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

IRC log for #mojo, 2017-01-03

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

All times shown according to UTC.

Time Nick Message
00:00 sri yes
00:27 genio suse++
00:36 sugar joined #mojo
00:44 asarch joined #mojo
00:52 genio jberger: I just realized I've completely forgotten about your requests wrt Bcrypt
00:52 genio do you want perms on that dist so you can do what you will with it?
00:57 jberger Hmmmm, no I'll file a PR
00:57 jberger Thanks for the nudge though
00:58 jberger This break has been so filled with family stuff I've basically forgotten everything I was working on/with
01:00 aborazmeh joined #mojo
01:38 sugar joined #mojo
01:56 karjala sri: tempdir from File::Temp doesn't rmdir the directory after the holding variable goes out of scope. Could you fix that, maybe with an optional option?
01:57 karjala it deletes the directory only when the program exits (which Mojo being a server doesn't do very often)
02:02 pink_mist if you pass DESTROY => 1 or something I thought it did
02:03 pink_mist oh wait, CLEANUP => 1
02:03 pink_mist oh you're right, that's for program exit
02:04 pink_mist I was thinking the File::Temp OO interface
02:06 karjala Maybe it's on purpose that way. Ok, it's not that bad to have to wait until process exit.
02:07 pink_mist no, I was saying that the File::Temp OO interface behaves like you wanted
02:07 karjala oh ok
02:09 karjala On this line, there's three shifts, but only two variables on the LHS: https://github.com/kraih/mojo/commit/f6451ce5ef4fda774ec129a6f58806a8c7431fcb#diff-23288abbf7581bb56bb0af7b60e03593R31
02:16 sugar joined #mojo
02:35 purl joined #mojo
03:13 noganex_ joined #mojo
04:12 stryx` joined #mojo
05:02 sh14 joined #mojo
05:04 dboehmer joined #mojo
05:48 kaare joined #mojo
05:53 tuna joined #mojo
05:58 kaare joined #mojo
06:08 kaare joined #mojo
07:00 dod joined #mojo
07:05 dod joined #mojo
07:12 batman thans sri and jberger :)
07:13 batman sri: i would say blocking ws support is the least exiting thing on your list to me...
07:14 batman *thanks
07:17 batman i can't remember where i placed my test setup for Mojo::Server::Morbo->new->modified_files
07:20 batman guess it shouldn't be too hard to set up? create a bunch of files in `templates/` and run something like `perl -MMojo::Server::Morbo -Mojo -E'$s=Mojo::Server::Morbo->new;n {$s->modified_files} 10000'` ?
07:21 batman what do you think sri?
07:55 Vandal joined #mojo
08:04 lluad joined #mojo
08:12 trone joined #mojo
08:27 Lee joined #mojo
08:27 tyldis joined #mojo
08:33 AndrewIsh joined #mojo
09:27 sri batman: think you tested with like 1000 files last time
09:27 sri should be interesting to see the difference
09:27 batman ok. but the test setup makes sense?
09:28 batman just need to tweak the numbers
09:29 sri looks ok
09:32 sugar joined #mojo
09:32 batman will test soon... just need to send an email first.
09:40 osfabibisi joined #mojo
09:43 stryx` joined #mojo
09:43 gregf_ joined #mojo
09:48 batman sri: https://ssl.thorsen.pm/paste/adb7fb965c2f
09:48 batman so it's a bit slower, but i'm not sure if it matters.
09:49 batman let me know if you want me to tweak the number of test runs/files/directories
09:51 sri the difference seems to be totally irrelevant
09:51 osfabibisi joined #mojo
09:51 sri perhaps test with an older perl too
09:52 sri for a File::Spec without XS
09:53 batman i don't have time to setup that right now :(
09:53 batman unless i can simply fetch an old version of File::Spec, instead of using the one shipped with perl...
10:59 tchaves joined #mojo
11:01 tchaves joined #mojo
12:18 batman anyone that has a trick to format this better with perltidy? https://ssl.thorsen.pm/paste/d7a33ee79c00
12:20 batman think i'll just create some helper methods to inject those methods...
12:20 batman s/methods/functions/
12:23 sri make an array before the for loop
12:24 batman ok.
13:05 gizmomathboy joined #mojo
13:37 aborazmeh joined #mojo
13:38 osfabibisi joined #mojo
13:57 marty joined #mojo
14:05 vicash joined #mojo
14:05 Pyritic joined #mojo
14:13 gryphon joined #mojo
14:38 jacoby joined #mojo
14:39 perlpilot joined #mojo
14:42 kivilahtio joined #mojo
14:52 dikim joined #mojo
15:24 zivester joined #mojo
15:43 * sri still loves the atom-sync plugin
15:43 sri not a single problem since i've used it
15:44 asarch joined #mojo
15:44 genio sri: ooooh.  that looks nice.  sshfs goes crazy on me so ofen that it's infuriating.
15:45 sri yea, i've tried sshfs, didn't work at all for me... way too slow
15:46 sri needed something based on rsync, there is no way i can wait more than a few milliseconds for stuff to sync after a write
15:47 genio It looks promising.  I'll definitely give it a try when I get home this evening.
15:48 sri i'm also getting surprisingly much value out of using Term::ProgressBar
15:50 sri was a bit sceptical of szabgab code at first, but it pretty much just works and it's now a default include in pretty much evert command line tool i build :)
15:51 genio I haven't given that a try.  I can imagine the usefulness though
15:52 genio sri++ # new, shiny tools (well, new-to-me!)
15:52 sri i've been ranting about so much stuff recently, guess i should also praise the things that worked :D
15:56 sri just used it in a download script https://github.com/openSUSE/desktop-file-translations/blob/git_support/50-tools/lib/SUSE/DesktopFileDownloader.pm#L63
15:57 sri the old script would just download 10k files with curl and be silent the whole time... it was infuriating
15:59 coolo on ranting... how is the keyboard coming along?
15:59 sri hahaha
15:59 sri razer folks don't talk to me anymore
16:00 sri ¯\_(ツ)_/¯
16:01 sri well, it's not like them sending me the right keycap would make a huge difference, the switch is still half open and won't stay usable for long i bet
16:04 * sri is pretty much done with razer, pretty products, but once you have a problem you're screwed
16:04 jberger I used Term::ProgressBar all the time in my scientific stuff
16:04 jberger the number crunching took quite a while and coworkers needed encouragement
16:05 genio I usually just have lots of log->debug() stuff in there and turn it on so I can see what's going on when I manually run things.
16:05 genio Term::ProgressBar would be a lot cleaner
16:06 sri oh, and talking product endorsements... i'm also really happy with the bose earphones, absolutely perfect for running
16:06 genio I need to invest in some noise cancelling (over the ear) headphones soon. We're moving into our new building in a few months and it's an open environment so my office goes bye-bye
16:07 sri https://cnet4.cbsistatic.com/img/ghR5huTL3Sp94LVHJiMffar965w=/770x433/2012/10/10/34cef09c-bb76-11e2-8a8e-0291187978f3/06BoseGreenEarbuds_35454671.jpg
16:07 genio I'm really, really not looking forward to that
16:07 sri those are the ones i got, look strange at first, but ridiculously comfortable
16:07 sri my ears usually hurt very quickly with earphones, but i can wear those for hours
16:08 sri ouch... <3 working from home
16:08 jacoby joined #mojo
16:09 disputin joined #mojo
16:09 sri the buzzing from the noise cancellation always bugs me, but i guess with an open floor office you have no choice
16:10 disputin joined #mojo
16:11 disputin joined #mojo
16:13 disputin joined #mojo
16:13 genio yea. I'm not sure whose bright idea the open floor plan was, but I'm not happy about it.  If I need to have a private conversation with someone now, I just close my door. Now we'll have to schedule time in the single office that's right in the middle of the entire department to have those conversations.
16:13 genio ugh.
16:14 PopeF joined #mojo
16:14 disputin joined #mojo
16:21 [vlad] lol
16:21 [vlad] i'm so glad i'm free
16:22 [vlad] <-- me has had a couple this evening so excuse him
16:24 genio Lots of people in the company have been linking every article they can find on why open floor plans suck for productivity.  They don't seem to care.
16:24 [vlad] heh
16:26 [vlad] i am a lone developer in africa so none of this applies to me
16:27 [vlad] but i remember being employed
16:31 khfeng- joined #mojo
16:32 khfeng_ joined #mojo
16:40 [vlad] you have my condolences
16:51 jberger all companies have ways to justify the open floor plan
16:51 jberger but it is just about cost
16:51 jberger pure and simple
16:51 jberger it is cheaper to throw desks in a big open room
16:52 jberger there aren't good enough metrics for the loss of productivity (though there are plenty of anecdotal articles)
16:52 jberger I'd say, ask the office manager to work from the break room for a week
16:52 jberger see how that goes
16:52 good_news_everyon joined #mojo
16:52 good_news_everyon [mojo] kraih pushed 1 new commit to mojo_file: https://github.com/kraih/mojo/commit/280eb0dd0314b6f7495662eb69f7c051f432e31a
16:52 good_news_everyon mojo/mojo_file 280eb0d Sebastian Riedel: avoid unnecessary catfile() calls
16:52 good_news_everyon left #mojo
16:53 * sri is still unsure about Mojo::File, but starts benchmarking a little to get a feel for things
16:53 sri regarding Path::Tiny, Mojo::File is definitely faster
16:53 [vlad] Mojo::File is a file object?
16:53 jberger [vlad]: a proposed one
16:53 jberger not in core yet
16:53 sri perl -Ilib -MMojo::File -MPath::Tiny -Mojo -E 'n { Path::Tiny::path("/foo/bar")->child("baz") } 100000; n { Mojo::File::path("/foo/bar")->child("baz") } 100000'
16:54 sri just tested that naively with vanilla perl 5.24.0 and 5.10.1
16:55 [vlad] would it be a bad thing(tm) to use use overload '""' => \&to_string;
16:55 [vlad] which is what i did with my very lame moose file object
16:55 sri https://gist.github.com/anonymous/a006865001e5fe3d2b5b3ca04c1f9413
16:56 sri i somehow doubt performance will be a big issue
16:57 sri at 200k ops per second, the slightest interaction with the actual file system will make all performance concerns moot
16:58 genio \o/
17:02 lluad joined #mojo
17:05 mishanti1 So how are people here finding the job market? Hard to find interesting employers who use perl/mojolicious?
17:08 sri [vlad]: re Mojo::File, see also https://groups.google.com/d/msg/mojolicious/8ld7jmH-qbg/wO2OCkCsEQAJ
17:20 blonewolfs joined #mojo
17:36 tyldis mishanti1: We did a mistake and hired a Java dev who said he'd be eager to learn Perl. Now all he does is piss and moan about bad IDEs - to the point of our management is doubting the rest of the team for using VIM and in consequence Perl - which is the core of our business. Fun stuff.
17:40 [vlad] have you shown him moose?
17:41 tyldis [vlad]: Indeed, but everything sucks because he can't find an IDE with a GUI towards the debugger and contextual help
17:42 tyldis And somehow the manglement now believes there is no way to do development efficiently without those featrues, and hence Perl and Mojo are dead ends. It was a surprising meeting since PErl is all we have done for 15 years in this shop.
17:42 tyldis Except for a few Java GUIs
17:43 genio Padre, ActiveState also makes an IDE
17:43 tyldis I left the meeting rather baffled and butthurt.
17:43 mishanti1 tyldis: How does his opinions have such much weight?
17:43 [vlad] not a very nice one
17:44 genio tell him to use those and be merry
17:44 tempire He's probably the loudest voice
17:44 [vlad] this guy is used to javabeans or whatever its called now
17:44 mishanti1 "such much" <-- jeez
17:44 [vlad] its like vb
17:44 tempire Combined with all the corporate advertising of java tools
17:44 tyldis genio: Padre: Released on 12 Feb 2012. "It's too outdated for me to even try it"
17:44 tempire It's hard not to see that as an executive and be swayed by it
17:45 genio tyldis: "shut up and try it"
17:45 genio and the ActiveState one isn't bad.  Some similar minded people here use it because they can't seem to work without an IDE
17:45 tempire He's going to be unhappy regardless if he's comparing it to the polish of an actively developed IDE
17:46 tyldis ActiveState.. I'll check that one out
17:46 mishanti1 tyldis: ActiveState offer all the things (including invoices) that you expect in Enterpricey $ettings.
17:46 tyldis I have tried IDEs with all the fluff, however I find it more useful to actually look up the real docs and get far more context and insight that a small tooltip with just method signatures
17:47 genio ActiveState's Komodo IDE for Perl
17:47 tyldis Oh, Komodo is AS now
17:47 tyldis Yeah, Komodo was the next stop for him to try
17:48 tempire If he can't build a customized IDE with either vim or emacs, he's being lazy, which probably translates into his code without him knowing.
17:48 mishanti1 tempire++
17:48 tempire Particularly for server side stuff.
17:48 [vlad] lazy is good
17:48 [vlad] :)
17:49 tempire If you're doing UI work, that's different. You need an IDE for that.
17:49 sri well, that he won't deliver good code pretty much goes without saying ;p
17:49 [vlad] perl does lack a vb type ide that pops up all the functions on an object when you postfix it with ->
17:49 tyldis tempire: That was my point as well. A debugger can certainly be powerful, however for a backend application I find it much easier to warn and dump stuff
17:50 sri why is komodo not good enough?
17:50 tempire What about eclipse?
17:50 tempire I know someone who used eclipse with perl quite successfully
17:50 [vlad] yeah eclipse is not bad
17:50 tempire It's ugly
17:50 tyldis tempire: Doesn't have the documentation tooltips and apparently the debugger in PErl doesn't work woth eclipse GUI
17:50 tempire Or at least it was when I tried it
17:50 tyldis He is an eclipse fan
17:50 mishanti1 After I started working more strctured with testing I have seldom used the debugger. If I need to whip out the debugger I am usually doing something overly complicated or stupid.
17:50 [vlad] its about the best you'll get for perl
17:51 tempire Indeed. The perl debugger is for a time since past for web stuff.
17:51 [vlad] i'm always using perl -d
17:51 [vlad] guess that means my stuff is overly complicated or stupid
17:52 tempire lulz
17:52 mishanti1 Hehe
17:52 tyldis ;)
17:52 [vlad] but i have yet to learn a test framework
17:52 [vlad] :(
17:52 tempire vlad: !!!
17:52 [vlad] i know
17:52 sri shame on you
17:53 * [vlad] feels shamed
17:53 mishanti1 [vlad]: Here is a pointy hat. Go stand in the corner.
17:53 tempire Use that shame for motivation
17:53 tempire Just do it. Get over that hump.
17:54 cromedome joined #mojo
17:54 [vlad] it'll take me three months
17:55 [vlad] thats what it takes me to learn a new languague to any degree of proficiency
17:55 mishanti1 [vlad]: But you'll get to keep the remains of your sanity.
17:55 jberger https://plugins.jetbrains.com/idea/plugin/7796-perl
17:55 [vlad] i know
17:55 [vlad] i must
17:55 jberger tyldis ^^
17:55 tempire Seriously. And don't even start with anything hard.
17:56 tempire Just Test::More.
17:56 tempire Once you get a taste you won't be able to stop.
17:56 tyldis jberger: GUI frontend to the debugger and "tool tip" documentation is apparently missing in that
17:57 tyldis I haven't tried, I just have to take his word for it.
17:57 sri ???????????? shame! shame! shame!
17:58 tyldis And after that meeting I wasn't too motivated to test and dispute.
17:59 jberger anyone that needs tooltips in order to program isn't worth keeping
17:59 jberger I understand the desire for cross-linking info
17:59 jberger but this isn't a static language, so some of that is going to suffer necessarily
18:00 jberger but to say "I can't program without tooltips" is patently insane
18:00 tyldis It comes in handy, because in Java you apparently write wrappers around everything (even the wrappers) which makes it hard to remember any syntax...
18:00 jberger he's saying "the IDE tells me what code to write"
18:00 jberger so he's saying he isn't necessary
18:00 tempire Indeed. Java is all about nouns, so functionality is hidden.
18:00 jberger yes, Java without snippets is very hard
18:01 jberger but most editors have snippets
18:02 tempire Then again this is an echo chamber
18:02 jberger I'm not saying that they aren't helpful
18:02 jberger sure they are helpful
18:03 sri this discussion is kinda boring
18:03 jberger but picking a language just because your other language is so obtuse that the tooling has gotten over the hump is silly too
18:03 jberger sri: noted
18:03 jberger lunchtime anyway :-P
18:14 cfedde joined #mojo
18:14 gryphon joined #mojo
18:15 good_news_everyon joined #mojo
18:15 good_news_everyon [mojo] kraih pushed 1 new commit to mojo_file: https://github.com/kraih/mojo/commit/b2cd66d995c76a7ea5be17d0b5356ec9ba05f5d7
18:15 good_news_everyon mojo/mojo_file b2cd66d Sebastian Riedel: use Mojo::File in Mojo::Home
18:15 good_news_everyon left #mojo
18:18 good_news_everyon joined #mojo
18:18 good_news_everyon [mojo] kraih pushed 1 new commit to mojo_file: https://github.com/kraih/mojo/commit/b27fad63e43106313f9371b389fd40b55c7ef9ef
18:18 good_news_everyon mojo/mojo_file b27fad6 Sebastian Riedel: use Mojo::File in Mojo::Server::Hypnotoad
18:18 good_news_everyon left #mojo
18:35 good_news_everyon joined #mojo
18:35 good_news_everyon [mojo] kraih pushed 1 new commit to mojo_file: https://github.com/kraih/mojo/commit/0b6dc49b8ade94ade428f1aa321c944c2212b001
18:35 good_news_everyon mojo/mojo_file 0b6dc49 Sebastian Riedel: use Mojo::File in a few more modules
18:35 good_news_everyon left #mojo
18:36 sri not all code gets better
18:36 sri it does help the documentation examples a lot though
18:36 sri https://github.com/kraih/mojo/compare/mojo_file#diff-65cf575515a0ad467194433b39cd30d3
18:38 dod joined #mojo
18:39 batman sri: do you really need to_string() all over the place?
18:40 sri hehe, was about to say that it's the appended ->to_string that usually makes it look verbose
18:40 dod joined #mojo
18:40 sri but yea, those places send the path elsewhere, and currently mostly return strings
18:40 sri suddenly returning an object would be awkward
18:41 sri Mojo::Home will have to be completely redesigned though
18:41 sri since app->home->rel_file(...) and app->home->parts are just weird now
18:41 batman also, do basename() and dirname() need to return new objects?
18:42 sri it will likely become a Mojo::File subclass with a few methods added with deprecation notices
18:42 sri batman: what else?
18:42 batman just strings
18:43 sri https://github.com/kraih/mojo/compare/mojo_file#diff-52d3ad944c9afec616d8ccbcc2f1ce2eR45
18:43 sri https://github.com/kraih/mojo/compare/mojo_file#diff-a9009b150b211dbbfc0a7453ca27e660R20
18:43 sri https://github.com/kraih/mojo/compare/mojo_file#diff-6be2b0f672c3cc12d7ecbad53dff1710R52
18:43 sri you get the idea ;p
18:45 batman i think i would let basename() and dirname() return strings, and then add sub parent { path(shift->dirname) }
18:46 sri be my guest, implement it
18:46 batman i won't do it, if you think it's a bad idea :/
18:47 sri you've not provided enough information to judge
18:48 sri for all i know it could be awesome or terrible
18:49 batman okey. cool. will have a look later i hope.
18:52 good_news_everyon joined #mojo
18:52 good_news_everyon [mojo] kraih pushed 1 new commit to mojo_file: https://github.com/kraih/mojo/commit/ccc14ac7f779abdcc4e1eab2d9b5d3e0a696fb44
18:52 good_news_everyon mojo/mojo_file ccc14ac Sebastian Riedel: missed a module
18:52 good_news_everyon left #mojo
19:00 Pyritic joined #mojo
19:01 good_news_everyon joined #mojo
19:01 good_news_everyon [mojo] kraih pushed 1 new commit to mojo_file: https://github.com/kraih/mojo/commit/6a21b74950c008b96f2e700765e1223d901ee10d
19:01 good_news_everyon mojo/mojo_file 6a21b74 Sebastian Riedel: allow subclassing of Mojo::File
19:01 good_news_everyon left #mojo
19:04 perlpilot joined #mojo
19:27 Pyritic joined #mojo
19:38 purl joined #mojo
19:48 ledet joined #mojo
20:06 jnbek joined #mojo
20:11 dod joined #mojo
20:23 bwf joined #mojo
20:23 perlpilot joined #mojo
20:26 sri anyone know how that split on a slash is not a bug? https://github.com/kraih/mojo/blob/master/lib/Mojolicious/Lite.pm#L16
20:29 sri interesting, in 1.0 it looked like this https://github.com/kraih/mojo/blob/v1.0/lib/Mojolicious/Lite.pm#L23
20:29 sri might be a very old bug that accidentally works
20:33 sri older than the first tag https://github.com/kraih/mojo/blob/v0.999920/lib/Mojolicious/Lite.pm#L26
20:33 jberger nice
20:34 sri yea, it's very old :) https://github.com/kraih/mojo/commit/a14662686cea1a343203cc680c25c0db53e1744e#diff-e6315209d70edce7210fba0dd37b78d6R23
20:35 sri note the commit description
20:35 jberger I'm not 100% sure I was using perl yet then :o
20:36 jberger this is my first public perl commit: https://github.com/jberger/MakeBeamerInfo-Legacy/commit/c7a251889215dd488b4caf3dd9b53bc4e2e96428
20:36 jberger I don't think I'd been using it much earlier than then
20:37 sri :)
20:40 good_news_everyon joined #mojo
20:40 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/bf6682aa6870ce2ec62ad87c11d772dd0215fcaf
20:40 good_news_everyon mojo/master bf6682a Sebastian Riedel: there should be no need to split the path
20:40 good_news_everyon left #mojo
20:40 sri would be funny if that causes a portability problem
20:42 stephan48 that change or the split which was in place beforehand?
20:43 disputin joined #mojo
20:46 sotona joined #mojo
20:47 jberger man, that first script of mine
20:47 jberger a greatest hits of bad perl
20:48 Grinnz joel's script archive
20:49 sri lol
20:49 sri ohoh, we have a real subprocess bug https://github.com/kraih/mojo/issues/1026
20:58 good_news_everyon joined #mojo
20:58 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/59f03213ffc38f3bc69c93c7320bbb8b2bebb20d
20:58 good_news_everyon mojo/master 59f0321 Sebastian Riedel: fix a bug in Mojo::IOLoop::Subprocess where the pipe used for IPC could disappear because of a timeout (closes #1026)
20:58 good_news_everyon left #mojo
20:59 jberger nice one
21:07 marty_ joined #mojo
21:37 PryMar56 joined #mojo
21:59 marty joined #mojo
22:05 disputin joined #mojo
22:20 marty_ joined #mojo
22:31 orev joined #mojo
23:26 batman https://metacpan.org/release/JHTHORSEN/Test-Mojo-Role-Selenium-0.01 :)
23:27 batman wonder if it'll work as good as i hope...
23:27 batman :)
23:39 pink_mist hmm, no announce of that by cpan_mojo?
23:41 tchaves joined #mojo
23:47 PryMar56 joined #mojo

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