Closed
Bug 664860
Opened 13 years ago
Closed 13 years ago
Rename several WebSockets classes/IDLs to match necko conventions before we ship WS
Categories
(Core :: Networking: WebSockets, defect)
Core
Networking: WebSockets
Tracking
()
RESOLVED
FIXED
mozilla7
People
(Reporter: jduell.mcbugs, Assigned: jduell.mcbugs)
References
Details
(Whiteboard: [inbound])
Attachments
(4 files)
5.53 KB,
patch
|
Biesinger
:
review+
|
Details | Diff | Splinter Review |
19.01 KB,
patch
|
Biesinger
:
review+
|
Details | Diff | Splinter Review |
75.25 KB,
patch
|
Biesinger
:
review+
|
Details | Diff | Splinter Review |
153.69 KB,
patch
|
Biesinger
:
review+
|
Details | Diff | Splinter Review |
Right now the network transport for Web sockets is called nsIWebSocketProtocol.idl, and it's implemented by the nsWebSocketHandler class.
Everywhere else in necko we reserve "Protocol" and "Handler" for singleton, protocol-wide classes (see nsHttpHandler, nsFtpProtocolHandler, both of which inherit from nsIProtocolHandler.idl). So this is needlessly confusing.
We should rename
nsIWebSocketProtocol -> nsIWebSocketChannel
nsWebSocketHandler -> WebSocketChannel
I'd especially like to see the IDL change before we ship WS for the first time, so we don't need to break XPCOM interfaces.
Assignee | ||
Comment 1•13 years ago
|
||
Attachment #543636 -
Flags: review?(mcmanus)
Assignee | ||
Comment 2•13 years ago
|
||
Attachment #543637 -
Flags: review?(mcmanus)
Assignee | ||
Comment 3•13 years ago
|
||
E10s classes are already named WebSocketChannelChild, etc.
Assignee | ||
Comment 4•13 years ago
|
||
Re-indent WebsocketChannel. It's now or never--we don't want to do this after the Aurora branch.
Contains some cleanup of LOG msgs too.
Attachment #543639 -
Flags: review?(mcmanus)
Assignee | ||
Comment 5•13 years ago
|
||
I've assigned these to Patrick, but there's actually no code logic changes here, so pretty much anyone can review. I'd really like to land them before we fork for Aurora. If anybody's got time to glance over them (just lots of class renaming) I'd appreciate it.
We don't have a backlog of WS patches right now, so patch breakage isn't an issue. The most "interesting" decision here is that we'll lose the hg blame for WebSocketChannel with the last, re-indent patch. But the hg log is pretty short now, so I'm not seriously worried about it.
Updated•13 years ago
|
Attachment #543636 -
Flags: review?(mcmanus) → review+
Updated•13 years ago
|
Attachment #543637 -
Flags: review?(mcmanus) → review+
Updated•13 years ago
|
Attachment #543638 -
Flags: review?(mcmanus) → review+
Updated•13 years ago
|
Attachment #543639 -
Flags: review?(mcmanus) → review+
Assignee | ||
Comment 6•13 years ago
|
||
http://hg.mozilla.org/integration/mozilla-inbound/rev/58c721a6bb1a
http://hg.mozilla.org/integration/mozilla-inbound/rev/0bfd2a8382c7
http://hg.mozilla.org/integration/mozilla-inbound/rev/3ebb4b775eee
http://hg.mozilla.org/integration/mozilla-inbound/rev/8df23f5d9709
Whiteboard: [inbound]
Comment 7•13 years ago
|
||
http://hg.mozilla.org/mozilla-central/rev/58c721a6bb1a
http://hg.mozilla.org/mozilla-central/rev/0bfd2a8382c7
http://hg.mozilla.org/mozilla-central/rev/3ebb4b775eee
http://hg.mozilla.org/mozilla-central/rev/8df23f5d9709
Status: ASSIGNED → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla7
You need to log in
before you can comment on or make changes to this bug.
Description
•