[Webkit-unassigned] [Bug 32452] Port of v8 to work with Gtk webkit

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=32452





--- Comment #27 from Michael Emmel <mike.emmel at gmail.com>  2009-12-17 17:31:54 PST ---
Created an attachment (id=45113)
 --> (https://bugs.webkit.org/attachment.cgi?id=45113)
Another way to generate naming conflicts


Ok I though of another way to generate naming conflicts assuming the use of
undef.

In this approach I use -include to force the conflicting headers to be include
in general with attempts 
to use undef to resolve the conflicts.

This generates yet another set of problems with that approach.

I'm now back to suggesting that the assumption of C++ namespaces as sufficient
to protect the use of common names is not a good one in general.

So your left with either removing the conflict via renaming or a case by case
refactoring often in seemingly unrelated headers.

I'm back to my generic observation that a library should try and be as
unobtrusive as possible and assume its own headers that its being included in
unknown projects and do its best to minimize naming conflicts this would
include the use of reasonable old style C prefixes to handle preprocessor
issues in its public headers.

Certainly this rule is broken by ancient code but the ancient code is one of
the reasons for the development of C++ namespaces in the first place, because
of C namespace conflicts. However the old code is still around causing
problems.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.