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

IRC log for #mojo, 2017-01-24

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

All times shown according to UTC.

Time Nick Message
00:22 Ryoga joined #mojo
00:42 bpmedley http://24.72.144.208:3000/minion_ui <-- The UI has been updated.  Any feedback welcome..
00:42 Grinnz not a fan of the tabs on the bottom. left side or top is the intuitive spot
00:43 genio on the bottom kind of makes sense in a mobile environment
00:43 bpmedley Grinnz : Thanks!  Hrmm, I will see if I can change the layout, and compare them..
00:44 Grinnz mm yeah in mobile maybe
00:44 bpmedley I should have mentioned, the UI is mobile 1st, or that's the goal..
00:44 Grinnz i do see that sometimes there, but there's also just as common, a tray that's swiped out from the left
00:44 bpmedley Grinnz : Are you referring to a drawer type menu?
00:44 Grinnz yea
00:44 genio true. I think I see more of that now
00:45 bpmedley Point taken.  Either one is possible, and I'll tinker..
00:55 bpmedley Is it easily discernible that the fingerprint is clickable / tappable?
00:55 pink_mist no
00:55 pink_mist it isn't
00:56 bpmedley pink_mist : Hrmm.  Thanks.  Back to the drawing board.. :)
00:57 htoetl write "menu" upon it )
00:57 htoetl it should do the trick of making it more intuitive )
01:20 sugar joined #mojo
01:31 disputin joined #mojo
01:55 jnbek joined #mojo
03:06 marty joined #mojo
03:21 kiwiroy joined #mojo
03:22 bpmedley http://24.72.144.208:3000/minion_ui <-- UI helper for fingerprint, updated for using drawer (rather than tabs), more information on display
03:48 noganex joined #mojo
03:58 disputin joined #mojo
04:29 zivester joined #mojo
04:43 prajith joined #mojo
05:04 dboehmer_ joined #mojo
05:46 inokenty-w joined #mojo
05:56 disputin joined #mojo
06:12 disputin joined #mojo
06:43 disputin joined #mojo
06:44 kiwiroy joined #mojo
06:51 CW joined #mojo
07:11 Kharec joined #mojo
07:12 dod joined #mojo
07:19 dod joined #mojo
07:51 anparker1 joined #mojo
07:52 AndrewIsh joined #mojo
08:06 kiwiroy joined #mojo
08:19 trone joined #mojo
08:45 sugar joined #mojo
08:51 rshadow joined #mojo
08:51 dotan_convos joined #mojo
09:00 rshadow joined #mojo
09:16 salva joined #mojo
09:31 stryx` joined #mojo
10:04 dod joined #mojo
10:12 Petru joined #mojo
11:01 sugar joined #mojo
11:07 tchaves joined #mojo
11:08 kiwiroy joined #mojo
11:17 tchaves joined #mojo
11:56 stryx` joined #mojo
12:06 Ralesk joined #mojo
12:08 kiwiroy joined #mojo
12:17 Petru joined #mojo
12:23 sugar joined #mojo
12:26 gregf_ joined #mojo
12:49 kes joined #mojo
12:51 kes sometimes server is configured for different header than: X-Forwarded-For, May you please do not hardcode it https://metacpan.org/source/SRI/Mojolicious-7.21/lib/Mojo/Transaction.pm#L39 and allow to configure?
12:59 irqq joined #mojo
13:16 mishanti1 kes: You can always handle other headers yourself using hooks, so it is easy to do what you need that way.
13:17 kes I prefer to use that something exists already and do not reinvent wheel
13:18 mishanti1 kes: That would be the X-Forwarded-For header :)
13:39 sri lol
14:05 sugar joined #mojo
14:05 mtths joined #mojo
14:08 kiwiroy joined #mojo
14:15 Petru joined #mojo
14:16 Pyritic joined #mojo
14:25 foursixnine joined #mojo
14:37 sugar joined #mojo
14:40 dikim joined #mojo
14:41 gryphon joined #mojo
14:57 marty_ joined #mojo
14:58 Petru joined #mojo
15:09 DavidSouza joined #mojo
15:11 Ryoga joined #mojo
15:14 kes mishanti1: then I should force the admin to configure http server to use this header. Is there RFC about that this header should be used?
15:19 mishanti1 kes: I am not an expert on this, but I do not believe it has been standardized. I am under the impression that XFF has been the de-facto standard in most proxies.
15:20 geheimnis` joined #mojo
15:20 mishanti1 There is also RFC 7239 that mentions 'Forwarded: ...' as a header, though with another format.
15:21 pink_mist https://tools.ietf.org/html/rfc7239 <-- proposed standard, not a standard yet
15:21 kes ok. thank you. I will look at
15:21 * sri kinda wants File::Path::remove_tree in Mojo::File, but it doesn't look like we have a use for it in core :(
15:22 kes I have found this helper: https://metacpan.org/pod/Mojolicious::Plugin::RemoteAddr but it also use different header names and are also hardcoded.
15:24 kes Under wheel I mean: I will prefer to use clear mojo if it will allow to configure headers or module like Mojolicious::Plugin::RemoteAddr, but because of both have hardcoded headers I am forced to reinvent wheel and create same code with just different header (((
15:24 sri nobody is forcing you
15:25 jberger ther
15:25 jberger there is also https://metacpan.org/pod/Mojolicious::Plugin::RequestBase
15:25 jberger it has yet another header
15:25 jberger pick the header you like best?
15:26 sri $req->headers->header('X-Forwarded-For' => $req->headers->header('My-Super-Custom-Unicorn-Forwarded-Header'))
15:27 sri that's so hard
15:28 kes in every place of use...
15:28 kes instead of: config( header => 'My-Header' ); $c->tx->remote_address
15:30 geheimnis` joined #mojo
15:31 jberger kes: it is documented here: http://mojolicious.org/perldoc/Mojolicious/Guides/Cookbook#Hypnotoad
15:31 jberger there are several easy work arounds
15:31 jberger I think it makes sense to stick with the common standards
15:33 mishanti1 kes: You can do what you propose, sort of. In a hook you can do what sri wrote, the everything else (like logging-plugins etc) that looks for XFF will find it. Then you can use XFF internally until upstream changes to a more common header.
15:50 geheimnis` joined #mojo
15:53 sugar joined #mojo
16:08 kiwiroy joined #mojo
16:18 perlpilot joined #mojo
16:31 disputin joined #mojo
16:40 howitdo joined #mojo
16:58 disputin joined #mojo
17:10 kiwiroy joined #mojo
17:23 dod joined #mojo
17:29 Petru joined #mojo
17:41 beyondcreed joined #mojo
17:42 sugar joined #mojo
17:45 sh14 joined #mojo
17:45 good_news_everyon joined #mojo
17:45 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vMNqX
17:45 good_news_everyon mojo/master b2c9b1a Sebastian Riedel: link to the blog example as well
17:45 good_news_everyon left #mojo
18:12 PryMar56 joined #mojo
18:18 good_news_everyon joined #mojo
18:18 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vMNGn
18:18 good_news_everyon mojo/master 6c890c2 Sebastian Riedel: mention a little more detail
18:18 good_news_everyon left #mojo
18:20 dod joined #mojo
18:28 sugar joined #mojo
18:38 dod joined #mojo
18:39 good_news_everyon joined #mojo
18:39 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vMNWC
18:39 good_news_everyon mojo/master 50863d8 Sebastian Riedel: just mention example tasks as well
18:39 good_news_everyon left #mojo
18:39 sri just hope that's not TLDR http://mojolicious.org/perldoc#SPIN-OFFS
18:43 good_news_everyon joined #mojo
18:43 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vMNWd
18:43 good_news_everyon mojo/master 1ce7d36 Sebastian Riedel: this sounds a little better
18:43 good_news_everyon left #mojo
18:44 Pyritic joined #mojo
18:44 sri btw. i'm warming up to the idea of having more formats for the built-in exception pages
18:45 sri you know, json error for REST clients
18:45 sri if anyone feels like working on a proposal
18:49 mishanti1 Json-format for REST-clients is pretty useful. We've hacked together our own stuff for that. Would probably give people here nightmares. :p
18:49 sri i noticed many people hack that up themselves
18:50 Grinnz yea, i have a hook that returns the stacktrace + the error message in JSON on development
18:50 Grinnz but only for the JSON endpoints
18:50 Grinnz so i have a hard time seeing how to make that generic
18:51 sri respond_to semantics i imagine
18:51 sri Accept: application/json
18:51 purl it has been said that application/json is approved IANA media type # http://www.iana.org/assignments/media-types/application/
18:51 * sri pats purl
18:51 * purl pats sri back, harder
18:51 Grinnz it's based on path though
18:52 Grinnz for this one
18:52 mishanti1 Our need is specificly one of not exposing any internal details to the outside world. So we render a generic error message on errors, and 40x etc when the client did a boo-boo.
18:52 Grinnz yes, on production
18:52 mishanti1 Yes
18:56 good_news_everyon joined #mojo
18:56 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vMN8b
18:56 good_news_everyon mojo/master f629862 Sebastian Riedel: more links
18:56 good_news_everyon left #mojo
18:56 Grinnz links... links everywhere
18:58 sri http://s2.quickmeme.com/img/05/05eb98341f66f41808a972753fe3a31f9a9362d97f0fc4f96422beacc3ea0492.jpg
19:12 sri interesting, next minor macos release will have f.lux built-in Oo
19:14 genio never used it
19:15 sri it gave me a healthy sleep schedule
19:15 stephan48 sri: the actual application or just the feature but reimplemented by apple?
19:15 sri the feature of course
19:16 sri ios already has a night mode
19:16 stephan48 that makes some sense, as yea android(atleast the former cyanogenmod) and as far as i heard ios...
19:16 stephan48 yea
19:16 stephan48 already have such modes. i was just suprised if apple actually would use an external app for that :D
19:33 jberger wait, the robots need sleep now?!
19:33 jberger :o
19:37 dod joined #mojo
19:50 mishanti1 Speaking of. While waiting in line at the pharmacy I observed that the cashiers implementation of serving customers was somthing like `sub serve( $me ) { $me->do_work(); $me->blank_stare(); sleep 150; $me->refocus(); $me->do_rest_of_work(); }`
19:52 mishanti1 Actually, a very good demonstration for students to learn the concepts behing non-blocking VS blocking would be to go there and compare the cashiers in the pharmacy and the cashiers in the kiosk next door. Worlds apart.
19:52 mishanti1 s/behing/behind/
19:53 mishanti1 </rant>
20:10 jberger Having worked for a small pharmacy i have one guess
20:10 jberger The insurance interop systems are REALLY  ancient and slow
20:16 Grinnz yeah. they might just be sitting at the terminal doing nothing, but they might be waiting
20:23 kiwiroy joined #mojo
20:29 mishanti1 You are probably correct in that it is a system slowing them down, but I'm placing my bet on it being a newer system being slow, not an old one.
20:30 jberger well that is always possible
20:30 jberger because at that level, everything has to be bug-for-bug compatible (see also banking software)
20:31 mishanti1 At least that is what I have seen in POS-systems for retail here in Norway. The newer the system the more of a steaming pile of rubbish it is.
20:32 lluad Well-tuned, battle-tested mainframe code vs some PoS POS running on XP.
20:34 mishanti1 lluad: Especially since a lot of them seem to be running pirated XP factory-installed right out of Hoo-Long Ding-Dong China.
20:53 kiwiroy joined #mojo
21:29 sri any more opinions on this? https://github.com/kraih/mojo/issues/1042
21:30 sri it's fine if some of you would like to extend the scope of Mojo::File a little
21:30 pink_mist well I think mst's suggestion seems very useful
21:30 sri is it? when would you use it?
21:31 pink_mist when I had lines stored in an array and wanted to write them all to a file?
21:32 sri does that happen a lot?
21:33 pink_mist not very often, but it does happen
21:33 pink_mist there's also his own example of readline in list context
21:38 sri this one i assume nobody really cares about https://github.com/kraih/mojo/issues/1040
21:39 sri the two people still using 5.10 and 5.12 ;p
21:39 Grinnz a lot more than that, but as long as it's just an errant warning
21:40 Grinnz at most i'd add a defined() check on the remove
21:40 jberger I have no interest in adding an XS dep or the other thing that might cause segfaults
21:41 Grinnz i think making Mojo::File useful for general file operations rather than only the things core needs would be useful for application development
21:42 sri so, we make it dependent on votes?
21:42 haarg jberger: Devel::GlobalDestruction does not require XS
21:43 Petru joined #mojo
21:43 jberger haarg: really?
21:43 jberger uses B:: then I guess
21:43 haarg yes
21:43 haarg my comment regarding Devel::GlobalPhase was mainly for mst, as a workaround in case mojo doesn't make any chages
21:43 Grinnz there is separately https://metacpan.org/pod/Devel::GlobalDestruction::XS
21:43 Grinnz but thats not required
21:44 haarg if nothing in mojo changes, Devel::GlobalPhase (once i make some changes) could serve as a viable workaround for people using older versions of perl
21:44 sri "use Devel::GlobalPhase -var;" is neat
21:45 jberger I suppose we could probably make Mojo::Util::_global_destruction include something like https://metacpan.org/source/HAARG/Devel-GlobalDestruction-0.14/lib/Devel/GlobalDestruction.pm#L33-35
21:45 haarg yes, that would be the proposal
21:46 jberger personally I'm 50/50 on it, but that's better than I had been when I thought there was XS involved
21:46 sri jberger: what are your thoughts on the Mojo::File scope?
21:46 jberger I've always been on the side of Mojo:: being a toolkit
21:46 sri ok, then Mojo::File features will depend on vote
21:47 pink_mist alright, I'll go vote +1 on mst's post then
21:47 jberger cautiously I'd say that I don't want to see huge feature creep, but I don't mind adding features that core doesn't need if they are broadly useful
21:49 sri yea
22:01 jberger sri / others: is there an example of making a ua call with a proxy only for a single request?
22:02 sri good question, not sure if that's a tested case yet
22:03 sri you'd make your own transaction and add the proxy setting to it
22:03 jberger on the request object it looks like
22:03 jberger and there are two relevant attributes, proxy and via_proxy
22:03 jberger I'm assuming I need both?
22:15 irqq_ joined #mojo
22:40 hernan604 was there a $clients hash that listed websocket clients ?
22:41 Grinnz pink_mist++ # getting me to try vmware again
22:43 Grinnz it's not perfect, but it works, and virtualbox is becoming unusable.. the last few fedora releases throw constant SELinux errors, cinnamon has stopped rendering properly, etc
22:43 Grinnz none of those problems in vmware
22:43 pink_mist glad it worked out for you =)
22:51 sugar joined #mojo
22:51 jberger hernan604 what do you mean?
22:55 hernan604 jberger: i have an old project here that uses a global $client hash ie: sub ws { ... $clients->{ $id } } ... so i wondered it was a var declared by mojo
22:55 hernan604 but probably not
22:55 jberger mojolicious definitely isn't declaring a global in your code, no :-P
22:55 hernan604 :)
22:55 pink_mist 'my' declares variables ... 'Mojo' isn't 'my' :P
22:55 jberger oh, it can be done :-P
22:56 jberger but we don't
22:56 pink_mist not sanely :P
22:56 jberger sure it can
22:56 jberger https://metacpan.org/pod/Tie::Select
22:56 pink_mist oh I suppose you can force export variables actually
22:56 pink_mist (though not 'my' variables)
23:08 jberger oh,yeah, injecting lexical would be rather insane
23:08 jberger if possible at all
23:15 pink_mist I recently saw someone demonstrate how to get back a real ref from a "0x98249284" stringified reference, so I'm thinking it might be possible
23:15 pink_mist haarg++ # it still haunts me
23:16 jberger done that before too :D
23:18 haarg pink_mist: i was hesitant to post that. i don't want anyone else to actually try it.
23:21 jberger I said similar things after posting Class::Anonymous
23:23 Petru joined #mojo
23:34 gizmomathboy joined #mojo
23:41 tchaves joined #mojo

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