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

IRC log for #mojo, 2017-03-31

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

All times shown according to UTC.

Time Nick Message
00:02 genio sri, jberger, or Grinnz: If either of you are around that could provide insight on https://github.com/libwww-perl/libwww-perl/issues/86   it seems both Mojo::UserAgent and LWP::UserAgent provide a 200 response, while HTTP::Tiny gives a 599
00:04 Grinnz I do not think it is a requirement of any spec, however the commenter does have a good point that there have been security issues related to it in the past
00:06 genio curl emits a warning and the discussion in #lwp is on whether a warning should be emitted there as well
00:11 genio Net::Async::HTTP errors with "Connection closed while awaiting body"
00:16 genio Grinnz: hrm... http://httpwg.org/specs/rfc7540.html#malformed
00:16 x86 joined #mojo
00:19 genio felipe: actually... "When a Content-Length is given in a message where a message-body is allowed, its field value MUST exactly match the number of OCTETs in the message-body. HTTP/1.1 user agents MUST notify the user when an invalid length is received and detected."
00:19 genio https://tools.ietf.org/html/rfc2616#section-4.4
00:19 genio seems like at least a warning is warranted
00:20 genio sorry, copy paste across channels. felipe isn't in here :)
00:21 genio so, maybe that's a bug in Mojo::UserAgent as well
01:14 cpan_mojo Mojolicious-Plugin-DBViewer-0.29 by KIMOTO https://metacpan.org/release/KIMOTO/Mojolicious-Plugin-DBViewer-0.29
01:29 x86_ joined #mojo
02:14 x86 joined #mojo
02:20 disputin joined #mojo
02:32 disputin joined #mojo
02:37 noganex joined #mojo
02:57 disputin joined #mojo
03:21 kiwiroy joined #mojo
03:54 x86_ joined #mojo
04:02 disputin joined #mojo
04:04 dboehmer_ joined #mojo
04:10 x86 joined #mojo
04:12 mib_cnl6j6 joined #mojo
04:16 x86_ joined #mojo
04:26 x86 joined #mojo
04:44 Vandal joined #mojo
04:45 x86 joined #mojo
04:47 disputin joined #mojo
05:29 x86_ joined #mojo
05:50 janl joined #mojo
05:57 inokenty-w joined #mojo
06:03 x86 joined #mojo
06:27 dod joined #mojo
06:33 dod joined #mojo
06:35 salv0 joined #mojo
06:42 prg joined #mojo
06:54 janl_ joined #mojo
07:03 AndrewIsh joined #mojo
07:22 janl joined #mojo
07:29 trone joined #mojo
07:33 dod joined #mojo
07:36 dod joined #mojo
07:37 disputin joined #mojo
07:39 stryx` joined #mojo
08:53 sri genio: warnings seem kinda pointless
08:54 sri you don't warn about a security issue, you throw a loud exception
08:54 rshadow joined #mojo
08:54 sri but i'll let the community decide
09:21 marcus I agree with sri
09:24 irqq joined #mojo
10:05 kiwiroy joined #mojo
10:12 kiwiroy joined #mojo
10:39 batman Me too
10:41 batman Or... how about setting $res->error ?
11:09 pink_mist is there an issue opened about this yet?
11:10 pink_mist ah there is
11:12 kiwiroy joined #mojo
11:15 tchaves joined #mojo
11:23 pink_mist sri: https://rt.cpan.org/Public/Bug/Display.html?id=112309 \patched/
11:23 sri pink_mist: yay
11:37 dod joined #mojo
11:43 dod joined #mojo
11:44 dod joined #mojo
12:08 sri btw. i could really use periodic jobs support for minion
12:08 sri if someone knows a good design that would fit into minion please let me know https://github.com/kraih/minion/issues/24
12:52 dantti_laptop joined #mojo
13:05 genio sri: I'm all for loud exceptions in this case.  no argument here
13:08 genio my suggestion for a warning was solely to be an "easing-into" type of change.
13:12 kiwiroy joined #mojo
13:12 maschine joined #mojo
13:14 pink_mist that's not how mojo dev usually is I think :P
13:14 maschine So I'm testing the form validation example shown here: http://mojolicious.org/perldoc/Mojolicious/Guides/Rendering#Form-validation
13:14 maschine But something seems off.  This is my test project:
13:14 maschine http://bcoder.westus.cloudapp.azure.com:3000/test
13:15 maschine I added a comparison to an existing username (because I want to use this to register new users).  That part works; if it matches an existing user it flashes a message
13:16 maschine but if you change the username and hit enter, it submits the last form result even though I've changed it.  And if you look at the URI, it still has the old "name" in the string
13:17 maschine for example, enter "testing" as the username (password doesn't matter).  it will say "testing is already used".  Then change it to anything else, and submit again.  It will still say "testing is already used".  It's not until you submit it again that it will pass.
13:18 maschine code: http://pasted.co/d3aaf782
13:19 genio Net::Async::HTTP throws an exception already.  It'd be kind of nice if every Perl-based useragent did the same thing
13:27 pink_mist genio: has it been tested what the different UAs do if it gets too *much* data rather than too little?
13:28 pink_mist *if they get
13:29 genio the wording of the RFC indicates that the same error should be thrown.  I haven't actually tested though
13:32 pink_mist the way the Net::Async::HTTP exception is worded makes me think that wouldn't happen if too much data was sent
13:45 Lee joined #mojo
14:25 itaipu joined #mojo
14:35 maschine I figured it out.  It probably seems obvious to many, but it was because I was using flash to flag the username being matched with an existing one, I switched to stash and now it works..
14:56 disputin joined #mojo
15:00 dave joined #mojo
15:15 stryx` joined #mojo
15:41 stryx` joined #mojo
16:11 perlpilot joined #mojo
16:40 perlpilot joined #mojo
16:42 lluad joined #mojo
16:49 PryMar56 joined #mojo
16:50 mib_4ay8pc joined #mojo
17:19 jberger maschine learning when to flash vs stash does take a bit
17:19 jberger when in doubt, stash
17:25 pink_mist almost never flash
17:25 pink_mist I'd say
17:25 stryx` joined #mojo
17:29 genio not unless you're being paid
17:30 lluad I don't think I've ever used flash for anything other than messages after a redirect.
17:31 * coolo read 'machine learning' and couldn't figure out the grammar. Took a bit to realize you talked to a machine :)
17:35 _dave_ joined #mojo
17:35 disputin joined #mojo
17:38 maschine it was actually after a redirect.. but it wasn't working the way I expected it to.  Anyway, lesson learned. :)
17:48 irqq_ joined #mojo
18:08 stryx` joined #mojo
18:13 kiwiroy joined #mojo
18:14 x86 joined #mojo
18:15 rshadow joined #mojo
18:57 stryx` joined #mojo
19:09 irqq joined #mojo
19:15 stryx` joined #mojo
19:32 jberger coolo: hehe
20:13 kiwiroy joined #mojo
21:07 PryMar56 joined #mojo
21:13 kiwiroy joined #mojo
21:41 arcanez when doing 'include', does the current stash get passed automatically?
21:42 pink_mist not sure what you mean by "the current" stash ... but the stash will be available there too, yes?
21:43 arcanez the way you phrased it is what I meant, thanks
22:00 batman arcanez: you can also pass on additional variables, which will get merged with the current stash
22:00 batman %= include "foo", x => 123
22:02 pink_mist s/current//
22:03 batman pink_mist: depends if you mean $c->stash or the $variables that is available in the template
22:04 batman but i guess you're more precise
22:20 arcanez % Perl code line, treated as "<% line =%>" (explained later)
22:20 arcanez is that really just "<% line %>"
22:20 arcanez ?
22:21 pink_mist no
22:21 pink_mist you missed the = at the end
22:21 pink_mist (and that's explained later)
22:21 arcanez trim whitespace, I see.
22:21 pink_mist (I think that could be more explicit that it's the = that'll be explained later
22:21 pink_mist )
22:22 maschine yeah I noticed that too.. it's not really clear where "later" is.
22:23 arcanez coming from HTML::Mason, I don't expect some magic like that (at least for "% Perl code" meaning <% =%>
22:23 arcanez also not used to having to use <%== %> but it's minimal
22:59 batman arcanez: should be very careful with `%==`
23:47 irqq_ joined #mojo

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