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

IRC log for #mojo, 2016-04-11

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

All times shown according to UTC.

Time Nick Message
00:28 yakudzo joined #mojo
01:03 woz joined #mojo
01:20 eltrump joined #mojo
01:41 ramortegui joined #mojo
02:09 ramortegui joined #mojo
02:15 noganex_ joined #mojo
03:05 woz joined #mojo
03:31 disputin joined #mojo
03:44 s1037989 bpmedley: I'm around if you want my help testing anything.
03:45 melo1 joined #mojo
04:26 umask001 joined #mojo
04:30 ivi joined #mojo
04:34 bpmedley s1037989: would be marvelous if you could test this one: https://github.com/brianmed/ForkAndGo/blob/master/lib/Mojolicious/Plugin/ForkCart.pm
04:37 s1037989 bpmedley: Mojo::Reactor::Poll: I/O watcher failed: Can't call method "on" on an undefined value at eg/sample_app.pl line 81.
04:38 s1037989 Started with: $ env PERL5LIB=lib:$PERL5LIB perl eg/sample_app.pl daemon -l http://*:8080
04:42 bpmedley s1037989: Would you try adding MOJO_REACTOR=Mojo::Reactor::Poll to the env mix..
04:43 s1037989 No errors on start up.  Loaded the page and got Hello:8 and 5 seconds later Joy showed in my log.
04:43 bpmedley Sweet!
04:44 bpmedley s1037989: Do you think this might be useful for your startup issue?
04:44 s1037989 What have you improved since we last spoke a weekish ago?  And what is meant by "ForkCart"?
04:45 disputin joined #mojo
04:45 bpmedley s1037989: Bug fixes and a more organized approach.  Also, the ability for the children and manger to watch themselves, and if one dies they go away.
04:45 s1037989 Not sure yet...  What are the forked calls at 11 and 48?
04:46 s1037989 Awesome!  That sounds like great progress!
04:46 bpmedley ForkCart is a "golf" cart of forked children..
04:46 bpmedley And the $caddy is responsible for watching over the children.
04:47 s1037989 Uh huh...  interesting metaphor!  :D
04:47 bpmedley Also, "forked" is a helper that allows setting up the arbitrary code that will execute on startup.
04:48 bpmedley Also, there is initial support for plugins: plugin ForkCart => { process => ["minion"] };  <-- For starting one minion with default arguments.
04:49 s1037989 Ah ha!  There we go.  :)  That would remove the blocks at 11 and 48?
04:49 bpmedley https://github.com/brianmed/ForkAndGo/blob/master/eg/minion.pl <-- s1037989
04:49 s1037989 Ah ha!  I'll run that.
04:51 s1037989 Ok, looking good.  I ran prefork with default 4 processes.  One of those processes has 3 minions.
04:51 bpmedley Nice
04:51 s1037989 I'll kill the prefork process that holds the 3 minions.
04:52 s1037989 Interesting.  I kill the parent of the minions, the first of the children of the original prefork.  And everything died, all preforks and minions.
04:53 bpmedley s1037989: That's almost certainly a bug of mine.  I'll have to defer fixing for a few days.  It's really late here.
04:55 s1037989 In the log, at startup, I get 7 workers and the last 3 of the workers also report "Checking worker registry and job queue" which leads me to believe that the minions put the same Worker message in the log.  Might be nice to be able to distinguish those ForkCalls separately from the app processes?  (If any of that makes sense...)
04:56 s1037989 Mmm...  I see, the first prefork that is the parent to the minions actually first forks 3 prefork processes and then each of those fork a minion.
04:56 bpmedley s1037989: I hear you.. https://github.com/brianmed/ForkAndGo/blob/master/lib/Mojolicious/Plugin/ForkCart.pm#L90 needs a better log msg.
04:56 bpmedley Not sure if my number of forks is optimal.
04:57 s1037989 It seems high, but I don't know what is minimally necessary.  At the least, it's clean (the process table) and that's important (IMO)!
04:58 s1037989 So, the minion.pl looks like an awesome minimalist example for a high usage.  Still needs more extensive testing, but the application looks really solid and terrific from an app dev POV.
04:59 bpmedley Spanx.. :)  I need sleep..
04:59 s1037989 sample_app.pl looks nice and simple as well from the perspective of wanting to do something a little more custom.  The two forked blocks are simple code that are necessary for custom.
05:00 s1037989 That you're going to support the notion of plugins (if I understand you correctly) is even better!
05:00 s1037989 Yep, good work!  Have a good night.  I hope to be on more this week.
05:06 inokenty-w joined #mojo
05:07 woz joined #mojo
05:42 stmovi joined #mojo
05:46 stmovi Hello. Anyone using Minion... I try to start workers on different queues, using the -q or --queue options, but it doesn't seem to take the option into account. Is there a bug on that or I do something wrong? ;)
05:51 trone joined #mojo
06:32 stmovi joined #mojo
06:42 stmovi joined #mojo
06:52 McA joined #mojo
06:53 Atog joined #mojo
06:58 kes joined #mojo
07:00 woz joined #mojo
07:01 berov joined #mojo
07:10 Jonis joined #mojo
07:22 AndrewIsh joined #mojo
07:24 dod joined #mojo
07:28 bpmedley stmovi: Do you have a small example script (possibly a Lite app)?
07:28 dod joined #mojo
07:28 bpmedley stmovi: Also, what version of Minion / Mojolcious do you have?
07:34 trone joined #mojo
07:38 stmovi Hey, I just found the problem... it works with: Mojolicious::Commands->start_app('CDTSync', 'minion', 'worker', '--queue', 'sync');
07:39 stmovi but not if I use just the -q ;)
07:41 stmovi Thanks anyway!
07:48 woz joined #mojo
07:50 disputin joined #mojo
07:51 disputin joined #mojo
08:01 osfabibisi joined #mojo
08:02 stmovi joined #mojo
08:20 Vandal joined #mojo
08:25 stephan48 joined #mojo
08:34 arthas joined #mojo
08:35 woz joined #mojo
08:52 woz joined #mojo
08:57 osfabibisi joined #mojo
09:03 disputin joined #mojo
09:25 woz joined #mojo
09:26 denny joined #mojo
09:40 and joined #mojo
09:41 avkhozov joined #mojo
09:58 avkhozov Hi! What is the best practices to handle bool datatype in Mojo::Pg in context of the JSON REST API? DBD::Pg returned 0 and 1 for false and true values and it looks bad in JSON responses.
09:59 avkhozov I think Mojo::Pg::Results should expand its to Mojo::JSON or JSON::PP boolean values. Or this has bad consequences?
10:04 disputin joined #mojo
10:05 woz joined #mojo
10:07 kaare joined #mojo
10:34 batman avkhozov: can't help you with this, but maybe you like to know that Mojo::JSON share booleans with JSON::PP
10:35 Kripton joined #mojo
10:38 avkhozov batman: Indeed, thanks
10:39 ashimema in dbic I use filters to map tiny ints (as returned by PG) to JSON::PP bools
10:39 ashimema perhaps similar can be done with the responses via Mojo::Pg
10:39 ashimema that's entirely conjecture on my part btw..
10:39 ashimema I've not played enough with Mojo::Pg yet
10:49 avkhozov Yes, I can scan an row and replace columns of type bool with JSON::PP::Boolean object.
10:50 avkhozov But Mojo::Pg already has expand method to convert json and jsonb types to perl hash.
10:51 avkhozov But with bool types is not so unambiguously. 0 and 1 is valid bools in Perl.
10:56 dvinciguerra_ joined #mojo
10:58 tchaves joined #mojo
11:05 disputin joined #mojo
11:11 tchaves joined #mojo
11:16 avkhozov By the way, pg float numbers stored as strings in perl scalars, instead of numbers. http://pastie.org/pastes/10793334/text
11:27 osfabibisi joined #mojo
11:36 melo joined #mojo
11:48 asarch joined #mojo
11:52 cpan_mojo JSON-Validator-0.67 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/JSON-Validator-0.67
12:02 melo joined #mojo
12:05 disputin joined #mojo
12:09 woz joined #mojo
12:15 woz joined #mojo
12:37 dvinciguerra joined #mojo
12:49 Foxcool joined #mojo
12:59 ramortegui joined #mojo
13:08 disputin joined #mojo
13:11 woz joined #mojo
13:16 woz joined #mojo
13:35 woz joined #mojo
13:37 osfabibisi joined #mojo
13:40 SmokeMachine bpmedley: Hi! I have a question for you... on Mojo::IOLoop::Tail how can I stop tailing?
13:41 woz joined #mojo
13:42 bpmedley SmokeMachine: I'll have to look later.  I'm on mobile.
13:42 SmokeMachine :) thanks...
13:45 woz joined #mojo
13:46 SmokeMachine bpmedley: now, I'm $tail->ioloop->reactor->remove($file)... but it doesn't look right to me...
13:48 Foxcool joined #mojo
13:48 bwf joined #mojo
13:51 melo joined #mojo
13:55 mcsnolte joined #mojo
13:56 osfabibisi joined #mojo
14:03 sri avkhozov: expanding booleans might be nice, but i don't see a good way to add it to the code
14:03 sri it's not worth making the expand code more complicated, that's for sure
14:04 jberger sri / avkhozov: funny, I was just thinking about that the other day
14:05 sri so, if someone would send a patch that ends up making expand look more elegant, i guess i'd accept it
14:06 jberger I'll try to take a look soonish
14:14 orev joined #mojo
14:19 disputin joined #mojo
14:24 avkhozov Nice. I also try to write pull request.
14:25 avkhozov And what about numbers? http://pastie.org/pastes/10793334/text Pull request are welcome?
14:28 disputin joined #mojo
14:35 sri hmm, if that's where things are heading i'm not so sure anymore
14:35 sri i don't want to add features that should really be in DBD::Pg
14:35 sri changing my stance to slightly -1 on all new expand features
14:36 sri that means bool support has to make the code cleaner, or i'm -1
14:37 cpan_mojo Mojolicious-Plugin-OAuth2-Server-0.24 by LEEJO https://metacpan.org/release/LEEJO/Mojolicious-Plugin-OAuth2-Server-0.24
14:39 Lee ashimema: i have split out the bulk of the OAuth2 logic into https://metacpan.org/release/LEEJO/Net-OAuth2-AuthorizationServer-0.01 # should make it easier to add other grant flows to the plugin if needs be
14:40 osfabibisi joined #mojo
14:40 zz_Foxcool joined #mojo
14:41 woz joined #mojo
14:43 sri i guess it should also roundtrip $db->query('select ?::bool', JSON::PP::true)->expand->hash
14:44 sri but yea, please don't expect a patch to be merged without a long discussion
14:45 sri the idea of number handling freaked me out
14:48 jberger I don't really like the idea of munging number types
14:48 jberger But as perl doesn't have a boolean type i don't think that's the same discussion
14:49 jberger Especially given the types available to json
14:51 jberger I have an increasing amount of code that selects data in a model then basically puts that data out to an api client as json
14:51 jberger With swagger for output validation
14:53 jberger On a related note pg's json_object_agg is really nice
15:01 PryMar56 joined #mojo
15:03 zz_Foxcool joined #mojo
15:05 avkhozov json_agg + json_build_object are also nice
15:05 avkhozov I will create an issue about number mangling in DBD::Pg
15:15 woz joined #mojo
15:18 disputin joined #mojo
15:21 lluad joined #mojo
15:30 zz_Foxcool joined #mojo
15:31 batman BinGOs: not a great success changing to "#!/usr/bin/env perl" :( http://www.cpantesters.org/cpan/report/552d2f9e-fea6-11e5-829f-59a5aef69d38
15:31 batman got any other ideas?
15:32 jberger batman: I thought released code shouldn't use env perl
15:32 jberger s/released/CPAN/
15:32 batman jberger: it's scripts in t/bin
15:33 melo joined #mojo
15:33 jberger you can't use $^X?
15:33 batman jberger: i could rewrite the scripts before executing...
15:34 jberger oh I see what you mean
15:34 jberger where do they get executed?
15:34 BinGOs I'm at $work at the moment, when I get $home I shall spin up a VM and see if I can manually test.
15:34 batman BinGOs: thank you :)
15:34 batman jberger: from M::P::CGI.pm
15:35 batman s/.pm//
15:35 batman jberger: i can't do ($^X => $script) since M::P::CGI doesn't know if it's a perl script
15:35 jberger can you cheat and do it here: https://metacpan.org/source/JHTHORSEN/Mojolicious-Plugin-CGI-0.29/t/not-modified.t#L10
15:36 jberger "$^X /path/tofile"
15:36 BinGOs it does appear to be that one FreeBSD box now rather than the NetBSD box too.
15:36 batman BinGOs: yeah... but we'll have a look later. no rush :)
15:37 batman jberger: no, because then the SCRIPT_NAME gets turned into /path/to/perl
15:37 batman not sure how to solve that
15:37 BinGOs I think I only looked at the NetBSD failure report previously.
15:37 batman BinGOs: i will try to generate the scripts.
15:40 meshl joined #mojo
15:45 woz joined #mojo
15:47 woz joined #mojo
15:50 ivi joined #mojo
15:51 woz joined #mojo
16:08 cpan_mojo Mojolicious-Plugin-CGI-0.30 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/Mojolicious-Plugin-CGI-0.30
16:08 * batman puts money on even more failing tests tomorrow :P ^^^
16:08 batman BinGOs: you don't have to do anything... maybe using $^X will fix it
16:11 jberger batman: did you mean to change the test files to have "#!/Users/jhthorsen/.perl/perl-5.22.1/bin/perl" ?
16:11 batman jberger: no.
16:12 batman i forgot to add them to MANIFEST.SKIP :(
16:12 jberger I'm looking over the release diff: https://metacpan.org/diff/file?target=JHTHORSEN%2FMojolicious-Plugin-CGI-0.30%2F&amp;source=JHTHORSEN%2FMojolicious-Plugin-CGI-0.29%2F
16:12 jberger ah
16:12 batman i only added them to .gitignore :P
16:13 batman but it doesn't matter, since the files will be created even if they exist
16:13 jberger sure
16:13 batman just looks stupid :P
16:13 batman hehe
16:19 cpan_mojo Mojolicious-Plugin-CGI-0.31 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/Mojolicious-Plugin-CGI-0.31
16:20 batman thanks jberger
16:21 jberger much better https://metacpan.org/diff/file?target=JHTHORSEN%2FMojolicious-Plugin-CGI-0.31%2F&amp;source=JHTHORSEN%2FMojolicious-Plugin-CGI-0.29%2F
16:21 jberger batman++
16:22 woz joined #mojo
16:39 osfabibisi joined #mojo
16:59 trone joined #mojo
17:02 dod joined #mojo
17:14 woz joined #mojo
17:23 meshl joined #mojo
17:45 woz joined #mojo
17:52 human39 joined #mojo
17:56 marty joined #mojo
17:56 marty joined #mojo
17:58 woz joined #mojo
18:24 woz joined #mojo
18:45 disputin joined #mojo
18:46 disputin joined #mojo
18:48 disputin joined #mojo
18:54 disputin joined #mojo
18:54 disputin joined #mojo
18:55 disputin joined #mojo
19:13 woz joined #mojo
19:15 thowe_work joined #mojo
19:25 sri oh wow, i had no idea what a horrible company maxmind is
19:26 sri "lets just default to random peoples homes, what could possibly go wrong?" http://fusion.net/story/287592/internet-mapping-glitch-kansas-farm/
19:27 vicash i am surprised they didn't pick somewhere in the middle of the ocean as a generic "not found" location.
19:28 sri or their own headquaters
19:28 vicash LOL
19:29 sri oalders: what you're doing there is messed up
19:29 oalders sri: it's being addressed as we speak
19:35 perlpilot vicash: that's exactly what I thought.
19:37 Grinnz_ "
19:38 Grinnz_ "They are picking new default locations for the U.S. and Ashburn, Virginia that are in the middle of bodies of water, rather than people’s homes." - now the crab people will suffer!
19:38 sri so, that's only two locations?
19:39 Grinnz_ FWIW in my only use of the Maxmind database, I never have tried to map the lat/long to physical addresses... it never occurred to me people would even think that rather than going through the ISP
19:39 mspo joined #mojo
19:40 jberger oh is that the problem with the house in florida too, the one that shows as where all the stolen cell phone goes?
19:40 mspo hey I just used mojo::lite to make a slack slash command in like five minutes and 50 lines of code :)
19:41 sri oalders: the article makes it sound like you're just sweeping it under the rug
19:41 perlpilot vicash: I figure they wanted to convey more information than "I dunno" so they chose the centerish of whatever area they did know.
19:41 oalders sri: no, it's being taken very seriously
19:42 vicash perlpilot: it makes sense that they may have assumed that the location might have been in the USA, so display the center to show the complete map visually.
19:42 perlpilot oalders: Just map the unknown locations to federal buildings ;)
19:43 vicash absolutely. map it all to the White House
19:43 vicash or even better the FBI building. this way FBI can knock on their own doors
19:43 meredith i'm surprised the database is only offering points, rather than point+radius, or an area, for something as fuzzy as "somewhere in the US"
19:43 Grinnz_ meredith: it is offering an area. the lat/long is just additional metadata
19:43 perlpilot meredith: even if it were, people would misuse that information
19:44 Grinnz_ normally, you would use the country/city/subdivisions, etc
19:44 meredith also if they do change these defaults, how many places aren't keeping local databases up to date :\
19:44 Grinnz_ https://metacpan.org/pod/GeoIP2::Record::City is what I work with usually
19:46 sri wonder how big the problem is outside the us
19:47 perlpilot That article did highlight something else quite scary though ... law enforcement is using IP mapping to obtain warrants.  That's so crazy.
19:47 jberger ^^ that
19:47 sri indeed
19:47 Grinnz_ yeah. they should only be going through the ISP
19:47 jberger they should point all the unknown IPs at the NSA headquaters
19:48 vicash aren't all the IPs "known" ?
19:48 Grinnz_ not publically
19:49 vicash but there has to be a provider somewhere in the chain ? like say Level3 who hide the .gov/.mil space for example
19:49 Grinnz_ there should be yes, I'm sure there's some ISPs which don't adequately record where their IPs are used though
19:49 sri wonder how many got killed already because of a bad default location in nigeria
19:50 jberger :o
19:50 vicash another possibility is that they might have gotten sent some money instead
19:51 perlpilot you know those videos of missles taking out trucks and buildings during the Gulf War?  What if they used IP mapping to target the buildings?
19:51 vicash in the 90s ?
19:51 perlpilot aye, I mean what if that happened today
19:51 vicash they're probably using cell tower mapping today
19:51 Grinnz_ they have a lot more sophisticated surveillance techniques than relying on public IP mapping
19:51 vicash Stingrays
19:51 perlpilot Who is this "they"?  Because people are stupid last time I checked.
19:52 Grinnz_ the military
19:52 vicash they're probably still using public IP mapping to find Tor exit nodes
19:53 sri most law enforcement and military is not properly trained, and will misuse whatever there is to misuse
19:54 vicash push-button style investigation
19:55 oalders perlpilot: :) the precision insights web service includes an "accuracy radius" for the lon/lat https://www.maxmind.com/en/geoip2-precision-insights
19:55 woz joined #mojo
20:09 perlpilot I wonder how law enforcement agencies actually come to get the lat/long for an IP address?  I mean, could companies such as MaxMind put some sort of disclaimer just for them?   "If you're looking at information for law enforcement purposes, please be aware that this data does not indicate a single point, but an area.  The IP address you're searching for could be anywhere within 5 miles of this location"  (or something)
20:09 perlpilot (yes, I dream a little bit)
20:10 perlpilot oalders: It wouldn't hurt for MaxMind to find a way to add such a disclaimer anyway  :)
20:10 sri maxmind is being incredibly irresponsible with the data
20:11 oalders perlpilot: yeah, i expect some website copy will be updated
20:12 sri law enforcement is just an annoyance, but i bet some mexican cartel has already used the data to locate bloggers they don't like
20:13 jberger that is a problem with more than just MaxMind
20:13 jberger anything that does any sort of geolocation can be misused in that way
20:14 sri but maxmind capitalizes on it, overselling the data, implying accuracy that's just not there
20:14 sri it's evil imo
20:15 perlpilot you know, in mapping apps I've been involved with in the past, we used 0,0 for the "I dunno" location of things.  I used to think that was pretty annoying.  "When I have an app specifically for Texas, it should show me somewhere in Texas!"  Now I'm thinking it was probably a pretty good choice.  :-)
20:16 thowe_work It appears that when a route function ends, the default behavior is to render...
20:17 thowe_work But if I call render before it would have returned, does it return there or keep going to the end?  In other words, if I want to render and end there do I need to return after the render?
20:18 thowe_work I can't seem to find someplace in the controller doc, or specifically under render, that says.  So I'm guessing it should be something I just know to infer in some way(?)
20:19 perlpilot thowe_work: Did you read Mojolicious::Guides::Rendering ?
20:19 mspo left #mojo
20:20 bpmedley thowe_work: I usually "return $c->render(text => 'Joy')"
20:20 Grinnz_ thowe_work: yes, render will not return itself (that would be overly magical)
20:21 thowe_work but it seems like saying $self->render; return;  is kind of silly since just saying return; will make me render.
20:21 Grinnz_ if you're relying on autorendering sure, the render call isn't needed
20:22 thowe_work but calling render directly, and only using return when one wants to return 1;  or return undef;  seems to be the practice if one goes by the examples int he docs.
20:22 Grinnz_ return value is ignored unless it's an under route
20:24 thowe_work huh, OK..
20:24 woz joined #mojo
20:24 oalders sri: accuracy levels are pretty clearly posted https://www.maxmind.com/en/geoip2-city-database-accuracy?country=&amp;resolution=city
20:25 disputin joined #mojo
20:25 PryMar56 joined #mojo
20:26 ribasushi perlpilot: speaking of 0.0 : https://goo.gl/maps/EVtLN6Y8GJm
20:28 melo joined #mojo
20:28 sri oalders: obviously not clear enough for all kinds of law enforcement agencies
20:29 sri not that i got it looking through the site
20:31 sri "These values are not precise and should not be used to identify a particular address or household."
20:31 sri i think what you really mean is "cannot be used"
20:33 * sri is still not sure what exactly those coordinates are supposed to mean
20:37 sri "99.8% accurate on a country level, 90% accurate on a state level, 81% accurate on a city level for the US within a 50 kilometer radius."
20:37 sri that's what your feature list says
20:37 sri that's so misleading
20:38 sri what you really mean is "mostly we can tell you the correct city"
20:46 sri anyway, that's what i mean when i say you're overselling your data, and it's obviously misused a lot
20:47 woz joined #mojo
20:48 sri for the record, i still don't know if those coordinates are supposed to always just point at a random location in a city or a household
20:52 bjakubski joined #mojo
20:53 s1037989 Can I ask what the catalyst for the maxmind sri / oalders discussion is?  Is oalders the author of an article about Maxmind, a module that uses Maxmind, or involved with the Maxmind company directly?
20:54 sri oalders works for maxmind
20:54 sri and i stumbled over that article in my twitter feed
20:54 sri brought it up because so far maxmind was regarded as one of the nicer companies in the perl community
20:56 jberger which is why I'd hope they will address this promptly and in the meantime, I'm sure maxmind is taking enough beating without us piling on
20:57 preaction how could they? they can't just randomize it, as people will take it seriously. people need to vent their internet rage, and they're not going to care that there's no real way to map an IP to a physical location
20:57 sri i only went on about it because oalders said it was clear on the site, and it just wasn't to me
20:58 jberger preaction: could be like the old GPS signal which was literally randomized within a certain range
20:58 preaction right, but even that has rules. this doesn't have even that degree of certainty
20:59 preaction country is about all you can rely on, period. because IANA is a thing
20:59 lluad Maxmind are reasonably honest about their data. Their users, OTOH, love maps with points on them and don't really care about accuracy.
20:59 sri somehow i doubt they are going to change all the coordinates in the database to point to bodies of water
20:59 preaction but average humans don't know that and will never understand it
21:00 preaction even if they put them in the middle of a lake, the user will just say "oh, must be the closest house on the lake"
21:00 lluad You can't even rely on country. Or even continent, 100%.
21:00 lluad But it's "good enough" for most purposes.
21:00 sri lluad: good enough for warrants
21:00 preaction which is why internet vigilantism is terrible
21:01 jberger reminds me of this: http://fusion.net/story/214995/find-my-phone-apps-lead-to-wrong-home/
21:01 lluad I've issued quite a lot of subpoeanas based on IP address. I'd never dream of using Geolocation data for that.
21:01 preaction jberger: same author, i think
21:02 thowe_work joined #mojo
21:03 lluad LEO understanding of tech ranges between "very competent security engineer" to "believes the internet is basically magic". I'm not surprised the latter end have more faith in geolocation than is warranted.
21:09 coolo joined #mojo
21:13 cpan_mojo Mango-1.28 by ODC https://metacpan.org/release/ODC/Mango-1.28
21:26 disputin joined #mojo
21:33 disputin joined #mojo
21:42 Mattjes joined #mojo
21:42 Mattjes Hello
21:42 meshl joined #mojo
21:43 Mattjes Can anyone help me plz?
21:43 woz joined #mojo
21:43 Mattjes How can i find with mojo:Dom an empty/blank tag
21:43 Mattjes <li>    </li>
21:43 Mattjes only if tag is blank?
21:44 Mattjes if ($tag->content eq '') dont work
21:49 Grinnz_ if ($tag->content =~ /^\s*$/) perhaps
21:50 jberger or trim first: http://mojolicious.org/perldoc/Mojo/Util#trim
21:50 jberger but it comes to the same thing
21:50 Mattjes @Grinnz_: works perfect! Thank you ;-)
21:51 Mattjes jberger: Thank you too!
21:51 jberger Grinnz_'s method is better, one regex vs two substitutions and a comparison
21:52 Mattjes thank you very much!
21:53 Mattjes Mojo rocks ;-)
22:03 dvinciguerra_ joined #mojo
22:10 sri there's a new css4 selector in the works for that too
22:11 sri but we do not yet support :blank
22:11 sri :blank is :empty with whitespace handling
22:39 jberger Interesting
22:46 sri oh, sidekiq is getting ported to node.js, will be interesting to follow the progress https://github.com/mperham/sidekiq.js
23:29 disputin joined #mojo
23:29 woz joined #mojo
23:29 disputin joined #mojo
23:41 avkhozov joined #mojo
23:50 woz joined #mojo
23:56 Phil21 hmm, anyone but tempire using jellybeans vim colorscheme? :) weird issue where it's not coloring some stuff with new vimrc
23:56 Phil21 was rocking tempire's dotvim earlier, trying to simplify
23:56 Phil21 bugging the hell out of me :)
23:59 Phil21 ok, that is fucking weird
23:59 Phil21 :color jellybeans = half the syntax is not colored
23:59 Phil21 setting colorscheme in .vimrc = works fine
23:59 Phil21 o.O

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