The patch has been improved, thanks for your review :) Regards, Kay
On Wed, Apr 4, 2012 at 9:11 AM, Eduardo Silva <[email protected]> wrote: > Thanks for the update, > > a minor improvement required: > > + if (sr->uri_processed.len == strlen(subprotocol_list[i].name) && > + strncmp(sr->uri_processed.data, subprotocol_list[i].name, > + strlen(subprotocol_list[i].name)) == 0) { > > if you have a fixed length string do not perform strlen() every time, > that is expensive. Once you fix that i will publish your patch. > > regards, > > On Wed, Apr 4, 2012 at 10:05 AM, Kay <[email protected]> wrote: > > Sorry about that, the new attached patch can work now. > > > > Regards, > > Kay > > > > > > > > On Wed, Apr 4, 2012 at 7:44 AM, Eduardo Silva <[email protected]> wrote: > >> > >> Hi, > >> > >> you forget to include the header: > >> > >> gcc -g -Wall -DDEBUG -DSAFE_FREE -I../../src/include -fPIC -c ws.c > >> ws.c:46:18: fatal error: echo.h: No such file or directory > >> > >> please also make sure to do not include trailing whitespaces: > >> > >> /home/edsiper/Downloads/websocket_echo_subprotocol_enabled.patch:183: > >> trailing whitespace. > >> unsigned char *frame_payload; > >> /home/edsiper/Downloads/websocket_echo_subprotocol_enabled.patch:207: > >> trailing whitespace. > >> else if (payload_length == 127) { > >> /home/edsiper/Downloads/websocket_echo_subprotocol_enabled.patch:259: > >> trailing whitespace. > >> > >> regards, > >> > >> On Mon, Apr 2, 2012 at 8:54 AM, Kay <[email protected]> wrote: > >> > This patch implements the function of sending frame and sample > >> > subprotocol: > >> > echo, it simply send back the payload it received. Now it can work > with > >> > the > >> > echo test in websocket.org. This can also be regarded as framework > for > >> > further development, it must have drawbacks can be improved, so any > >> > suggestions are appreciated. > >> > > >> > This patch is based on the one applied to master, but not my previous > >> > one(Includes it). If the previous one has not be applied, then forget > >> > it. > >> > > >> > This is the first time I use epoll, and I want to ask a question about > >> > it. I > >> > think the socket should be set to edge triggered, or the write event > >> > will be > >> > kept invoking. Right? But why I didn't see any code in monkey change > the > >> > mode to MK_EPOLL_EDGE_TRIGGERED. Thanks for your explanation. > >> > > >> > --Kay > >> > > >> > > >> > > >> > _______________________________________________ > >> > Monkey mailing list > >> > [email protected] > >> > http://lists.monkey-project.com/listinfo/monkey > >> > > >> > >> > >> > >> -- > >> Eduardo Silva > >> http://edsiper.linuxchile.cl > >> http://www.monkey-project.com > > > > > > > > -- > Eduardo Silva > http://edsiper.linuxchile.cl > http://www.monkey-project.com >
websocket_echo_subprotocol_enabled.patch
Description: Binary data
_______________________________________________ Monkey mailing list [email protected] http://lists.monkey-project.com/listinfo/monkey
