespace std;' being "inherited" from another source file) while others require std::isnan(), like the mingw/msys for me. Please do not drop those constants.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20141023/b5dbfd2f/attachment-0002.html>