[asterisk-dev] [Code Review] 3248: Fix for WebRTC over WSS not working

[wdoekes]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3248/#review11068
-----------------------------------------------------------

Ship it!


Minor stylistic issues left:


/branches/11/res/res_http_websocket.c
<https://reviewboard.asterisk.org/r/3248/#comment20712>

    Generally we use (rlen == 0), not (0 == rlen), and you do too, on the same line with errno != EAGAIN.
    
    But I'm not sure there is a strong consensus there.
    
    
    
    And below I see a bunch of excess parentheses around pointer dereferences:
    
      (*opcode) => *opcode
      (*payload) => *payload
      (*payload_len) => *payload_len
    
    Here it makes sense (the first reference):
    
      if ((*payload_len) && (new_payload = ast_realloc(session->payload, *payload_len))) {
    
    Here it doesn't:
    
      if (ws_safe_read(session, &buf[frame_size], (*payload_len), opcode)) {
    
    Consistency is nice.



/branches/11/res/res_http_websocket.c
<https://reviewboard.asterisk.org/r/3248/#comment20711>

    *Exiting


- wdoekes


On March 3, 2014, 1:19 a.m., Moises Silva wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/3248/
> -----------------------------------------------------------
> 
> (Updated March 3, 2014, 1:19 a.m.)
> 
> 
> Review request for Asterisk Developers and rnewton.
> 
> 
> Bugs: ASTERISK-21930 and ASTERISK-23099
>     https://issues.asterisk.org/jira/browse/ASTERISK-21930
>     https://issues.asterisk.org/jira/browse/ASTERISK-23099
> 
> 
> Repository: Asterisk
> 
> 
> Description
> -------
> 
> Several fixes for the WebSockets implementation in res/res_http_websocket.c
> 
> * Flush the websocket session FILE* as fwrite() may not actually guarantee sending
>   the data to the network. If we do not flush, it seems that buffering on the SSL
>   socket for outbound messages causes issues
> 
> * Refactored ast_websocket_read to take into account that SSL file descriptors
>   may be ready to read via fread() but poll() will not actually say so because
>   the data was already read from the network buffers and is now in the libc buffers
> 
> This should fix an issue that I have experienced and other users may have reported [1][2][3], where
> secure websockets wouldn't work, messages seem to not make it into Asterisk
> 
> [1] http://lists.digium.com/pipermail/asterisk-users/2013-August/280175.html
> [2] https://issues.asterisk.org/jira/browse/ASTERISK-21930
> [3] https://issues.asterisk.org/jira/browse/ASTERISK-23099
> 
> 
> Diffs
> -----
> 
>   /branches/11/res/res_http_websocket.c 409360 
> 
> Diff: https://reviewboard.asterisk.org/r/3248/diff/
> 
> 
> Testing
> -------
> 
> See ASTERISK-21930 for details on other users testing these changes. I did both WS and WSS calls, confirmed audio works with chrome. This patch is for Asterisk 11 as the issue is reported on Asterisk 11, but I tested a few months ago and same issue existed on 12 and trunk. I created my own team branches for those too (/team/moy/webrtc-11, /team/moy/webrtc-12, /team/moy/webrtc-trunk)
> 
> Confirmed working by Sean Bright on Jan 20, 2014 on Asterisk 11 (see ASTERISK-21930 comment)
> 
> 
> Thanks,
> 
> Moises Silva
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140305/1e4699ee/attachment-0001.html>