| Summary: | Remove WebSocketHandshakeRequest class | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Anders Carlsson <andersca> | ||||||
| Component: | New Bugs | Assignee: | Anders Carlsson <andersca> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | ap, eflews.bot, gyuyoung.kim, kling, sam, webkit-ews | ||||||
| Priority: | P2 | ||||||||
| Version: | 528+ (Nightly build) | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Attachments: |
|
||||||||
|
Description
Anders Carlsson
2013-05-15 12:44:17 PDT
Created attachment 201867 [details]
Patch
Comment on attachment 201867 [details] Patch Attachment 201867 [details] did not pass qt-wk2-ews (qt-wk2): Output: http://webkit-queues.appspot.com/results/476631 Comment on attachment 201867 [details] Patch Attachment 201867 [details] did not pass efl-wk2-ews (efl-wk2): Output: http://webkit-queues.appspot.com/results/473621 Created attachment 201871 [details]
Patch
Comment on attachment 201871 [details]
Patch
I'ma go ahead and trust you on this one, Carlsson.
Committed r150142: <http://trac.webkit.org/changeset/150142> > Turns out WebSocketHandshakeRequest is just used by the web inspector, and there's no reason
why we can't just use a ResourceRequest instead.
This is not accurate. WebSocket has different semantics than HTTP, in particular, order of header fields matters.
I'm puzzled that you are making such changes without any attempt of talking to me. Andreas, I suggest waiting at least 24 hours before reviewing patches in areas you are not sure that you are an expert in.
(In reply to comment #7) > > Turns out WebSocketHandshakeRequest is just used by the web inspector, and there's no reason > why we can't just use a ResourceRequest instead. > > This is not accurate. WebSocket has different semantics than HTTP, in particular, order of header fields matters. > > I'm puzzled that you are making such changes without any attempt of talking to me. Andreas, I suggest waiting at least 24 hours before reviewing patches in areas you are not sure that you are an expert in. Alexey, I trusted Anders's judgment on this change and it looked like a straightforward removal of an unnecessary abstraction. I set my r+ without fully understanding the implications, and for that I apologize. > order of header fields matters
This is no longer the case, the spec no longer requires that. I could not quickly find any other differences that would require using a distinct class.
|