[Webkit-unassigned] [Bug 44261] [GTK] Add HTTP caching support

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=44261





--- Comment #79 from Sergio Villar Senin <svillar at igalia.com>  2010-09-25 04:06:51 PST ---
(In reply to comment #75)
> Created an attachment (id=68717)
 --> (https://bugs.webkit.org/attachment.cgi?id=68717) [details]
> patches to Sergio's patches
> 
> OK, the reason SoupRequester needed a special case to get file URIs working right was because there was already an incorrect special case for file URIs in ResourceHandleSoup::startGio() that was messing the URL up before it got to SoupRequester. So this patch removes both special cases (I didn't bother attaching it in the form of a git commit since it actually needs to be split between both of Sergio's existing patches.).
> 

Well that's not really the reason. The special case in souprequester needs the special case in startGio. One cannot work without the other. With both of them it works fine.

> Additionally, to deal with the problem the startGio() hack was originally attempting to work around, we need to make soup_uri_decode() accept badly-%-encoded URIs rather than returning NULL (which is something we'd already discussed in gnomebug 523100 anyway). There is now a patch for that in https://bugzilla.gnome.org/show_bug.cgi?id=630540 but it's not yet committed because we're in hard code freeze for 2.32.

That's the real issue IMO. That's why my first thought was to modify the decoding in soup-uri.c although your changes seem indeed better than adding a special case for file: URL's. I also considered removing that fixup parametter but I was not sure about breaking current URI parsing in soup. Note that glib APIs do return NULL when they cannot decode URIs

One last question Dan, did you run the tests with both patches?

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