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

IRC log for #mojo, 2015-11-09

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

All times shown according to UTC.

Time Nick Message
00:13 sri https://github.com/kraih/minion/commit/79b6260e7ab7626cb35c13346aad584e628b6fbd
00:13 sri so that's minion 4 ;p
00:14 sri downside is more work for backend authors
00:41 cpan_mojo Mojolicious-Plugin-AutoRoute-0.18 by KIMOTO https://metacpan.org/release/KIMOTO/Mojolicious-Plugin-AutoRoute-0.18
00:45 lluad joined #mojo
01:24 aborazmeh joined #mojo
01:37 sri and once again... if anyone has any more ideas for minion, now is the time ;p
01:42 nicomen I don't know if it's in the scope, but is it possible to have recurring jobs, and ensuring jobs are run in a certain order?
01:43 sri when you say order, do you mean first job has to be finished before the second starts?
01:43 nicomen with order I meant preferably after another job has finished for instance
01:43 nicomen yup
01:43 sri you can just enqueue the second job from the first job
01:43 nicomen ah right, good point
01:44 nicomen although perhaps a shortcuts for that might be nice, like adding an array of tasks?
01:44 sri recurring could be done similarly
01:44 sri "$job->finish; $job->retry({delay => 60});" from within your job
01:45 nicomen ok, I think I see what you mean, I think this might be better suited as a separate module for syntax sugar
01:45 sri a job can finish and resquedule itself
01:45 nicomen sweat, all set then
01:45 nicomen *sweet
01:45 sri both could also be done natively in the job queue itself, but implementations mostly suck
01:46 sri job dependencies especially make the code very very complicated and slow
01:46 nicomen ok, thanks for the enlightenment, I think I know now how to replace our home-made poc "job" system
01:47 sri bpmedley has done some experiments, and i think it was an order of magnitude slower even with the best version
01:48 bpmedley My experiments didn't even have the satisfaction of failed smoke.. :).  Also, sorry for my lack of input lately.
01:49 sri actually, i think i like a job that creates a new job of the same type for recurring more
01:49 nicomen can I run the jobs in one process, and add them in another?
01:50 sri of course
01:50 nicomen ah right
01:50 nicomen sorry
01:50 nicomen doc reading foobar
01:53 Zoffix joined #mojo
01:59 sri bpmedley: wait till you see all the recent minion changes ;p
01:59 bpmedley Cool
02:00 sri https://github.com/kraih/minion/compare/v2.05...master
02:01 sri i think that's what the mysql backend has to catch up with
02:02 bpmedley Thanks.  I'll try and get that updated within a week or so.
02:03 sri i'm still not entirely sure on the attempts feature, so feedback would be very welcome
02:05 sri quite a bit of new logic pushed into the backend
02:06 sri tried making it work in Minion::Job, but it didn't
02:29 vervain joined #mojo
03:21 kaare joined #mojo
03:53 aborazmeh joined #mojo
04:19 vervain joined #mojo
05:19 vervain joined #mojo
05:39 CIAvash joined #mojo
05:52 inokenty-w joined #mojo
06:20 vervain joined #mojo
06:24 irqq joined #mojo
06:44 cheesekun joined #mojo
07:13 crab [debug] Template "not_found.production.ARRAY(0x3215890).ep" not found
07:13 crab huh
07:14 Grinnz that must be one of those new web 3.0 formats
07:15 crab wonder how that happened
07:22 vervain joined #mojo
07:27 McA joined #mojo
08:09 eseyman joined #mojo
08:14 Vandal joined #mojo
08:17 trone joined #mojo
08:23 vervain joined #mojo
08:31 osfabibisi joined #mojo
08:31 AndrewIsh joined #mojo
09:24 vervain joined #mojo
09:25 nic crab is probing all the attack vectors
09:25 CandyAngel Mojo's routing is nifty
09:26 nic sri: I had an implementation of job dependencies
09:26 nic well, two implementations
09:26 nic the first one used a separate queue per real queue
09:27 nic so a job destined for queue abc, but with dependencies, would sit in something like depends:abc until its dependencies were seen to be resolved
09:28 nic (by a separate dependency checker, which behaved essentially like a worker but checked dependencies, but you guessed that bit)
09:29 nic That one had the least impact on existing Minion, but I really didn't like it cos the job queuer had to take responsibility for putting the jobs on the pseudo-queue
09:29 nic The second one used an additional state: 'depends'
09:29 CandyAngel Well, I say that.. I just ended up with a regex in the address bar :P
09:29 nic Upon queueing, if it arrived with dependencies then it went straight into 'depends' state
09:30 nic So that implementation had a bigger impact through the code, but was much nicer to work
09:30 nic Trouble was, it worked in File but I didn't have it quite working in pg
09:30 nic and then Minion moved on a bit
09:31 nic If you're at all interested, I'll get it working in pg
09:32 nic Otherwise I'll leave it
09:34 nic I've discovered I don't really like working with pg, so I'm planning to do a connector-based Mysql backend, or a hybrid with Redis
09:35 nic "don't really like" means "have twenty years experience with MySQL and anything else doesn't map as easily to my brain"
09:35 nic <-  is on the MySQL developer list
09:53 bjakubski joined #mojo
09:57 berov joined #mojo
10:11 sri well, pg is now the reference backend
10:19 sri a depends state does sound better for performance
10:20 sri i guess ->repair would then have to make sure the job it depends on is still there, or fail the job
10:21 sri which ends up being a recursive search with multiple dependencies, which might be a little expensive
10:22 nic In my mind, jobs with dependencies are second-class jobs and shouldn't impact repair
10:22 nic if you use a separate process to check dependencies then each implementation can tune how frequently that should be checking
10:23 sri that's where i disagree
10:23 nic and I'm imagining more than one dependency checker
10:23 nic so checking jobs destined for the 'high-priority' queue can be done more frequently
10:24 sri not even sure i want attempts to stay, need to figure out how smart i want the queue to be
10:24 nic k
10:25 vervain joined #mojo
10:25 nic Once I get my own backend implemented I'll be able to play more, so nothing from me for this round of release
10:26 sri btw. there's already https://metacpan.org/pod/Minion::Backend::mysql
10:27 sri which is a little outdated now
10:27 CIAvash joined #mojo
10:30 nic yeah, that's a good start
10:31 nic but we use Mojar::Mysql here; been battle-tested for more than 10 years and now most of our infrastructure relies on it
10:37 nic (I'm expecting it'll be trivial to backport anything I come up with)
10:41 HtbaaPi joined #mojo
10:42 irqq joined #mojo
10:50 * sri would really like to hear some opinions on the attempts feature from minion users (even if it is just "don't care")
10:51 * sri pokes CandyAngel, Grinnz, bpmedley, jberger and nic
10:52 sri at this point, it is basically this https://github.com/kraih/minion/compare/v3.0...master
10:53 sri as explained here http://mojolicio.us/perldoc/Minion#enqueue
10:55 nic I think the feature is great
10:56 nic I'm expecting a _lot_ of my day-to-day issues will disappear once I have access to {attempts => 2}
10:56 Mattjes joined #mojo
10:56 Mattjes Hello
10:57 Mattjes Where can i find the dokumentation for older Mojo-Dom-Versions?
10:57 nic but my current (non-Minion) solution does some 'stuff' (logging into a db table) when it detects the start of a re-attempt
10:57 CandyAngel sri: The doc implies that the delay is always exponentially increasing, regardless what 'backoff' actually does
10:57 nic so I need to figure out how to port that to the Minion feature
10:58 nic Mattjes: github.com/kraih/mojo gives you access to every commit
10:59 nic and cpan gives you access to every release
10:59 Mattjes Thank you nic!
10:59 nic so the easiest two ways are to browse the commit history of github or checkout the old version from cpan
10:59 nic as ever, cpanm is your friend
10:59 sri nic: that's actually pretty easy, there's an event where you can check $job->info http://mojolicio.us/perldoc/Minion/Job#failed
11:00 CandyAngel sri: Looks good to me, I'm sure I'll end up with a use case for it eventually :)
11:00 Mattjes ty ;-)
11:00 nic cpanm -l /tmp/mojo Mojolicious@4.01
11:00 sri CandyAngel: i think referring to the exponential default is fine, every such description has a link to the backoff attribute, which goes further
11:01 sri Mattjes/nic: there's also "jump to version" on https://metacpan.org/release/Mojolicious
11:01 nic sri: yes, I need to make more use of events.  And rely more on Minion's own stats rather than my current way of collecting them in my main db
11:01 nic Mattjes: what sri said; much easier :)
11:02 Mattjes Cool! That was what I was looking for!
11:02 Mattjes Thanky oyu very much!
11:03 CandyAngel sri: It's that it says "exponentially increasing *based on* backoff", which makes it sound like uses the return value of backoff in some additional calculation to get the next delay
11:03 Mattjes sri: Darf man hier Deutsch sprecehn?
11:03 CandyAngel Like $delay = exp($job->backoff)
11:03 sri Mattjes: no
11:05 Mattjes sri: ok - my English is not so good but I would like to ask once how much time you have been in the programming of Mojo invest
11:06 Mattjes Mojo is amazing ;-)
11:06 sri Mattjes: don't think that's quantifiable
11:06 Mattjes It must be a lot of time
11:06 sri since https://github.com/kraih/mojo/blob/master/Changes#L4455
11:07 sri and catalyst prepared me for that before
11:07 CandyAngel \o/
11:07 * CandyAngel is victorious
11:07 Mattjes sri: You are creating Mojo in your spare time?
11:07 Mattjes only?
11:08 sri yes
11:08 Mattjes wow
11:08 Mattjes amazing
11:08 sri CandyAngel: how would you change that sentence then?
11:09 Mattjes sri: Mojo is great and helped me a very often! With mechanize and Mojo i spider a lot of different website, normalize the content then and process it further!
11:10 CandyAngel "transition back to inactive with a delay calculated by "backoff"
11:10 Mattjes Without Mojo I could not
11:10 CandyAngel Mattjes: I use Mojo in the same way :)
11:10 CandyAngel But with Minion (ab)use too
11:11 CandyAngel Drives sri up the wall :)
11:11 Mattjes Thanks for their efforts regarding Mojo
11:11 Mattjes ;-)
11:12 nic CandyAngel: Is your point that the time sequence is polynomial rather than exponential?
11:12 CandyAngel nic: No, it's that it might not be either. Backoff could just be sub { return 1; }
11:13 CandyAngel Or sub { return 60 / $attempts; }
11:13 CandyAngel (less delay each attempts)
11:14 nic those are polynomial; so you mean it is not necessarily increasing?
11:14 CandyAngel Mattjes: Right now, I'm making a Lite app for parsing daily/indie royale pages. Providing easy download links and .md5 files (which also contain steam keys etc.)
11:15 CandyAngel nic: It could also just return a random number
11:16 Mattjes CandyAngel: Cool thing
11:16 sri CandyAngel: actually... https://github.com/kraih/minion/commit/b3b7857fa38c4bc3291e20dba2d84da5422ac5f3
11:17 sri (re driving me up the wall ;p)
11:17 CandyAngel :P
11:17 nic driving you up the wall on a non-exponential curve
11:19 nic sri: just a technicality, but x**4 is polynomial rather than exponential
11:19 sri https://github.com/kraih/minion/commit/cbe8d5cea5bf88be2dc9aec1d08cd09139411982
11:19 nic \o/
11:19 CandyAngel If we keep going, it's going to end up as
11:20 CandyAngel "Does a thing"
11:21 sri i think "(retries ** 4) + 15" is stolen from queue_classic
11:21 sri "think" because i stole it from sidekiq ;p
11:22 Zoffix joined #mojo
11:26 vervain joined #mojo
11:26 CandyAngel I wonder if I should set up a polipo cache or something while I am developing these things
11:30 CandyAngel And that reminds me, I need to modify DownThemAll
11:32 melo joined #mojo
11:33 sri another stolen example ;p https://github.com/kraih/minion/commit/fd995d7441e83c248219f595caa5b2e2aa3e89d1
11:36 jontaylor joined #mojo
11:42 bpmedley sri : We currently have some custom logic for retrying jobs.  Visual inspection seems to indicate that the new attempt feature will reduce the amount of custom code we need.. :)
11:45 sri ok, that sounds like three times +1 so far for attempts
11:46 CandyAngel Yup
11:47 CandyAngel Hm, is there a simple way of holding onto a variable in case another request for the same page comes in for another format?
11:51 nic app->{cache}{myvar} = $val
11:53 nic (Using Mojo::Cache is slightly better)
12:02 neilhwatson joined #mojo
12:10 * sri wonders if $minion->reset actually serves a purpose anymore
12:11 nic isn't it useful in testing?
12:11 sri for testing we mostly have better solutions, with temporary schemas in postgres, temp files for sqlite/file backends.. and so on
12:13 sri nic: or do you have a scenario where you actually need to reset the queue in the middle of a bunch of tests?
12:14 * nic is in an office with a strict dress code and has to keep quietly sweeping sapphire glitter off his desk
12:15 nic sri: no, I was just querying my previous understanding, having read and played with the existing unit tests (as they were about a month ago)
12:15 sri in the beginning i used $minion->reset at the end of my test files
12:16 sri but that doesn't work if you run multiple tests parallel
12:16 nic I see
12:16 sri so, we now have better ways to handle that
12:16 nic yeah, and it could be viewed as dangerous to leave it in
12:16 sri like every test getting its own postgresql schema, which gets dropped recursively at the end
12:17 sri it's nice being able to run all tests in parallel, even those messing with a database
12:24 * CandyAngel holds hand up
12:24 CandyAngel I reset minion queues
12:24 sri CandyAngel: why?
12:25 CandyAngel I'm not using the test suite yet, I guess
12:26 CandyAngel I test by actually using it >.<
12:26 nic CandyAngel: which backend are you currently using?  (just curious)
12:26 vervain joined #mojo
12:27 CandyAngel nic: I still use File
12:27 nic k
12:27 sri so you could just "rm my.db"
12:27 sri btw. the File backend doesn't exist anymore
12:27 CandyAngel -.-
12:29 sri anyway, i think that's actually a reasonable use case for $minion->reset
12:30 sri messing around with one-liners and the like against a real database backend
12:31 sri CandyAngel: the file backend is up for adoption, if you want it
12:32 sri it hasn't learned about attempts yet though
12:33 nic The Storable backend sounded more interesting
12:33 sri well, feel free to adopt that too
12:35 * sri will release minion 4.0 later today if no reasonable new feature requests pop up
12:37 nic sri: yes please, I'd like to adopt Storable.  How do I go about doing that?  Is it just 'fork' on github?
12:38 sri i think minion 0.38 was the last version using the Storable backend
12:39 sri just rip out the stuff you need and give it a new name like Minion::Backend::Storable
12:39 nic k
12:39 sri some stuff will be outdated though
12:42 nic yes, 0.38, got it.  thanks
12:43 nic it'll be a v useful learning exercise, maybe tomorrow
12:56 kes joined #mojo
12:58 kes May anyone give a link to guide how to display table data in *.ep template?
12:58 CandyAngel Well if the File backend is gone, I can take something off my TODO
12:58 CandyAngel Silver lining and all that :P
12:59 nic kes: *.html.ep is html+embedded perl
12:59 nic do you mean how do you generate html inside a perl loop?
13:00 kes ok. thank you
13:00 CandyAngel Not releasing a tool that crawls filesystems that requires a database server, that's just ridiculous
13:03 vervain joined #mojo
13:03 nic kes: in my controller I fetch a result set and pass it to 'render' as a stashed arg.  In my template I have "% for (@{stash('rs') // []}) {"
13:04 nic and then in the enclosed html I can embed "<%== $_->{mycol} // '' %>" and so on
13:06 nic (whether you need/want '=' or '==' depends on your data)
13:06 jberger kes: see also Mojolicious::Plugin::ReplyTable
13:07 jberger Either to use or to see the code
13:09 kes ok
13:12 sri CandyAngel: there's Minion::Backend::SQLite
13:13 CandyAngel Is that new?
13:14 CandyAngel Must be, because I would have used that instead of File :P
13:14 sri https://github.com/kraih/minion/blob/master/Changes#L23
13:16 * CandyAngel readds to TODO then
13:16 CandyAngel :P
13:17 ajr_ joined #mojo
13:28 aborazmeh joined #mojo
13:32 vervain joined #mojo
13:42 shadowpaste "kes" at 217.168.150.38 pasted "Is it expected error message?" (27 lines) at http://paste.scsys.co.uk/501214
13:42 mattastrophe joined #mojo
13:43 gryphon joined #mojo
14:06 vervain joined #mojo
14:07 ichi joined #mojo
14:15 Akron joined #mojo
14:20 CandyAngel My royale parser is working :D (for daily royale stuffs)
14:22 kaare joined #mojo
14:38 cpan_mojo JSON-Validator-0.60 by ASHIMEMA https://metacpan.org/release/ASHIMEMA/JSON-Validator-0.60
14:42 * sri still thinks some kinda way to list currently active queues in minion would be nice
14:44 nic is that: a queue that currently has an active or pending job?  or a queue that has had such a job within a specified timeframe?
14:45 sri all queues we can find in the minion_jobs table
14:45 sri so a ui could show a list/drop-down
14:45 nic might be clearer to drop the word 'active' in that context
14:47 sri actually scratch that, i have no idea how it should work at all
14:47 nic In that context you probably want all queues in the domain, regardless of whether they've ever had a job
14:48 nic in which case it's only the app that knows, not Minion
14:48 CandyAngel Until some crazy person lets the end users specify the queue name to put a job into
14:48 nic eg If someone has done a 'reset' you don't want the list of queues to become empty
14:48 CandyAngel >.>
14:48 CandyAngel <.<
14:49 sri perhaps the queue name will always work more like a tag
14:50 nic I suppose minion could have an attribute that means 'expected queues' and then there could be an option to reject jobs not in the 'expected' list
14:51 nic yeah, I suppose it makes sense to store that in Minion rather than the app, even if Minion does almost nothing with it
15:01 punter joined #mojo
15:04 jberger you certainly don't want to require that a queue name be predeclared
15:05 jberger because of the idea of a "parkinglot" queue which intentionally has no worker
15:05 jberger sri: I think you're right, its basically a tag
15:10 cpan_mojo Mojolicious-Plugin-Captcha-reCAPTCHA-0.05 by HJANSEN https://metacpan.org/release/HJANSEN/Mojolicious-Plugin-Captcha-reCAPTCHA-0.05
15:10 marty joined #mojo
15:11 * sri is also rather hesitant to add more stuff to $minion->stats, even though he has some fun ideas
15:12 sri like, it could be nice to have the number of currently delayed jobs
15:13 espent joined #mojo
15:13 sri so you don't just see 1000 inactive jobs, but also that 999 of those are delayed
15:14 * sri just added a list of available stats http://mojolicio.us/perldoc/Minion#stats
15:15 abra joined #mojo
15:23 sri or maybe even uptime, based on oldest worker or job
15:24 sri you can then make more statistics based on the time, like jobs per minute enqueued/processed
15:26 sri oh, maybe uptime of the postgresql server :)
15:29 sri ah
15:29 sri select extract(epoch from now()) - extract(epoch from pg_postmaster_start_time());
15:29 CIAvash joined #mojo
15:53 vervain joined #mojo
16:05 Akron joined #mojo
16:08 jb360 joined #mojo
16:11 vervain joined #mojo
16:14 melo joined #mojo
16:25 Akron Design question: I've got multiple plugins, that should add complex html to a certain section of a page (e.g. <= content_for 'user_actions' %>). How can I trigger loading the snippets from the plugins each time content_for is called?
16:26 mattastrophe joined #mojo
16:29 ajr_ joined #mojo
16:40 mattastrophe1 joined #mojo
16:43 mattastrophe2 joined #mojo
16:48 mattastrophe joined #mojo
17:01 irqq_ joined #mojo
17:06 vervain joined #mojo
17:13 vervain joined #mojo
17:15 hernan605 joined #mojo
17:15 cpan_mojo Minion-4.0 by SRI https://metacpan.org/release/SRI/Minion-4.0
17:20 sri allright, no more new features this year ;p
17:23 * batman wonder when the minion backends will catch up
17:25 sri Grinnz has been pretty fast with Minion::Backend::SQLite, bpmedley not so much with Minion::Backend::mysql
17:25 sri the mongodb backends seemed abandoned last i looked
17:25 batman ok
17:26 sri can't actually think of that many new features that would require backend changes
17:27 sri aside from more stats and list_jobs options
17:27 Grinnz_ server uptime wouldnt make any sense in SQLite anyway :P
17:28 sri most commonly requested features right now are job dependencies, recurring jobs, and throttling
17:28 sri and i think all those can already be done with workarounds
17:29 odc looks like the mongodb backend was updated last week :)
17:29 sri job dependencies and recurring jobs by enqueing followup jobs from inside a job, and throttling with dedicated queues that lead to a limited number of workers
17:32 sri ./myapp.pl minion worker -j 3 -q only_three_at_a_time
18:10 absolut_todd joined #mojo
18:14 eitz joined #mojo
18:17 stigo joined #mojo
18:19 eitz Hello. Is it possible to use a static file as the root of a Mojo lite application? Or make a config setting index.html as the default file to load when accessing a folder?
18:20 batman eitz: get '/' => sub { shift->reply->static('index.html') };
18:23 eitz Thanks, batman!
18:26 Mattjes joined #mojo
18:26 Mattjes Hello
18:27 Mattjes If you have a Spider-Script (with Mojo and Mechanize) even wants to run per hour, which goes exclusively with a cron or is the other way possible?
18:28 Phil21 cron seems simplest to implement, or you could wrap your script in a sleep loop if cron is inconvenient for whatever reason
18:28 Phil21 I'm a fan of KISS :)
18:29 Mattjes thank you!
18:29 Mattjes What is KISS?
18:29 Phil21 Keep It Simple Stupid
18:29 Mattjes ah - ok ;-)
18:31 pink_mist https://en.wikipedia.org/wiki/KISS_principle it's fairly widely known, Mattjes =) I suggest you take a look at it; it has many fans =)
18:31 mattastrophe1 joined #mojo
18:32 Mattjes thank you
18:34 Mattjes i am a beginner ... not a coder-profi
18:35 Mattjes i must ask often ;-)
18:36 sri there's also job queues like minion ;) http://mojolicio.us/perldoc/Minion
18:47 Mattjes sri: you write Minion in your spare time too?
18:53 PryMar56 joined #mojo
18:57 trone joined #mojo
19:03 ichi joined #mojo
19:12 sivoais joined #mojo
19:18 fitnerd joined #mojo
19:21 ZoffixW joined #mojo
19:23 ZoffixW Is it possible to redirect to another route handler without issuing an HTTP redirect? Basically, I want /q/foobar and /?q=foobar behave the same. I'd just set up the / route handler and I'm hoping to somehow redirect to it from /q/foobar
19:23 Grinnz have both call a common method in a module or a common helper method
19:24 ZoffixW That's a plan. Thanks. Grinnz++
19:24 sivoais joined #mojo
19:38 val joined #mojo
19:50 bwf joined #mojo
20:28 * sri guesses that everobody is now playing fallout 4
20:29 Grinnz_ not for 8.5 more hours D:
20:29 lb mine says 5
20:29 * lb will not be playing in 5 hours, or 6 or 18
20:35 * genio should try to get into gaming one day
20:35 genio If it isn't Zelda/Metroid, I haven't played it
20:36 jberger <3 cities skylines lately
20:36 jberger tried ARK, my computer can barely handle it
20:36 buu genio: Games have advanced a bit since then
20:38 * Grinnz_ has been playing assassin's creed unity, arkham knight, fallout new vegas, witcher 2 (in preparation for 3), ...
20:38 * Grinnz_ GTA V...
20:39 buu I can't play games any more unless they involve another person and end in less than an hour
20:39 buu I feel old
20:39 sri last big game i played was the tomb raider reboot i believe, which was really good
20:40 Grinnz_ the last games i played that involved other people were MMOs
20:40 Grinnz_ otherwise i like single player... no people rushing me through content >:(
20:40 * sri heads off for a round of hearthstone :)
21:02 espent_ joined #mojo
21:15 ZoffixW joined #mojo
21:40 borkur joined #mojo
22:01 cosimo_ joined #mojo
22:19 cosimo joined #mojo
23:13 ZoffixW Mojo docs spoiled me :) I feel so lost in DBIx::Class docs now :P
23:15 mattastrophe joined #mojo
23:16 jzawodn joined #mojo
23:24 cfedde (in cleanup) Can't call method "tx" on an undefined value at /home/cfedde/.plenv/versions/5.22.0/lib/perl5/site_perl/5.22.0/Mojo/UserAgent.pm line 42 during global destruction.  I get the above error when I run this code from the command line but not in the debugger.  Looks like I'm exiting before all the distruction is done?  Can I trigger a stack dump so I can figure out where I'm exiting early?
23:28 disputin joined #mojo
23:34 disputin joined #mojo
23:35 ajr_ joined #mojo
23:38 FatalNIX ZoffixW, hehe. The Java of DBI? 5GB of ram, for a table schema :>
23:38 mattastrophe1 joined #mojo
23:39 ZoffixW RAM is cheap
23:40 ZoffixW See: only 25K: http://www.ramtrucks.com/en/lineup/?app=bmo
23:40 ZoffixW badum ts
23:45 CandyAngel Mojo::Loader is awesome o.o
23:50 CandyAngel I'm using my 3 favourite projects right now :P
23:50 CandyAngel I shall call it.. Mojonnex MD
23:50 Grinnz_ PHP, mysql, and apache?
23:50 * Grinnz_ runs
23:51 * jberger realizes that he no longer uses any of the LAMP stack (as I first understood it)
23:51 FatalNIX Zoffix, well, with http://downloadmoreram.com you don't need to pay for RAM anymore.
23:51 CandyAngel Bad Grinnz_
23:52 Grinnz_ LNPP > LAMP
23:52 Grinnz_ the last P being perl of course
23:52 FatalNIX FNSP
23:52 Grinnz_ ahem.. i mean prolog
23:52 Grinnz_ of course
23:52 FatalNIX FreeBSD Nginx SQL of your choice Perl
23:59 FatalNIX What the hell? I just got a huge static jolt from a potato
23:59 cfedde as long as your choice is not MySQL.

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