[Webkit-unassigned] [Bug 26242] [GTK] The way we handle optional features is not quite right

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=26242





------- Comment #3 from gns at gnome.org  2009-06-07 13:07 PDT -------
+1

I was happy to see that patch landed (in fact, I did not r+ it only because Jan
had already worked on it, so I decided to leave it for him), but I agree we
should have a better solution, and that this is a good time to design it, both
as a developer and as a Debian maintainer - this would make my life much easier
if designed right.

So my take on this, would be:

/usr/lib/webkit/1.0/modules

There we would have gnome-keyring.so, soup-gnome.so. I think we want these to
be global API, so not attached to webview or settings at all. So, say:

webkit_use_feature(string, boolean) -> boolean

Those strings would be the names of the modules, without '.so'. Our apps would
be optimists - just request whatever features they would like to enable, our
code would try to load the module, and if the symbols are not resolvable,
return FALSE, returning TRUE when they are, and init worked OK, and they'll
know which are not available through the return value.

How does that sound?


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