Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-lwp-gsoc, 2014-07-18

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

All times shown according to UTC.

Time Nick Message
01:24 FROGGS_ joined #perl6-lwp-gsoc
02:10 dalek joined #perl6-lwp-gsoc
02:10 sergot joined #perl6-lwp-gsoc
02:21 sergot joined #perl6-lwp-gsoc
02:39 dalek joined #perl6-lwp-gsoc
02:47 dalek joined #perl6-lwp-gsoc
03:07 d4l3k_ joined #perl6-lwp-gsoc
03:09 sergot joined #perl6-lwp-gsoc
03:14 sergot joined #perl6-lwp-gsoc
03:30 dalek joined #perl6-lwp-gsoc
04:01 dalek joined #perl6-lwp-gsoc
04:07 d4l3k_ joined #perl6-lwp-gsoc
04:27 dalek joined #perl6-lwp-gsoc
04:46 dalek joined #perl6-lwp-gsoc
04:51 dalek joined #perl6-lwp-gsoc
04:58 sergot joined #perl6-lwp-gsoc
05:08 dalek joined #perl6-lwp-gsoc
06:48 sergot o/
07:05 FROGGS_ hi
07:34 sergot FROGGS: let's talk about our project :)
07:34 FROGGS yeah?
07:34 sergot What do you think, which thing is the most important now?
07:35 sergot to do
07:36 sergot You are probably busy right now, arent you?
07:36 FROGGS just a bit
07:36 FROGGS I think we should add more tests to http::useragent, perhaps from the Perl 5 module
07:37 FROGGS so that we have more use cases, and maybe spot problems
07:37 sergot true, I like this idea :)
07:39 moritz write a crawler that extracts links from some pages, and randomly follows one or two of them
07:39 moritz you'll find lots and lots of problems that way
07:40 moritz (preferably follow cross-domain links, so that you'll get to talk to many different web servers)
07:40 FROGGS +1
07:41 FROGGS and you might want to record the links you followed, so you can replay
07:43 sergot yeah! I like it :)
07:50 moritz when I implemented the IRC logs, I was surprised by the amount that my (implicit) assumptions were broken again and again by the real world data :-)
07:59 FROGGS do not assume anything then :P
08:12 moritz that's kinda hard :-)
08:13 moritz for example, there are codepoints that aren't valid in XML
08:13 moritz (iirc unsigned codepoints, and a few others)
08:13 moritz but they can appear in IRC input
09:29 sergot uh, SSL is still very buggy
09:30 FROGGS in what way?
09:31 sergot I deosnt work :D
09:32 sergot I tried https://github.com and it failed
09:32 FROGGS eww
09:32 sergot $ http-request https://github.com
09:32 sergot Cannot look up attributes in a type object
09:33 sergot it works only for filip.sergot.pl - probably because this site doesnt use ssl at all (so 403 is there)
09:33 sergot I tried some sites like https://home.pl
09:33 sergot yea, a lot of debugging work
09:35 sergot $ http-request https://perlgeek.de
09:35 sergot Cannot look up attributes in a type object
09:35 sergot :)
09:35 sergot in method write at lib/OpenSSL.pm6:53
09:35 sergot we'll see
10:00 sergot something wrong with $!ssl there
10:02 FROGGS so the connection was not established?
10:03 sergot perhaps
10:15 sergot yeah, the fd is 0
10:16 FROGGS then I guess the handshake does not succeed
10:17 sergot it's 0 before we do the ssl stuff
10:17 sergot client_connect returns 0
10:17 FROGGS ohh
10:17 FROGGS well, might be even easier to solve
10:17 sergot yes :)
10:19 sergot https://github.com/sergot/io-socket-ssl/blob/master/src/libclient.c#L20
10:19 sergot fails here
10:20 FROGGS huh
10:20 sergot maybe something is wrong with the char *hostname
10:21 sergot working on this
10:29 sergot perlgeek.de�3
10:29 sergot yes, here is the fault
10:41 sergot FROGGS: should we cut somehow this string?
10:42 sergot why p6 pass it longer than it actually is
10:42 sergot ?
10:42 sergot passes*
10:44 sergot can I add another argument (int len) which will specify where C should end reading the string? (add \0 after it)
10:45 FROGGS hmmm
10:45 FROGGS and there is no \0 already?
10:45 sergot looks like that
10:45 sergot we got perlgeek.de�3
10:45 sergot we expect: perlgeek.de
10:46 sergot when I pass "perlgeek.de" as an argument to failing funtion it passes
10:47 FROGGS sub client_connect(CArray[uint8], int32) returns int32 { * }
10:47 FROGGS why isnt it Str, int32?
10:47 sergot hmm, I dont know...
10:48 sergot seems like I forgot about Str or something :D
10:48 FROGGS because when you tell it that it is a Str, it will null terminate it
10:48 FROGGS :o)
10:49 sergot ok, so here is the fix :")
10:51 sergot yep, it works, FROGGS++
10:53 FROGGS nice
10:56 sergot it prints https://perlgeek.de's content!
10:56 sergot maybe it will even work with github then
10:57 FROGGS \o/
10:57 sergot checking
10:57 sergot no, it doesnt... :(
10:58 FROGGS yeah, somehow I had expected that
11:08 dalek io-socket-ssl: 09501e3 | sergot++ | lib/IO/Socket/SSL.pm6:
11:08 dalek io-socket-ssl: carray fix
11:08 dalek io-socket-ssl:
11:08 dalek io-socket-ssl: if we put Str there, it is null terminated automatically
11:08 dalek io-socket-ssl: FROGGS++
11:08 dalek io-socket-ssl: review: https://github.com/sergot/io-socket-ssl/commit/09501e334f
11:14 dalek io-socket-ssl: a5d199c | sergot++ | src/libclient.c:
11:14 dalek io-socket-ssl: port fix tadzik++
11:14 dalek io-socket-ssl: review: https://github.com/sergot/io-socket-ssl/commit/a5d199cc4f
11:27 sergot I became ACM member today.
11:27 sergot because of GSoC
11:27 sergot how about you FROGGS as a mentor?
11:27 sergot or moritz
11:28 FROGGS what is ACM?
11:28 sergot Association for Computing Machinery
11:28 moritz I wouldn't object to an ACM membership.
11:28 FROGGS never heard of it :o)
11:29 sergot with a big resource of books, videos and courses.
11:29 sergot Didnt you get any email from google?
11:29 sergot Maybe it's only for students.
11:29 moritz nope, no email
11:29 FROGGS no email
11:29 moritz seems it's just for students
11:29 sergot ok :(
11:30 sergot I think mentors should get some benefits from being a mentor.
11:30 moritz we get a t-shirt :-)
11:30 moritz and our org gets 50USD or something for each (primary) mentor
11:31 sergot hmm, it's not a lot, you deserve much more :)
11:32 moritz FROGGS does; I mostly just hang out here and make a smart-ass remark now and then :-)
11:32 moritz no really; our benefit is the student's involvment, and resulting code
11:32 FROGGS I thought it was 500USD
11:33 FROGGS but yeah, the shirts are nice too :P
11:33 moritz oh, possible
11:33 FROGGS and I do not expect or want anything, FOSS, you know :o)
11:34 moritz oh, I want lots of things, including world peace :-)
11:34 FROGGS yeah
11:34 FROGGS I just do that here so that there will be a Star Trek-ish future some day
12:04 sergot Nice :)
12:11 sergot FROGGS: I have a question.
12:12 sergot or.. wait
12:12 FROGGS hehe
12:14 sergot FROGGS: https://github.com/sergot/http-useragent/blob/master/lib/HTTP/UserAgent.pm6#L86
12:14 sergot here is the bug
12:14 sergot when I use https://github.com
12:14 sergot it fails because:
12:15 sergot it cant find header/content separator
12:15 FROGGS :/
12:15 sergot because header section is to long
12:15 sergot too long
12:15 FROGGS okay, simple solution is simple
12:15 FROGGS as long as we do not hit the sep, fetch more
12:15 sergot yeah :)
12:16 FROGGS and then adjust that comment: https://github.com/sergot/http-useragent/blob/master/lib/HTTP/UserAgent.pm6#L81
12:16 FROGGS :o)
12:16 sergot okok :)
12:22 sergot almost done
12:23 sergot brb
12:23 sergot &
13:29 sergot FROGGS: how can I make one buf from two (or more) buffs?
13:30 sergot bufs*
13:30 sergot I mean, the easiest way
13:31 sergot or, add elems from one buf to another
14:02 dalek http-useragent: c43b091 | sergot++ | lib/HTTP/UserAgent.pm6:
14:02 dalek http-useragent: (header too long) fix
14:02 dalek http-useragent: review: https://github.com/sergot/http-useragent/commit/c43b091da3
14:03 sergot FROGGS: what about this?
14:06 sergot hmm, malformed utf now..
14:06 sergot FROGGS: could you take a look at this?
14:06 sergot have to go, I will be later $
14:06 sergot &
14:32 FROGGS I've commented on your commit
15:31 sergot ok, thanks, and what about the utf problem?
15:31 FROGGS I dunno... where does it explode?

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