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

IRC log for #mojo, 2014-09-27

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

All times shown according to UTC.

Time Nick Message
00:00 ceej I could but I’d rather have less dependencies, especially not having to require ruby to install foreman :)
00:01 zgarnog joined #mojo
00:03 D4RK-PH0ENiX joined #mojo
00:06 marty joined #mojo
00:06 Mso150 joined #mojo
00:26 marty_ joined #mojo
00:26 ua1 joined #mojo
00:26 disputin1 joined #mojo
00:26 Mikey joined #mojo
00:27 kbenson1 joined #mojo
00:28 chansen_ joined #mojo
00:28 Gedge_ joined #mojo
00:28 ralesk_ joined #mojo
00:28 dexteruk_ joined #mojo
00:28 allison_ joined #mojo
00:28 robins joined #mojo
00:28 suy_ joined #mojo
00:30 bd_ joined #mojo
00:30 heytitle1 joined #mojo
00:30 rofl__ joined #mojo
00:30 jeff_ joined #mojo
00:30 cosimo_ joined #mojo
00:30 stephanj joined #mojo
00:30 caeles joined #mojo
00:30 mikegrb_ joined #mojo
00:30 jmmills_ joined #mojo
00:30 DaniBunn1 joined #mojo
00:30 michael_ joined #mojo
00:31 jzawodn_ joined #mojo
00:31 elik_ joined #mojo
00:32 throughnothing_ joined #mojo
00:32 Rallias joined #mojo
00:32 bjoernfan joined #mojo
00:33 Claudio joined #mojo
00:33 DaTa joined #mojo
00:34 yusukebe joined #mojo
00:34 bzero joined #mojo
00:34 preaction joined #mojo
00:36 dahl joined #mojo
00:37 Zx3 joined #mojo
00:41 mudler joined #mojo
00:41 inokenty joined #mojo
00:42 howitdo joined #mojo
00:43 xtetsuji joined #mojo
00:43 jojo joined #mojo
00:43 xevz joined #mojo
00:43 go|dfish joined #mojo
00:43 Mso150 joined #mojo
00:45 Kundun joined #mojo
00:47 sri joined #mojo
00:47 ssm joined #mojo
00:47 dotan_ok joined #mojo
00:48 lestrrat joined #mojo
00:49 abhishekisnot joined #mojo
00:49 d4rkie joined #mojo
01:47 ceej I ended up doing this https://gist.github.com/f00c8087a8e6cb56ea14
01:50 marty interesting:   http://blogs.enterprisedb.com/2014/09/24/postgres-outperforms-mongodb-and-ushers-in-new-developer-reality/
01:50 ceej how come it’s not possible to just do "env.$self->mode.sh” ?
01:54 marty postgres might be my DB of choice.  Being able to combine structured and non structured data in a single DB is very compelling
02:11 noganex_ joined #mojo
02:15 ceej marty: yee, pgsql is pretty nice :)
02:16 irq joined #mojo
02:18 sujithm joined #mojo
02:32 bpmedley joined #mojo
02:49 good_news_everyon joined #mojo
02:49 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/Bhj2ww
02:49 good_news_everyon mojo/master 9f76dc7 Sebastian Riedel: more consistent examples
02:49 good_news_everyon left #mojo
02:49 good_news_everyon joined #mojo
02:49 good_news_everyon [mojo] kraih tagged v5.45 at 2de02d7: http://git.io/UZdlOw
02:49 good_news_everyon left #mojo
02:50 sujithm joined #mojo
03:02 basic6_ joined #mojo
03:07 good_news_everyon joined #mojo
03:07 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/hLL02g
03:07 good_news_everyon mojo/master 93b33b0 Sebastian Riedel: bump version
03:07 good_news_everyon left #mojo
03:15 sri i hate PAUSE
03:16 sri it ate my upload
03:20 good_news_everyon joined #mojo
03:20 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/DxthXw
03:20 good_news_everyon mojo/master eb2ce87 Sebastian Riedel: uploading to CPAN is fun
03:20 good_news_everyon left #mojo
03:21 good_news_everyon joined #mojo
03:21 good_news_everyon [mojo] kraih tagged v5.46 at 89ac8e4: http://git.io/Tq2hRQ
03:21 good_news_everyon left #mojo
03:22 good_news_everyon joined #mojo
03:22 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/IFkLMw
03:22 good_news_everyon mojo/master 108cd9d Sebastian Riedel: bump version
03:22 good_news_everyon left #mojo
03:22 jberger sri-- # burning through our version numbers :-P
03:23 sri YOU MEAN PAUSE
03:23 * jberger is looking at couchdb
03:23 jberger Anyone have thoughts?
03:23 sri relax
03:23 jberger So i hear
03:24 sri OH C'MON PAUSE
03:24 sri https://metacpan.org/release/SRI/Mojo-5.45
03:25 * jberger smacks pause
03:48 sri rofl
03:48 sri https://metacpan.org/recent
03:48 sri Mojolicious-5.45, Mojolicious-5.46, Mojo-5.45...
03:50 sri Mojo-5.45 and Mojolicious-5.45 actually have different sizes :S
03:51 disputin joined #mojo
03:52 sri and metacpan is totally confused now showing Mojolicious-5.45 as the latest release
04:01 KCL_ joined #mojo
04:08 woz joined #mojo
04:14 phillipadsmith joined #mojo
04:17 chansen joined #mojo
04:47 rem_lex|pivo joined #mojo
04:49 jalie joined #mojo
04:51 ceej sri: I just did an update… Successfully installed Mojolicious-5.46 (upgraded from 5.44)
04:56 ceej surprised there’s no plugin for http://rethinkdb.com/ :)
05:10 woz joined #mojo
05:15 KCL joined #mojo
05:20 KCL_ joined #mojo
05:34 rwp joined #mojo
06:01 rem_lex| joined #mojo
06:11 woz joined #mojo
06:46 jalie joined #mojo
06:47 ceej_ joined #mojo
06:51 Vandal joined #mojo
06:58 ceej__ joined #mojo
07:02 ryozi joined #mojo
07:04 sujithm joined #mojo
07:12 woz joined #mojo
07:14 batman what's bad about rethinkdb?
08:13 woz joined #mojo
08:13 denis_boyun joined #mojo
08:16 sh4 joined #mojo
08:20 denis_boyun_ joined #mojo
08:27 hertzhaft joined #mojo
08:31 dod joined #mojo
08:39 hertzhaft https://github.com/njlg/perl-rethinkdb makes use of Mojo::Base :-)
08:53 denis_boyun joined #mojo
08:57 berov joined #mojo
09:13 denis_boyun_ joined #mojo
09:14 woz joined #mojo
09:16 denis_boyun__ joined #mojo
09:27 denis_boyun joined #mojo
09:30 denis_boyun_ joined #mojo
09:39 denis_boyun joined #mojo
09:45 berov Some may be interested to know that Atom now supports has,extends,use etc.. Modern Perl keywords . Browsing the documentation is also supported if you do not have __END__ label above it.
09:47 berov http://blogs.perl.org/users/krasimir_berov/2014/09/extended-rules-to-support-modern-perl-in-atom-symbols-view-package.html
09:56 odc joined #mojo
09:59 meshl joined #mojo
10:02 basiliscos joined #mojo
10:12 denis_boyun_ joined #mojo
10:15 woz joined #mojo
10:25 sujithm joined #mojo
10:45 meshl joined #mojo
10:48 hertzhaft left #mojo
11:16 woz joined #mojo
11:16 D4RK-PH0ENiX joined #mojo
11:28 woz joined #mojo
11:30 D4RK-PH0ENiX joined #mojo
11:59 woz joined #mojo
13:00 woz joined #mojo
13:01 basiliscos joined #mojo
13:10 _eugen joined #mojo
13:37 sujithm joined #mojo
13:52 * sri yawns
13:59 marty o/
14:01 woz joined #mojo
14:13 dotandimet joined #mojo
14:22 jberger |o|
14:23 jberger -o-
14:34 bpmedley_ joined #mojo
14:37 dexteruk joined #mojo
14:37 dvinciguerra joined #mojo
14:45 dvinciguerra joined #mojo
14:46 Akron joined #mojo
14:54 GabrielVieira joined #mojo
15:03 woz joined #mojo
15:06 meshl joined #mojo
15:10 dotandimet joined #mojo
15:21 basiliscos joined #mojo
15:42 disputin joined #mojo
15:54 laouji joined #mojo
15:55 dotandimet joined #mojo
15:57 jberger sri: what was the opinion on Pg::PQ
15:58 jberger other than the big alpha warning
15:58 sri not looked into it, i'm done with dealing anything not proven 100% reliable when it comes to databases
16:00 sri *+with
16:01 jberger my google foo has really been failing me this morning
16:04 sri actually... s/with dealing/dealing with/
16:04 woz joined #mojo
16:04 sri my typo foo is today strong
16:05 jberger the typo is strong with this one
16:05 jberger (humorous to note that I almost sent that message with the word "string" rather than "strong")
16:13 jberger databases are hard
16:13 * jberger goes shopping
16:14 sri Go sure is verbose, but those screencasts are pretty nice https://gophercasts.io/lessons
16:16 jberger I think that for my personal use I'm going to start using DBM::Deep mostly again
16:21 sri i guess the whole mongodb thing has made me super conservative when it comes to databases :o
16:26 jberger https://metacpan.org/pod/Protocol::CassandraCQL
16:26 jberger sri: don't blame you
16:34 bobkare joined #mojo
16:37 * sri is still annoyed by the pause upload thing from yesterday
16:38 sri too bad there is no way to get rid of the 5.46 release
16:40 mst pause upload thing?
16:40 mst jberger: I've had pretty good fun using sqlite with a YAML::XS blob column for the actual object
16:40 sri yea, look how 5.45 appeared after 5.46 on metacpan, and metacpan thinks 5.45 is the latest release ;p
16:41 mst sri: that sounds like "order metacpan noticed it" rather than anything to do with pause?
16:41 sri i'vw had an upload to pause time out yesterday
16:41 sri s/w/e/
16:42 sri of course i couldn't upload it with the same name again
16:43 sri i never went through the whole repair upload process before... tried uploading Mojo-5.45.tar.gz then and rename it with that
16:44 sri when that didn't work i just said fsck it, and bumped the version number...
16:46 sri and after that a Mojolicious-5.45 appeared on pause somehow
16:46 irq joined #mojo
16:46 sri soo... i ended up with Mojo-5.45, Mojolicious-5.46 and Mojolicious-5.45 in that order -.-
16:47 sri and i still have no fricking clue what that repair upload step is supposed to do
16:47 jberger sri: I never figured that one out either
16:47 jberger anyway, it all goes away with a 5.47 right?
16:48 sri yea
16:48 sri we need new features to add quickly!
16:48 sri GO FIND BUGS!
16:48 * jberger BREAKS ALL THE THINGS
16:48 sri some super critical bug would be nice right now
16:49 jberger lets ask
16:49 jberger purl: bug?
16:49 purl jberger: bugger all, i dunno
16:49 jberger well that's no help
16:51 sri omg, mysql now speaks http and json :O
16:51 sri http://de.slideshare.net/nixnutz/http-plugin-for-mysql-39598656
16:52 jberger mst: what made you pick YAML::XS vs all the other things?
16:53 mst jberger: fastish, handles objects without fucking about, but results in a text representation that's not completely horrible to read for debugging
16:54 mst Sereal would probably be a better choice performance-wise, but I'll happily trade a bit of speed for ease of debugging if I can afford to
16:55 jamesaxl joined #mojo
16:55 odc all right, Mango is forked. I hope i won't regret this...
16:57 basic6_ joined #mojo
16:58 jberger mst: was going to ask about Sereal, thanks
16:59 sri hmm... interesting... mysql folks just let you send the sql query with the path
16:59 sri curl --user basic_auth_user:basic_auth_passwd --url "http://…/sql/db/SELECT+col_float+FROM+sql_types"
16:59 mst basically, being able to do a SELECT and actually read the output made it worth using YAML::XS for me
17:00 sri curl --user basic_auth_user:basic_auth_passwd --url "http://new.mysql.com/sql/db/SET+@no='session'"
17:01 sri jberger, mst: how come you two don't care about querying at all?
17:02 sri i mean, if you just want a key/value store, why bother with an SQL database?
17:02 mst well, the trick for me is that the things I want to query, I materalise as columns
17:03 sri i really like the idea of jsonb in postgresql
17:06 woz joined #mojo
17:20 KCL joined #mojo
17:41 Mso150 joined #mojo
17:47 bobkare joined #mojo
18:07 woz joined #mojo
18:08 hertzhaft joined #mojo
18:20 berov joined #mojo
18:26 dotandimet joined #mojo
18:28 jberger decides to read about redis
18:34 sri it's ridiculous how much higher the quality of conversations in #postgresql is than in #mongodb
18:36 jberger sri: I know very little about postgres, but one of the few things seems to be that the community is strong
18:46 sri lol, found the first thing to hate about postgresql
18:46 zgarnog joined #mojo
18:46 sri it does multiple inheritance for tables :O
18:50 jberger http://www.moviefancentral.com/images/pictures/review52692/_0a0aaagarth.jpg?1365708710
18:51 moritz sri: afaict, table inheritance in its current form should be avoided
18:51 sri :D
18:51 moritz sri: at least I haven't found a use case where the documented limitations aren't a complete show-stopper
18:52 moritz it looks like a really cool and useful feature, until you read the fine print
18:52 sri the 32tb limit for tables is also interesting
18:53 moritz http://www.postgresql.org/docs/9.3/static/ddl-inherit.html#DDL-INHERIT-CAVEATS
18:53 jberger 32tb, yeah, that's not going to work for me
18:54 jberger :-P
18:56 mst sri: multiple table inheritance is not multiple inheritance for tables
18:56 mst also, table inheritance is basically useful for partitioning
18:56 mst for everything else, views with triggers tends to be nicer
19:04 Mso150 joined #mojo
19:07 jberger 32TB, HA! DBM::Deep has an overall limitation of 16XB! Take that!
19:08 woz joined #mojo
19:13 moritz what's the 32bit limitation? for columns in a table? or total number of tables?
19:14 dotandimet joined #mojo
19:17 sri mst: reading through the docs moritz linked to i can see how it's totally different
19:18 sri the fact that you select from *descendants* is freaky
19:19 moritz the way I'd like to use table inheritance (if it weren't for those limitations) is to have a "pages" table for my pages (for a blog or a CMS), and then several tables inheriting from it that contain more specific page types (polls, posts, login, forms, ...)
19:20 moritz but that's just not possible to do in a safe way right now
19:22 sri what also freaks me out a bit is some of the more crazy looking operators
19:24 mst moritz: that's what views with backing tables work well for
19:24 davido___ joined #mojo
19:38 marcus You can write postgresql stored procedures in Perl?
19:38 basiliscos joined #mojo
19:40 sri yea http://www.postgresql.org/docs/9.4/static/plperl.html
19:43 mst course you can. timb submitted a bunch of patches so you can NYTProf 'em as well
19:43 mst I fatpacked an EXPLAIN server at one point so I could do 'SELECT explain_server(8080);'
19:52 sri there's a lot to like about postgresql
19:52 sri moritz: it's a 32TB size limit on tables
19:54 moritz sri: oh. Doesn't sound like it's a problem for my usual use cases :-)
19:56 mst and, y'know, you can use table inheritance to have multiple 32Tb tables selectable via a single name ;)
19:58 sri just reading about partitioning :)
19:59 mst you have to be quite careful to get the constraint exclusion stuff to kick in
19:59 mst but it's fucking clever when it does
19:59 sujithm joined #mojo
20:05 sri now inheritance makes sense
20:06 sri can't say i like it particularly, but i get it
20:07 mst right. for transparent partitioning e.g. by date, it's fuckin' ace
20:07 mst it's not as easy to configure as I might like
20:07 mst but once it works the app code can mostly pretend it isn't therre
20:07 mst and that part I do like
20:08 sri the docs have gotten really good too
20:09 sri the example makes the use case very clear http://www.postgresql.org/docs/9.4/static/ddl-partitioning.html
20:09 woz joined #mojo
21:02 trone joined #mojo
21:10 woz joined #mojo
21:11 chansen jberger: checkout CBOR (<http://cbor.io/>) also a nice serialization format, especially if you want cross language support
21:14 jberger chansen: thanks I'll take a look
21:14 jberger my focus is really on db choice now
21:14 jberger but if/when I get to storing blobs
21:20 chansen ok, I really like the extensibility of CBOR using tags (<http://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml>)
21:25 chansen example <https://metacpan.org/source/CHANSEN/Time-Moment-0.18/eg/cbor.pl>, using Time::Moment
21:27 lb joined #mojo
21:35 Akron joined #mojo
21:51 sri guess i'll try this for a new minion backend http://johtopg.blogspot.se/2010/12/queues-in-sql.html
21:55 sri batman/marcus: /list seems still broken btw.
22:11 woz joined #mojo
22:15 nicomen sri: is minion blocking?
22:15 sri yes
22:15 nicomen ah ok ;)
22:15 nicomen tell me again, are we still not able to do non-blocking SQL?
22:17 sri i explained that yesterday in detail i believe
22:17 nicomen ah, goodie, will check logs
22:18 sri this sentence covers it well though http://irclog.perlgeek.de/mojo/2014-09-25#i_9411072
22:19 Akron joined #mojo
22:21 sri as long as you have a low latency connection to the database (which you hopefully always have), blocking i/o is good enough
22:23 sri DBD::Pg and DBD::mysql allow async waiting for queries to finish, which is usually where you encounter latency from the database
22:24 sri so what i'll be trying is a hybrid solution of blocking i/o and non-blocking waiting for queries
22:27 nicomen makes sense
22:31 chansen why use a RDBM for non relational data?
22:33 sanya_com_ua joined #mojo
22:36 lb implementing acid yourself gets boring after a while
22:40 chansen It's simply the wrong tool for the job, a RDBM is optimized for relational data.
22:41 chansen Implementing ACID isn't that hard
22:42 jamesaxl joined #mojo
22:58 denny joined #mojo
23:04 Mso150 joined #mojo
23:04 sri of course nobody tries to pitch me a new feature the day i really want a new mojolicious release -.-
23:07 chansen heh
23:19 janus joined #mojo
23:31 meshl joined #mojo
23:43 nicomen sri: well, I need a way to alter the actual method line sent by Mojo::UserAgent, is there a way I can do that? (namely I want to reproduce a faulty "GET path HTTP/1.1" line
23:43 nicomen (no slash at the beginning of the path)
23:53 sri yea, lets not do that

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