Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-lwp-gsoc, 2014-07-06

| Channels | #perl6-lwp-gsoc index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:48 FROGGS__ joined #perl6-lwp-gsoc
03:19 dalek joined #perl6-lwp-gsoc
12:21 sergot hi o/
12:24 sergot will nativecast be merged to the master branch?
12:26 sergot :)
12:55 FROGGS__ sergot: it will
12:55 FROGGS__ :o)
12:55 FROGGS__ jvm was done as you know, parrot passes the first three tests before blowing up
12:56 FROGGS__ and btw, porting it to parrot was worse than to jvm...
15:00 FROGGS[mobile] or rather still is
15:37 sergot ohh, I wouldn't say so.
16:03 sergot FROGGS[mobile]: does it make sens to write functions in C for tcp connections?
16:03 sergot sense
16:03 FROGGS[mobile] we have that in MoarVM
16:06 sergot Can I call it directly from p6 script?
16:07 sergot somehow, maybe usinh nqp if it's mapped.
16:08 sergot I just need a connection's fd.
16:10 FROGGS[mobile] look at IO::Socket...
16:10 FROGGS[mobile] there must be nqp ops
16:11 sergot true.. I'm sorry for those questions, just expecting info about the best way to do this, or something like that. I'll do some reaserch now, thanks :)
16:47 FROGGS[mobile] what are you about to do anyway?
17:37 dalek http-useragent: f9d0b7d | sergot++ | lib/HTTP/UserAgent.pm6:
17:37 dalek http-useragent: use HTTP::Cookies added
17:37 dalek http-useragent: review: https://github.com/sergot/http-useragent/commit/f9d0b7ddc6
17:41 dalek http-useragent: 02e63bb | sergot++ | t/03-cookies.t:
17:41 dalek http-useragent: failing tests removed (TODO)
17:41 dalek http-useragent: review: https://github.com/sergot/http-useragent/commit/02e63bb532
18:47 FROGGS__ sergot: in IO::Socket the $!PIO might contain a fd, so you could use nqp::readfh and nqp::printfh
18:50 sergot FROGGS__: yes, this looks promising
18:51 FROGGS__ but I would encourage you to use that IO::Socket role and its methods...
18:52 FROGGS__ because you end up with duplicated code otherwise, and backend specific fudges are always painful in modules
18:55 sergot true, I would, but I wanted to start with some maybe-not-clear but workable code.
18:56 sergot I will be harder to do this with IO::Socket than with pure C I think.
18:56 sergot It*
18:58 sergot because:
18:59 sergot how can I use SSL_set_fd without knowing connection's fd?
18:59 sergot I mean
18:59 sergot hm, wait
18:59 sergot Ok, I'll try
19:00 FROGGS__ there must be a C function in libSSL that can return the fd
19:01 FROGGS__ or... it relies on you setting up the connection, and then pass the fd along to libSSL
19:01 sergot yes, the second one
19:02 sergot I must set up the connection and then use SSL_set_fd to let it know about this connection.
19:03 FROGGS__ my $fd = nqp::getattr($connection, IO::Socket, '$!PIO');
19:03 FROGGS__ that should do I think
19:50 sergot ok, thanks :)
19:52 dalek joined #perl6-lwp-gsoc

| Channels | #perl6-lwp-gsoc index | Today | | Search | Google Search | Plain-Text | summary