[Webkit-unassigned] [Bug 43204] Add symbol export macros to C++ DOM bindings

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Jul 30 23:27:38 PDT 2010


https://bugs.webkit.org/show_bug.cgi?id=43204





--- Comment #6 from Kevin Ollivier <kevino at theolliviers.com>  2010-07-30 23:27:39 PST ---
(In reply to comment #5)
> (From update of attachment 62999 [details])
> Patch looks fine: Though I'd suggest to add a new wtf/VisibilityMacros.h file or (wtf/ExportControl.h whatever you can come up with), that tries to unifies the existing macros.
> I'd feel more comfortable reviewing such a patch...
> 
> What do you think? Can you try it?

I would like to, at least at some point, but this issue is thornier than it looks on the surface. (Something I learned in #27551.) To use these macros pretty consistently throughout the codebase, we need to know for each ports which headers they allow a particular project to include, which headers they may (or may not) copy over from JSCore, etc., and we may also need to modify at least a couple of the ports' build systems, particularly for things like JSCore's testAPI, which is supposed to only be able to access JSCore public API (JavaScriptCore/API) headers. As I was somewhat surprised to discover, several ports treat wtf as an internal implementation detail of JSCore, so in certain areas including headers from it will break some intentionally designed public / private separation. This is part of the reason we've ended up with several duplications of these macros in the first place, and part of the reason that I'm not 100% clear on how I'd even go about implementing the sol
 ution.

Given the issues involved, I think the appropriate thing to do here is to file a new bug on the consistency issue and perhaps get some feedback on the idea and suggestions on how to handle it best. I think we should consider this patch by itself for now.

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



More information about the webkit-unassigned mailing list