Bug 109817 - Stop placing std::isfinite and std::signbit inside the global scope
Summary: Stop placing std::isfinite and std::signbit inside the global scope
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Zan Dobersek
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-02-14 05:14 PST by Zan Dobersek
Modified: 2013-02-18 09:15 PST (History)
27 users (show)

See Also:

Attachments
Patch (83.87 KB, patch)
2013-02-14 05:21 PST, Zan Dobersek 		darin: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Zan Dobersek 2013-02-14 05:14:00 PST 
Stop placing std::isfinite and std::signbit inside the global scope
Comment 1 Zan Dobersek 2013-02-14 05:21:47 PST 
Created attachment 188328 [details]
Patch
Comment 2 Eric Seidel (no email) 2013-02-14 09:33:40 PST 
Comment on attachment 188328 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=188328&action=review

> Source/WTF/ChangeLog:12
> +        On Solaris and OpenBSD platforms or when using Visual C++ the two methods
> +        are now defined (as incompatibility workarounds) inside the std namespace.

So this is a compile fix for Solaris and OpenBSD?
Comment 3 Eric Seidel (no email) 2013-02-14 09:34:03 PST 
When did Solaris/OpenBSD break? or have they always been broken?
Comment 4 Darin Adler 2013-02-14 09:59:41 PST 
The basic approach on this in WebKit for years has been to work around these kinds of issues in the MathExtras.h header. Please put something in that header instead of changing every single call site for isfinite and signbit.

Unless you want to change our strategy, which probably requires some webkit-dev conversation.
Comment 5 Alexey Proskuryakov 2013-02-14 10:21:47 PST 
Mu understanding is that we agreed to qualify names with std:: in this e-mail thread: <https://lists.webkit.org/pipermail/webkit-dev/2012-May/020656.html>. So this seems like the right thing to do.
Comment 6 Zan Dobersek 2013-02-15 12:43:47 PST 
Is everyone now OK with that consensus and the proposed changes or would anyone like to repeat the discussion?
Comment 7 Zan Dobersek 2013-02-18 09:15:00 PST 
Committed r143232: <http://trac.webkit.org/changeset/143232>