The web in a box - a next generation web framework for the Perl programming language

IRC log for #mojo, 2016-01-21

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:10 binlei joined #mojo
00:29 binlei joined #mojo
00:38 sri this is insane http://codepen.io/Yakudoo/pen/YXxmYR/
00:39 Grinnz nothing happens for me
00:40 Grinnz ah, webGL error
00:41 genio That's pretty awesome
00:42 Grinnz also errors on my desktop.. are either of you using chrome?
00:43 sri yes
00:43 genio yes
00:43 Grinnz works in canary o_0
00:52 kablamo joined #mojo
00:55 zivester joined #mojo
01:34 nicomen ;)
01:56 marty joined #mojo
01:57 marty_ joined #mojo
02:15 kaare joined #mojo
02:45 lluad joined #mojo
03:05 jberger I can even make it work in Firefox mobile beta
03:22 jb360 joined #mojo
03:46 noganex_ joined #mojo
04:09 davido joined #mojo
04:10 c--_ joined #mojo
05:17 inokenty-w joined #mojo
05:58 zivester joined #mojo
06:44 panshin joined #mojo
06:57 kes joined #mojo
07:11 kes I run application under morbo. Why $c->req->env is empty? I want to access to 'HTTP_REFERER'
07:37 Vandal joined #mojo
07:43 panshin joined #mojo
07:43 Lee kes: $c->req->headers->referrer
07:44 kes ok. Thank you, Lee
07:45 dod joined #mojo
07:50 dod joined #mojo
08:01 sugar joined #mojo
08:07 AndrewIsh joined #mojo
08:11 sugar joined #mojo
08:45 trone joined #mojo
09:24 panshin joined #mojo
09:30 osfabibisi joined #mojo
09:34 dod joined #mojo
09:39 berov joined #mojo
10:08 rmc joined #mojo
10:15 dod joined #mojo
10:19 rmc Hi. I get an error when running 'make test' against current Mojo 6.41 with TEST_EV=1. 't/mojo/reactor_ev.t' exists with 'not ok 51 recurring was triggered again'.
10:22 rmc Digging deeper i've found out, that the issue went away when reverting the change to lib/Mojo/Util.pm 'update HTML5 entities in Mojo::Util' dated 2015-10-29
10:27 rmc To be more precise...restoring the deleted html entities in the __DATA__ section makes t/mojo/reactor_ev.t run fine again
10:32 sawtooth joined #mojo
10:34 jontaylor joined #mojo
11:34 pink_mist do you know if this was true on 6.40 and 6.39 as well?
11:41 rmc yes...tried 6.26 -> works, 6.27-> error, 6.30->error, 6.32-> error, 6.35 -> error, ....6.40->error
11:42 rmc i must admit and as you can see i haven't tried every version, but it's pretty obvious
11:46 kaare joined #mojo
12:09 binlei joined #mojo
12:19 panshin joined #mojo
12:31 jberger rmc: can you give us some background please, os, perl version etc
12:31 jberger Those things seem unrelated
12:36 rmc @jberger: 5.20.2, OpenBSD 5.9
12:56 panshin joined #mojo
12:57 sawtooth joined #mojo
12:58 binlei joined #mojo
13:15 binlei joined #mojo
13:18 panshin joined #mojo
13:21 panshin joined #mojo
13:32 ajr_ joined #mojo
13:33 binlei joined #mojo
13:43 neilhwatson joined #mojo
13:44 sri rmc: t/mojo/reactor_ev.t does not actually use the entities
13:45 sri there's more known openbsd issues, considering we have nobody on the core team using that os, it will be very hard to debug https://github.com/kraih/mojo/issues/898
13:48 sri is this just a language problem? as in he misunderstands the meaning of "Server available at ..."? https://github.com/kraih/mojo/issues/899
13:50 sri BinGOs: is your openbsd test box running with very limited memory maybe?
13:51 sri it seems to die consistently when the PODRenderer plugin has to render a big file
13:53 BinGOs 896M is hardly limited, but it is being a right arse at the moment.
13:54 sri how odd
13:54 binlei joined #mojo
14:03 BinGOs I've just upped to the available RAM to 1024M
14:23 mspo 899 looks pretty valid to me
14:23 mspo saying I'm litening on 127.0.0.1 when I'm actually on the internet is kind of a problem
14:24 zivester joined #mojo
14:28 vicash just wanted to say that Mojo::Util is awesome. i appreciate the hard work put in creating the hmac_sha1 and other useful functions
14:30 sri mspo: but that's nto what it says
14:31 sri you're mixing up the messages
14:32 sri it's "Listening on "http://*:3000"." in the log and "Available at "http://127.0.0.1:3000"" on STDERR
14:32 sri umm, STDOUT
14:32 sri s/on/at/
14:34 asarch joined #mojo
14:35 sri it's an inconsistency, but 100% correct
14:35 mspo :)
14:35 mspo the best kind of correct?
14:36 moritz btw my chromium doesn't like http://0:3000/ URLs :(
14:36 moritz firefox does
14:37 sri yes, they don't work consistently
14:38 mspo 0:3000?  is that some kind of ipv6 shortcut?
14:38 sri moritz: prolly means 0.0.0.0 and ::
14:39 sri 0.0.0.0 is supposed to be unroutable
14:43 sri it says 127.0.0.1 since 2011 https://github.com/kraih/mojo/commit/a9ec6cfe916f9b27264ecae8bf4b4c452c0ffbf5
14:44 sri hehe, back then we still had Mojo::Server::FastCGI, and it was actually popular :o
14:51 nic for many it was a gateway into the modern world
15:16 odc joined #mojo
15:21 disputin joined #mojo
15:27 ajr_ joined #mojo
15:36 vicash hello. Does Mojo have an email address regex verifier utility function ?
15:42 sri no
15:42 preaction no. frankly there's no such thing. send an e-mail to the address, make them click a link inside
15:42 genio You can verify the structure of the email address with Email::Valid, but as preaction said, there's only one real way to verify an email address
15:44 vicash understood. thank you.
15:48 dod joined #mojo
15:49 rmc sri : thanks for the info. I also get the 'uninitialized value' as seen in #898. Seems to be triggered during 64-bit binary web socket tests
15:51 sri seems pretty harmless
15:52 rmc t/mojolicious/websocket_lite_app.t
15:54 rmc sri : yes indeed. will go on with the reactor_ev test
16:03 binlei joined #mojo
16:03 cpan_mojo Business-Payment-SwissESR-0.11.2 by OETIKER https://metacpan.org/release/OETIKER/Business-Payment-SwissESR-0.11.2
16:06 n16gel joined #mojo
16:11 bwf joined #mojo
16:29 zivester joined #mojo
16:34 n16gel joined #mojo
16:42 lluad joined #mojo
16:43 sh4 joined #mojo
16:57 kaare joined #mojo
17:04 kaare_ joined #mojo
17:04 binlei joined #mojo
17:06 disputin joined #mojo
17:24 dod joined #mojo
17:40 bowtie_ joined #mojo
17:46 jb360 joined #mojo
17:52 ajr_ joined #mojo
18:26 meredith vicash you may be interested in jberger's article here: http://perltricks.com/article/193/2015/9/18/How-to-send-verification-emails-using-Mojolicious
18:30 jberger hmmm, I should update that with SQLite backend :s
18:34 sri ouch, yes, you should
18:35 abra joined #mojo
18:38 jberger done
18:40 sri jberger++
18:40 jberger pluggable backends make that really easy
18:41 jberger sri++
18:42 jberger if and when I ever get my websockify module out I want to do another perl-tricks post about vnc over websocket
18:42 aitap joined #mojo
18:42 sri considering that nobody has adopted the file backend, i guess not that many cared after all
18:42 jberger the SQLite backend makes it kinda moot
18:45 aitap Hi! Is url_escape supposed to work with unicode characters?
18:48 Grinnz I believe it should be utf-8 encoded first, looking at the source
18:50 aitap Ah, that must be the cause, thanks.
18:56 PryMar56 joined #mojo
18:58 aitap Well. When I encode unicode characters to utf8 before url_escape()ing, I get unescaped mojibake (double-encoded utf8) in output.
19:01 aitap By output I mean something like $html = '<a href="'.url_escape("some unicode").'">...</a>'; and later in a template <%== $html %>. Looks ugly, but I need it for wiki-style link processing.
19:02 jberger aitap: you can wrap a string in Mojo::ByteStream to prevent it from being escaped later
19:02 jberger that's basically how all the built-in tag helpers work
19:04 aitap jberger: sorry for not bringing a self-contained example; I hope this will show you what I'm trying to do: https://github.com/aitap/nanowiki/blob/master/nanowiki.pl#L251
19:07 jberger aitap: my comment was just about <%== $html %>
19:07 bowtie joined #mojo
19:07 jberger as to utf8 encoding, you should only encode on the application boundary, which mojolicious does for you
19:09 Grinnz_ aitap: url_escape shouldn't be applied to a whole url, it's for items you'd put in a url parameter striing for example
19:09 Grinnz_ aitap: if you already have a full url, you should xml_escape it to put it in a html tag
19:10 Grinnz_ but i guess if the url has unicode, it's not a real url
19:10 Grinnz_ maybe Mojo::URL can help with constructing the url
19:11 aitap Grinnz_: Is it safe to have utf8 non-ascii letters in href="..." attribute? Most of the browsers just urlencode everything for me, but links2 tried to send utf8 bytes to webserver.
19:12 Grinnz_ aitap: no, they would be percent encoded
19:12 n16gel joined #mojo
19:12 Grinnz_ by Mojo::URL or whatever you use to construct the url
19:12 Grinnz_ browsers usually percent encode what you give them behind the scenes
19:13 aitap Grinnz_: thanks! I'll go with Mojo::URL for encoding
19:26 panshin joined #mojo
19:28 panshin joined #mojo
19:39 aitap Works as expected. Should have stopped to think before trying to nail it with url_escape.
19:41 jb360 joined #mojo
20:00 disputin joined #mojo
20:57 n16gel joined #mojo
21:01 berov1 joined #mojo
21:06 disputin joined #mojo
21:33 berov1 is there any reason why after_dispatch hook can be invoket twice per request?
21:34 * berov1 is wondering what is wrong
21:38 c--__ joined #mojo
21:40 mullagain joined #mojo
22:20 sri crazy how quickly the language Nim vanished again
22:21 * pink_mist must have blinked
22:21 pink_mist because I missed it
22:21 jberger I have a friend who said he's been playing with it
22:21 jberger but yeah, **ffffft** and then it was gone again
22:21 sri it got some hype like 2 years ago
22:22 jberger keeping interest levels up is hard when there are so many cool things to play with
22:24 sri comes to show that a language needs a niche to stick around
22:24 sri at least to reach critical mass
22:29 mcsnolte joined #mojo
22:48 cpan_mojo Clustericious-1.15 by PLICEASE https://metacpan.org/release/PLICEASE/Clustericious-1.15
22:55 c--_ joined #mojo
23:05 n16gel joined #mojo

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary