--- Log opened Tue Aug 01 00:00:20 2006 01:03 < Wombert> v-dogg! 01:04 < Wombert> arrrgghhhh 01:04 < Wombert> you broke trunk :| 01:04 < Wombert> at least the tests 01:10 < CIA-6> david * r855 /trunk/ (3 files in 3 dirs): changed sample app, code templates and testing to supply names as arguments when initializing environments and contexts 01:11 < CIA-6> david * r856 /trunk/src/config/AgaviLoggingConfigHandler.class.php: use context directly instead of getter method in LoggingConfigHandler so test doesn't break 01:25 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"] 01:25 < CIA-6> david * r857 /trunk/tests2/ (4 files in 2 dirs): tests are working again 01:32 < CIA-6> david * r858 /trunk/tests2/ (19 files in 14 dirs): changed tests to use the 'test' context everywhere 01:43 < CIA-6> david * r859 /trunk/tests2/config/LoggingConfigHandlerTest.php: fixed LoggingConfigHandler tests 01:57 < CIA-6> david * r860 /trunk/ (4 files in 3 dirs): made passing of env/context names to Agavi::bootstrap()/AgaviContext::getInstance() mandatory. as an alternative, set 'core.environment'/'core.default_context' directives 02:17 * Macca ++ 03:13 < Macca> [Tue Aug 01 10:13:27 2006] [error] [client 203.206.205.212] PHP Fatal error: You must supply an environment name to AgaviContext::getInstance() or set the name of the default environment to be used in the configuration directive "core.default_context". in /var/www/agavi/src/core/AgaviContext.class.php on line 206, referer: https://development.maccasoft.com.au/lld/direct/www/client 03:13 < Macca> how do i fix? :D 03:15 < Wombert> getInstance('web') 03:37 -!- Wombert [n=Wombert@dslb-088-064-006-044.pools.arcor-ip.net] has quit [] 03:45 < Macca> right 06:21 -!- shoan_ [n=shoan@59.92.145.138] has joined #agavi 06:35 -!- shoan [n=shoan@59.92.158.11] has quit [Read error: 110 (Connection timed out)] 07:17 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 07:41 -!- raidman|Work [n=raidman@62.60.194.3] has joined #agavi 08:08 < v-dogg> woot 08:11 -!- impl [n=impl@httpcraft/php/nforbes] has quit ["Leaving"] 08:31 < Macca> i'm not meannt to forward in the action, am i? 08:31 < v-dogg> no 08:31 < v-dogg> you're not meant to forward anywhere anymore :) 08:32 < v-dogg> return array('module', 'action') from the view and you'll be forwarded 08:32 < v-dogg> and you probably should unset the template too 08:33 < v-dogg> don't know if that's necesarry 09:24 < CIA-6> v-dogg * r861 /trunk/ (3 files in 3 dirs): Added AgaviContext to AgaviLayout, fixed AgaviLoggingConfigHandler to initialize layouts correctly and also remembered to fix tests this time. 09:31 < CIA-6> v-dogg * r862 /trunk/src/logging/ (AgaviAppender.class.php AgaviLayout.class.php): To make commit statistics look better I make at least one typo per commit.... 09:44 < v-dogg> anyone has anything against it if I break AgaviLogger b/c ? 09:49 -!- shoan [n=shoan@59.92.145.138] has joined #agavi 09:55 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"] 10:02 -!- eremit [n=eremit@p5490D3B4.dip.t-dialin.net] has joined #agavi 10:02 < eremit> heya ... 10:02 < v-dogg> hey 10:02 -!- shoan_ [n=shoan@59.92.145.138] has quit [Read error: 110 (Connection timed out)] 10:02 < eremit> sorry for being absent such a long time ^^ ... 10:03 < eremit> i was in a rehabilitation hospital because of a heavy rsi attack :/ 10:05 -!- Wombert [n=Wombert@dslb-088-064-006-044.pools.arcor-ip.net] has joined #agavi 10:05 < eremit> v-dogg: anything really important happened the past 3 month ? 10:05 < eremit> heya Wombert 10:06 < v-dogg> eremit: not much 10:06 < v-dogg> just about everything has changed :) 10:06 < eremit> lol ... how stable is trunk atm ? 10:06 < v-dogg> pretty good 10:06 < eremit> kk... i'll check it out and test ... 10:06 < Wombert> validation is still in a branch 10:06 < v-dogg> check out the sample app 10:07 < Wombert> almost finished though 10:07 < Wombert> the rest is done 10:07 < Wombert> only CachingExecutionFilter is broken right now 10:07 -!- eremit [n=eremit@p5490D3B4.dip.t-dialin.net] has quit [Read error: 104 (Connection reset by peer)] 10:07 < v-dogg> and I'm about to break Logger :) 10:07 < Wombert> btw I just made it yesterday 10:07 < Wombert> 231 commits ;) 10:07 < v-dogg> yeah, saw that :) 10:07 < Wombert> yeah I fixed some tests and such... 10:08 < v-dogg> sorry about that :) 10:08 < Wombert> I say we rewrite (and I mean rewrite) tests between 0.11 and 1.0 10:08 < Wombert> no problem 10:08 -!- eremit [n=eremit@p5490D3B4.dip.t-dialin.net] has joined #agavi 10:08 < Wombert> the tests are _so_ messy 10:08 < eremit> dmn ... this computer is quite unstable :/ 10:10 < v-dogg> Wombert: is it ok if I: (a) rename Logger::priority to Logger::level (and everything that goes with that) 10:10 < v-dogg> (b) change priorities to bitwise levels 10:12 * Macca wants to do some dev :( 10:12 < Wombert> ? 10:12 < Macca> on agavi 10:12 * Macca feels left out 10:12 < Wombert> aaah 10:12 < Wombert> I read "Macca wants some dev" 10:12 < Wombert> +s 10:12 < Macca> please mr wombat sir let me do some devel! 10:12 < Macca> haha 10:12 * Wombert scratches head 10:12 < Wombert> well 10:12 < Wombert> what else is there 10:13 < Macca> the broken form filter? 10:13 < Macca> :) 10:13 < Wombert> v-dogg: hmmmh 10:13 < Wombert> v-dogg: yeah right change that 10:13 < Wombert> Macca: it's not broken!? 10:13 * Macca shrugs 10:13 < Macca> i just don't know how to use it 10:13 < Wombert> look at the sample 10:13 < Wombert> app 10:13 < Macca> that app sucks :( 10:13 < Wombert> put it in the global filters list, and you're done 10:14 < Wombert> also, tell it to run on "write" method 10:14 < Wombert> just copy that from sample app 10:15 < Macca> so what dev can i do? :D 10:15 -!- AgaviBot [n=dzuelke@dslb-088-064-006-044.pools.arcor-ip.net] has joined #agavi 10:16 < Wombert> !seeen eremit 10:16 < Wombert> !seen eremit 10:16 < AgaviBot> Wombert: I'm sorry, but eremit never spoke a single word here, he is online right now though 10:17 < shoan> !seen Wombert 10:17 < AgaviBot> shoan: Wombert is online right now, last message on Tue Aug 1 09:16:16 2006 saying "!seen eremit" (joined Fri Jul 7 21:04:19 2006 10:17 < shoan> ;) 10:19 < Wombert> !info 10:19 < AgaviBot> Wombert: I'm AgaviBot version 0.3, powered by Agavi :) My IRC Bot components and the Web Interface to the Logs share quite some code, thanks to Agavi's flexibility and clean structure! 10:20 < Macca> how is it powered by agavi? 10:21 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 10:21 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/862: Changeset [862]: To make commit statistics look better I make at least one typo per ... 10:21 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/861: Changeset [861]: Added AgaviContext to AgaviLayout, fixed AgaviLoggingConfigHandler to ... 10:21 < AgaviBot> 3) http://trac.agavi.org/trac.cgi/ticket/297: Ticket #297 (defect) created by v-dogg 10:21 < AgaviBot> 4) http://trac.agavi.org/trac.cgi/changeset/860: Changeset [860]: made passing of env/context names to ... 10:21 < AgaviBot> 5) http://trac.agavi.org/trac.cgi/changeset/859: Changeset [859]: fixed LoggingConfigHandler tests 10:21 < Macca> tf? 10:21 < Wombert> Macca: it's written in agavi 10:21 < Macca> you mean PHP? 10:21 < Wombert> PHP, and Agavi 10:22 < Wombert> the routing parses the messages 10:22 < Macca> well it'd be written in PHP for agavi? :) 10:22 < Wombert> there are routes setup like "!seen " 10:22 < Wombert> then it dispatches SeenAction 10:22 < Wombert> !seen Macca 10:22 < AgaviBot> Wombert: Macca is online right now, last message on Tue Aug 1 09:22:06 2006 saying "well it'd be written in PHP for agavi? :)" (joined Sun Jul 16 23:18:56 2006 10:22 < Wombert> the nice thing is that I can use SeenAction from the web interface, too 10:22 < Macca> oh noes 10:22 < Macca> there's no closing bracket! 10:23 < Macca> where's the sauce for that? 10:23 < Wombert> on my hard drive ;) 10:23 < Macca> i see. 10:23 < Wombert> it'll be a sample app once I finished it 10:24 < Macca> ooooooOOOOOOOoooooooo 10:24 < Macca> oo 10:24 < Macca> i like 2.0 10:24 < Macca> Get rid of decorators and slots and invent something biblically good to replace them 10:24 < Macca> :D 10:25 < Macca> Make it PHP6 only .. that seems a little odd tho 10:25 < Macca> :S 10:25 < Wombert> !seen Macca 10:25 < AgaviBot> Wombert: Macca is online right now, last message on Tue Aug 1 09:25:03 2006 saying ":S" (joined Sun Jul 16 23:18:56 2006) 10:25 < Macca> better. 10:25 < Macca> :) 10:27 < Wombert> well 10:27 < Wombert> we'll see what's up with that PHP6 thing then 10:27 < Macca> :) 10:37 < Macca> do you use many pear things? 10:45 < v-dogg> I try to avoid them 10:58 < ttj> Any specific reason? 10:59 < ttj> Technical or political? 11:00 < v-dogg> php4 mainly 11:10 -!- shoan [n=shoan@59.92.145.138] has quit [] 11:14 < v-dogg> Wombert and others 11:14 < v-dogg> log message _level_ or _type? 11:14 < v-dogg> FATAL, ERROR, WARN, INFO and DEBUG 11:17 < Wombert> iirc, you wanted to call it level, right? 11:17 < Wombert> leave large gaps between them, too, so users can add their own 11:17 < Wombert> bbiab 11:19 < v-dogg> I don't think we need gaps 11:20 < v-dogg> class AgaviLogger { const FATAL = 1; const ERROR = 2; const WARN = 4; const INFO = 8; const DEBUG = 16; const ALL = 31; 11:22 < v-dogg> ALL is the only thing that's a bit problematic 11:22 < v-dogg> because if you wanted something else, you could define MyLogger::FOO = 64 11:23 < v-dogg> and set level = AgaviLogger::INFO | MyLogger::FOO 11:23 < v-dogg> of course we could set ALL to be, say, 4096 11:24 < v-dogg> then you would have 7 "gaps" 11:25 < v-dogg> err, ALL = 4095 11:26 < v-dogg> yeah, that works 11:28 < v-dogg> lunch time 11:31 < v-dogg> Has anyone just cause to halt these log levels, speak now or forever hold your piece :) 11:32 < v-dogg> or before I get back from lunch 11:41 < Wombert> ? 11:42 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit [Read error: 110 (Connection timed out)] 11:42 < eremit> uh... the new router is nice ... 11:43 < v-dogg> Wombert: ?? 11:43 < Wombert> I didn't get that ahm 11:43 < Wombert> halt blah thing 11:44 < Wombert> ALL is a problem tho, that's right 11:44 < Wombert> why not make it 2^31 or whatever 11:45 < v-dogg> that was a jokethingy, you know when you get married and the priest ask "has anyone a just cause to halt this union, speak now..." :) 11:45 < Wombert> oO 11:46 < Wombert> lol :> 11:46 < v-dogg> but gotta run 11:46 < v-dogg> bbl 11:47 < Wombert> k 11:47 < eremit> uh ... maybe i'm some kind of stupid ... but how to setup a generic route like // ? 11:50 < Wombert> 11:50 < Wombert> but why do you want that? 11:50 < Wombert> or do you want 11:50 < Wombert> foo.com/Default/ShowProducts/id/123 11:50 < Wombert> ? 11:51 < eremit> no i just want for testing and development purposes things like foo.com/Guest/Register foo.com/User/Logout etc 11:52 < eremit> there are no get params used in the test setup ... 11:52 < Wombert> well 11:52 < Wombert> adding a new route is fairly easy 11:52 < eremit> ah your route does the trick ...ty 11:52 < Wombert> eremit: wait 11:53 < Wombert> if you have that route 11:53 < Wombert> you have to give it a name, too 11:53 < Wombert> and then do 11:53 < Wombert> $r->gen('blah', array('module' => 'Default', 'action' => 'Register')); 11:53 < Wombert> assuming that route's name is foo 11:53 < Wombert> err 11:53 < Wombert> blah 11:53 < Wombert> if you create a route 11:53 < Wombert> you do 11:54 < Wombert> $r->gen('guest.register'); or so 11:54 < Wombert> I'd do proper routes right away 11:54 < Wombert> you can keep the route names 11:54 < Wombert> but change the patterns 11:54 < Wombert> and it generates the correct stuff for you 11:54 < Wombert> always 11:54 < eremit> nice ^^... 11:54 < Wombert> if you want to have simple urls for testing, don't use that at all 11:54 < eremit> i really appreciate the work on this ... *happy dancing* 11:54 < Wombert> because you could just as well do index.php?module=Default&action=ajda 11:55 < eremit> which doesn't work for me 11:56 < eremit> if i try that, i get a 404 11:56 < Wombert> did you enable the .htacces? 11:57 < eremit> *thinking sound* 11:57 < Wombert> see ;) 11:57 < Wombert> don't do that 11:57 < Wombert> because it will rewrite then 11:57 < Wombert> leave it disabled 11:57 < eremit> there's no htaccess and mod rewrite is disabled ... atm 12:01 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has joined #agavi 12:02 < digitarald> morning 12:06 < Wombert> hi 12:06 < Wombert> eremit: set use_routing to false then in settings.xml 12:07 < eremit> ah that's the trick ... 12:07 < eremit> ty ... 12:07 < eremit> missed that one ^^ 12:09 < Wombert> eremit: your routing.xml will still be parsed 12:09 < Wombert> and you can still gen() routes 12:10 < Wombert> the urls will then be like index.php?module=Default&action=ViewProduct&id=312 12:10 < eremit> really nice ... 12:36 -!- shoan [n=shoan@61.95.206.107] has joined #agavi 12:37 < shoan> Wombert: is the trunk stable again? i thought there were some merge issues earlier? 12:38 < Wombert> no 12:38 < Wombert> no issues 12:38 < shoan> cool 12:38 < shoan> updating :) 12:38 < Wombert> some changes got lost when I merged changes from trunk to branches/dominik-validation_mods 12:38 < Wombert> trunk is fine ;) 12:51 -!- raidman|Work [n=raidman@unaffiliated/raidman] has joined #agavi 13:00 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit [Read error: 60 (Operation timed out)] 13:14 < v-dogg> howdy 13:16 < Wombert> http://blog.bitxtender.com/ 13:16 < Wombert> :) 13:17 < Wombert> specifically: http://blog.bitxtender.com/2006/08/01/agavi-project-activity/ 13:19 < eremit> Wombert, at which moment is the renderer created? 13:19 < Wombert> after View::execute() returns 13:19 < Wombert> you can return a renderer instance from execute() if you like 13:20 < Wombert> but usually, that's not necessary 13:20 < Wombert> output types are a better idea 13:20 < eremit> hmm :/ ... 13:20 < Wombert> the last occasion to set an output type is in View::initialize() 13:20 < Wombert> but you typically set it in the routing 13:20 < eremit> i need to modify the smarty renderer instance 13:20 < Wombert> you could use a rendering filter 13:20 < Wombert> what do you need to modify 13:21 < eremit> i want to add customer plugins ... 13:21 < Wombert> are they always the same? 13:21 < Wombert> create a MySmartyRenderer extends AgaviSmartyRenderer 13:21 < eremit> basically yes .. it's a bunch of about 20 custom tags ... 13:21 < Wombert> and register them there 13:21 < Wombert> in getEngine 13:21 < Wombert> that's how I would do it 13:22 < Wombert> and 13:22 < Wombert> hey 13:22 < eremit> ok ... i only thought there could be a better way ... 13:22 < Wombert> surprise ;) 13:22 < Wombert> now that can be nested 13:22 < Wombert> you can specify the plugins to be loaded in the config file 13:23 < eremit> oO *rethings the way i understood the configs* 13:23 < Wombert> you define the renderer in output_types.xml 13:23 < Wombert> the renderer itself may have init parameters 13:23 < Wombert> for instance, you can specify the var name where the template vars are stored 13:23 < Wombert> default is $template 13:23 < Wombert> you can change that to $t 13:24 < Wombert> or you can tell the renderer to extract the vars 13:24 < Wombert> now you could have another parameter, called "plugins" 13:24 < Wombert> and then 13:24 < Wombert> in there 13:24 < Wombert> more parameters 13:24 < Wombert> each specifying plugin name and method 13:24 < Wombert> that's the way it works in smarty, iirc, right? 13:25 < eremit> yea 13:25 < Wombert> is that register_function? 13:25 < eremit> yes 13:26 < Wombert> I'm thinking about addind that to the default smarty renderer 13:26 < Wombert> -d+g 13:26 < eremit> hmmm ... 13:26 < Wombert> not that easy though, because it could be callbacks 13:26 < Wombert> to an object 13:27 < eremit> yeah thats the main problem ... 13:27 * Wombert scratches his head 13:27 < eremit> thats why i normally only define an additional plugin dir 13:27 < Wombert> I think I could work that out 13:27 < Wombert> all I need is dominik to change parameters to allow numerical indices 13:27 < Wombert> hang on a second 13:29 < Wombert> yeah I think it's possible 13:30 < Wombert> how do you typically load and organize your plugins? 13:30 < eremit> smahed into one directory like the plugins dir of smarty ... 13:30 < Wombert> ah 13:30 < eremit> one file per function ... 13:30 < Wombert> does it load them automatically? 13:31 < eremit> it should ... 13:31 < Wombert> long time since I used smarty ;) 13:31 < Wombert> what's the naming convention 13:31 < eremit> i'm currently checking my sources about this ... 13:31 < Wombert> I'll sit down later and put SmartyRenderer on steroids a bit 13:31 < eremit> haven't used smarty myself for some type 13:32 < eremit> mom ... i'll check that ^^ 13:34 < v-dogg> Wombert: 13:34 < v-dogg> Strict standards: Declaration of TRTestSampleView::initialize() should be compatible with that of AgaviView::initialize() 13:34 < Wombert> ty 13:34 < v-dogg> should I fix these or is dominik already on to it? 13:34 < eremit> hmm basically you only need to remove the hardcoded plugins_local directory ... in the renderer and put a config able path in... 13:37 -!- raidman|Work [n=raidman@unaffiliated/raidman] has joined #agavi 13:48 < Wombert> I can fix it, or you, don't care 13:52 < v-dogg> I'll do it 13:52 < v-dogg> logging is almost done 13:52 < v-dogg> I think :) 13:53 < Wombert> very good work man 13:53 < Wombert> very good 14:07 < v-dogg> hmph 14:08 < v-dogg> Warning: file_get_contents(\log1391.tmp): failed to open stream: No such file or directory in C:\workspace\agavi-trunk\tests2\logging\LoggerManagerTest.php on line 74 14:09 < v-dogg> aa... my mistake 14:12 < v-dogg> these tests are a bit hard to figure out :) 14:16 < v-dogg> but they sure are nice to have when you change something 14:17 < Wombert> as I said 14:18 < Wombert> we'll have to re-write them for 1.0 14:18 < v-dogg> yep 14:23 -!- benny`work [n=benny@p54AC3CD4.dip0.t-ipconnect.de] has joined #agavi 14:28 < v-dogg> heads up 14:28 < v-dogg> I'm going to break trunk 14:28 < v-dogg> kidding... :) 14:28 < v-dogg> but I am going to break b/c 14:29 < v-dogg> does agavi.org use logging? 14:34 < eremit> oO, there's an error in WebRouting caused by usage of an uninstanciated variable ... 14:36 < v-dogg> eremit: where? 14:36 < eremit> WebRouting:210 ... i don't understand the whole routine starting at 191 not ... 14:37 < eremit> $defaults gets initialized using array(), $myDefaults gets not initialized, none of them both are used before ... but both are merged ... 14:38 < eremit> why merging $defaults and $myDefaults if $defaults is definately empty ... an mydefaults isnt event instanciated if the if's don't match ... 14:42 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 14:42 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/297: Ticket #297 (defect) closed by v-dogg 14:42 < eremit> anyway i will create a patch with all my changes later... 14:43 < eremit> is there any kind of registry in agavi to retrieve instanciated objects from ? 14:43 < CIA-6> v-dogg * r863 /trunk/ (9 files in 6 dirs): 14:43 < CIA-6> Re-wrote logging priority levels. They are now treated as n-bit integers so you 14:43 < CIA-6> can define e.g. AgaviLogger::ALL ^ AgaviLogger::DEBUG to log everything but 14:43 < CIA-6> debug messages, or AgaviLogger::INFO | AgaviLogger::WARN to log only info and 14:43 < CIA-6> warning messages. AgaviLogger::ALL is defined as 2^32-1 so you can define level 14:43 < eremit> i remember a discussion about it ... 14:43 < CIA-6> constants from 1 to 2^31. NOTE! This will break backwards compatibility! Sorry 14:43 < CIA-6> about that. closes #296. 14:45 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 14:45 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/296: Ticket #296 (enhancement) closed by v-dogg 14:45 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/863: Changeset [863]: Re-wrote logging priority levels. They are now treated as n-bit integers ... 14:51 < shoan> v-dogg: nice work with loggers :) 14:51 < shoan> will check it out tonight 14:51 < v-dogg> thanks, hope it works :) 14:51 < shoan> didn't you write any test cases? ;) 14:52 < v-dogg> there were a few already and I extended those a bit 14:53 < shoan> btw is there a way to edit a svn commit message? 14:54 < shoan> i just committed saying that it does not work 100% instead of saying that it did :| 14:54 < v-dogg> not that I know of 15:05 < eremit> shoan there is ... 15:05 < shoan> how? revert and commit again? 15:05 < Wombert> no 15:05 < Wombert> you can edit commit messages on the server, afaik 15:05 < eremit> i'm just searching for it ... 15:06 < shoan> ok 15:06 < Wombert> looking ;) 15:06 < shoan> its not really critical, but I prefer not looking stupid :D 15:06 < Wombert> http://subversion.tigris.org/faq.html#change-log-msg 15:07 < eremit> uhm wombert was faster ^^ 15:08 < shoan> :) 15:08 < shoan> thanks guys 15:11 < v-dogg> I can't get rid of all strict standars errors in tests 15:11 < v-dogg> without multiple inheritance :) 15:16 < CIA-6> v-dogg * r864 /trunk/tests2/ (3 files in 2 dirs): Strict standards fixes. 15:16 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 15:16 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/864: Changeset [864]: Strict standards fixes. 15:32 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit [Read error: 104 (Connection reset by peer)] 15:34 -!- raidman|Work [n=raidman@62.60.194.3] has joined #agavi 15:50 < Wombert> so... which tickets don't need to be finished for 0.11b1? 15:53 < v-dogg> wooot? waht this be: http://trac.agavi.org/trac.cgi/ticket/286 15:53 < Wombert> #65, #183, #78 (maybe), #107, #137, #184 (?), #34 15:54 < Wombert> http://forum.agavi.org/viewtopic.php?id=106 15:55 < v-dogg> lemme see... 15:55 < v-dogg> wtf... 15:55 < Wombert> does it work for you? 15:55 < v-dogg> no 15:55 < Wombert> I thought I'd push this back until we roll b1 15:55 < v-dogg> but what the f.. broke it 15:55 < Wombert> I have no idea 15:55 < Wombert> aaah hey 15:56 < Wombert> didn't we want to rename webapp to app? 15:56 < Wombert> and www to... pub? 15:56 < v-dogg> you did :) 15:57 < v-dogg> but yeah, we agreed that it's a go-go 15:57 < Wombert> so is "pub" good? 15:57 < Wombert> benny`work? digitarald? eremit? Macca? shoan? ttj? 15:57 < shoan> ? 15:57 < Wombert> rename "www" folder to... "pub" 15:57 < Wombert> purely cosmetic thing 15:57 < eremit> no 15:58 < Wombert> why :< 15:58 < shoan> no 15:58 < shoan> i like www and webapp :) 15:58 < Wombert> which are both wrong because you can build other things than web apps with agavi 15:58 < Wombert> also, you could just set core.app_dir and core.pub_dir to those 15:58 < Wombert> so you don't have to change everything 15:58 < Wombert> deal? 15:59 < shoan> +1 :) 15:59 < eremit> hehe ... i take back my no ... and say i dont care :P 15:59 < Wombert> is there anything else we forgot 15:59 < eremit> i'm changing most pathes anyway in my retail apps 15:59 < Wombert> aside from i18n 15:59 < Wombert> which we have a ticket for 16:01 < shoan> yay 16:01 < eremit> i havnt found anything yet ^^ 16:01 < Wombert> good 16:01 < Wombert> just to remind you guys 16:02 < Wombert> 0.11 betas are the last chance for you to step forward and say "there's something I want to do, but it's not possible with agavi" etc 16:02 < v-dogg> gaaah... 16:02 < v-dogg> wanna know who broke package building? 16:02 < Wombert> me? 16:02 < Wombert> :p 16:02 < v-dogg> yeah, you guessed it, PEAR did 16:02 < Wombert> ahahah 16:02 < Wombert> wtf 16:02 < Wombert> ah yeah I remember 16:03 < eremit> hmm i have a really really special issue , but i dont think it's solvable using agavi :P 16:03 < Wombert> back then I had to go back to an old package_fuck_pear_whatever_the_name_is version 16:03 < Wombert> to make it work 16:03 < v-dogg> release and api versions cannot have letters anymore 16:03 < Wombert> ahahahahaha 16:03 < Wombert> right 16:03 < Wombert> FUCK PEAR 16:03 < Wombert> seriously 16:03 < Wombert> such a bunch of morons 16:03 < v-dogg> so 0.11-DEV is now 0.11 16:03 < Wombert> they change stuff in between 2.1.1 and 2.1.2 16:03 < eremit> pear is crap ^^ 16:03 < Wombert> just break bc and... zomg 16:03 < Wombert> Z 16:03 < Wombert> O 16:03 < Wombert> M 16:03 < Wombert> G 16:04 < Wombert> aah no 16:04 < Wombert> v-dogg: wait 16:04 < Wombert> I think they only allow lowercase letters 16:04 < Wombert> but I could be wrong 16:04 < Wombert> I read something about this 16:04 < Wombert> not sure 16:04 < v-dogg> testing... 16:05 < v-dogg> Error: Version type is not a valid version (0.11-dev) 16:06 < CIA-6> dominik * r865 /trunk/src/config/ (AgaviConfigHandler.class.php xsd/default_tags.xsd): support for nested parameters without the name attribute (so the array created will be int indexed) 16:06 < Wombert> I read something about lowercase shit and all 16:06 < Wombert> I'm 100% sure 16:06 < Wombert> maybe it's the packages themselves or so 16:06 < v-dogg> 0.11dev works 16:07 < v-dogg> stupid piece of crap 16:07 < Wombert> indeed 16:07 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:07 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/865: Changeset [865]: support for nested parameters without the name attribute (so the array ... 16:08 < v-dogg> 0.11.DEV doesn't work 16:08 < Wombert> PEAR 16:08 < Wombert> SUCKS 16:08 < Wombert> MUST 16:08 < Wombert> SHOOT 16:08 < Wombert> SOMEONE 16:08 < Wombert> it just occured to me 16:08 < Wombert> you know 16:09 < Wombert> pear being shitty might be the #1 reason why php has such a "bad" image 16:09 < v-dogg> 0.11.dev doesn't work 16:09 < eremit> its one of the majors yes ^^ 16:09 < Wombert> besides the core dev's ignorance, of course 16:10 < eremit> ^^ --- that cant be called ignorance anymore 16:11 < v-dogg> 0.11.1b works... 16:11 < Wombert> b1? 16:11 < v-dogg> so what's teh current version? 16:11 < Wombert> I mean 16:11 < Wombert> WTF 16:11 < Wombert> current version is 0.11 dev 16:12 < Wombert> good question by the way 16:12 < Wombert> will we release 0.11.0 16:12 < Wombert> or 0.11 16:12 < Wombert> and then 0.11.1 16:12 < Wombert> (if necessary) 16:12 < v-dogg> Version type is not a valid version (0.11.b1) 16:12 < v-dogg> tragicomedy :) 16:13 < eremit> lol v-dogg try to find the regex ... and figure out what it allows instead of guessing ... 16:13 < eremit> :P 16:13 < v-dogg> can't be bothered :) 16:13 < v-dogg> we can use 0.11dev for now 16:14 < v-dogg> ok? 16:14 < Wombert> go 16:14 < eremit> preg_match('/^\d+(?:\.\d+)*(?:[a-zA-Z]+\d*)?$/' 16:17 < eremit> uh if i interpret this right 0.11.dev1 should work and 0.11-dev1 too 16:17 < v-dogg> nope 16:17 < v-dogg> - is not in [a-zA-Z] 16:17 < CIA-6> v-dogg * r866 /trunk/etc/phing/AgaviPackageTask.php: Fixed api and release version numbers to please Pear PackageManager (*sigh*). closes #286. 16:18 < eremit> ah right ... 16:19 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:19 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/286: Ticket #286 (defect) closed by v-dogg 16:19 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/866: Changeset [866]: Fixed api and release version numbers to please Pear PackageManager ... 16:25 < eremit> is there any way to tell agavi to autoload all files in a specific path ? 16:26 < eremit> or to add autoloading to all files in a path ? 16:26 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:26 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/292: Ticket #292 (enhancement) closed by dominik 16:26 < Wombert> no 16:26 < eremit> ty ^^ 16:27 < Wombert> use spl_autoload_register 16:27 < v-dogg> what's that sound... 16:27 < v-dogg> aaa... it's the lovely sound of tickets beeing closed 16:27 < v-dogg> *clickety click* 16:29 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:29 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/35: Ticket #35 (defect) closed by david 16:30 < eremit> Wombert, i know what i'm missing atm ... the possibility to change the include path of php using agavi config files :P 16:30 < Wombert> eremit: config.php is your friend 16:30 < Wombert> include path is none of agavi's business 16:31 < eremit> yeah but i would favor something like instead ;) 16:31 < eremit> but isnt necessary ^^ 16:32 < eremit> it's more a matter of taste ^^ 16:35 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:35 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/298: Ticket #298 (task) created by david 16:37 < eremit> oeh ... 16:37 < eremit> cant the Action::validate() method access the params anymore ? 16:38 < Wombert> uh 16:38 < Wombert> yes it can!? 16:38 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:38 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/299: Ticket #299 (enhancement) created by david 16:38 < Wombert> or did I break something there 16:38 < eremit> not using $this->getContext()->getRequest()->getParameter() 16:39 < Wombert> oh bloody hell 16:39 < Wombert> my pad 16:39 < Wombert> bad 16:39 < eremit> cant it be changed to be given to validate as $parameters ? like execute() gets'em ? 16:40 < Wombert> execute() gets filtered parameters 16:40 < Wombert> whereas validate() uses the original ones 16:40 < eremit> ah ok ^^ 16:41 < eremit> btw... the new exception screen is nice ... i really start to like it the more i see it :P 16:42 < eremit> ok, how to unlock the parameters ? 16:42 < Wombert> you don't unlock parameters 16:42 < Wombert> do not try 16:42 < Wombert> right now, you could do $this->getContext()->getRequest()->unlock(); 16:42 < eremit> yeah that sounds like a solution atm ... 16:42 < Wombert> I swear if I catch someone doing this I will make that impossible. I already know how, I just couldn't be bothered yet ;) don't try! 16:42 < Wombert> no 16:42 < Wombert> no 16:42 < Wombert> no 16:42 < Wombert> well 16:43 < Wombert> seems like it's time for plan b then 16:43 < eremit> ok, i ll take a look about how to get the parameters into the validate without using unlock directly :P 16:43 < v-dogg> Wombert: lock it and throw away to key! 16:43 < Wombert> [15:39] Wombert: oh bloody hell 16:43 < Wombert> [15:39] Wombert: my pad 16:43 < Wombert> [15:39] Wombert: bad 16:43 < Wombert> v-dogg: that's exactly what I had in mind 16:43 < Wombert> lock() returns a key 16:44 < Wombert> and you can only unlock it with that key 16:44 < Wombert> and the key is stored... well... I don't know yet, but I'll find a way ;) 16:46 < eremit> did i get it right ... you are locking the params, so nobody can change them ? ... 16:46 < Wombert> no 16:47 < Wombert> a) I'm fixing the issue where you can't access params in validate() right now, as I said, just give me a minute 16:47 < Wombert> b) locking is done so you can't use getParameter() on the request in an action or view 16:47 < eremit> sure i have all time of the world, i'm just trying to get back into the code and exspec. into the ideas behind it :P 16:51 < CIA-6> david * r867 /trunk/ (3 files in 3 dirs): Changed FormPopulationFilter to use list of parameters for the 'methods' parameter in the config. That should be it for now. Closes #79 16:52 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:52 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/79: Ticket #79 (enhancement) closed by david 16:52 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/867: Changeset [867]: Changed FormPopulationFilter to use list of parameters for the 'methods' ... 16:52 < v-dogg> *clickety click* 16:53 < CIA-6> david * r868 /trunk/src/filter/AgaviExecutionFilter.class.php: Fixed an issue where accessing original request parameters inside Action::validate() was not possible. Refs #281 16:55 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit [Read error: 110 (Connection timed out)] 16:55 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 16:55 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/868: Changeset [868]: Fixed an issue where accessing original request parameters inside ... 16:55 < eremit> yea ... ty 16:57 -!- shoan_ [n=shoan@61.95.206.107] has joined #agavi 16:58 * Wombert meanwhile continues to carry out his evil plan 17:01 < CIA-6> david * r869 /branches/dominik-validation_mods/src/request/AgaviRequest.class.php: made all parameter related methods in Request final. Refs #281 17:01 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 17:01 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/869: Changeset [869]: made all parameter related methods in Request final. Refs #281 17:01 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/ticket/280: Ticket #280 (enhancement) closed by dominik 17:02 < CIA-6> david * r870 /branches/dominik-validation_mods/src/buildtools/ (2 files in 2 dirs): WTF WENT WRONG WITH THESE MERGES.... ARRGH 17:03 < Wombert> I said FUCKING REFS #281 17:03 < Wombert> you stupid postcommit hook 17:03 < Wombert> ZOMG 17:03 < Wombert> wtf why is this listed at closed here 17:03 * Wombert kicks trac 17:03 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 17:03 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/870: Changeset [870]: WTF WENT WRONG WITH THESE MERGES.... ARRGH 17:04 < v-dogg> keep'em coming lads, http://cia.navi.cx/stats/project/Agavi :) 17:04 < Wombert> cia uses sunday as first day of the week 17:04 < Wombert> how... american 17:04 < v-dogg> how... stupid 17:08 < v-dogg> I'm off, catch (YouLater $e) { ; } 17:10 -!- shoan [n=shoan@61.95.206.107] has quit [Read error: 110 (Connection timed out)] 17:11 -!- impl [n=impl@httpcraft/php/nforbes] has joined #agavi 17:12 < digitarald> what about a deleteCookie() function ;) 17:14 -!- raidman|Work [n=raidman@62.60.194.3] has joined #agavi 17:14 < eremit> the FormPopulationFilter is broken too atm ? 17:14 < Wombert> no 17:15 < Wombert> or is it? 17:15 < Wombert> works for me 17:15 < eremit> do i only need to set enabled="true" in global_filters? ... 17:15 < eremit> if thats the only thing to do, it's broken ... because of the in_array check 17:17 < Wombert> ... 17:18 < Wombert> http://trac.agavi.org/trac.cgi/changeset/867 17:19 < eremit> i just made an svn update ^^ 17:21 < eremit> ah stupid me ... 17:26 < Wombert> hah! 17:26 < Wombert> there you go 17:26 < CIA-6> david * r871 /trunk/src/ (2 files in 2 dirs): made it impossible to unlock the Request without knowing the secret key :) 17:26 < Wombert> tadaaaa 17:26 * Wombert does the rain dance 17:27 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 17:27 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/871: Changeset [871]: made it impossible to unlock the Request without knowing the secret key :) 17:35 -!- Wombert_ [n=Wombert@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 17:35 -!- Wombert [n=Wombert@dslb-088-064-006-044.pools.arcor-ip.net] has quit [Nick collision from services.] 17:35 -!- Wombert_ is now known as Wombert 17:39 < eremit> Wombert, maybe i still stupid ... 17:39 < eremit> but changing the configs throws a parser exception 17:40 < Wombert> then fix your configs 17:40 < impl> Wombert: Did you fix the xsd to match your changes? 17:41 < eremit> ^^ 17:41 < Wombert> of course 17:42 * impl doesn't even see a schema for global_filters.xml 17:43 < impl> oh, just filters.xsd 17:44 * impl nods in agreement 17:44 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has joined #agavi 17:50 < CIA-6> dominik * r873 /trunk/samples/webapp/config/compile.conf: removed unneeded compile.conf from sample app 17:50 < CIA-6> david * r872 /trunk/src/ (agavi.php core/Agavi.class.php): use spl_autoload_register() instead of __autoload() if possible, closes #299 17:50 -!- AgaviBot [n=dzuelke@dslb-088-064-006-044.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 18:00 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 18:02 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 18:02 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/873: Changeset [873]: removed unneeded compile.conf from sample app 18:02 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/ticket/299: Ticket #299 (enhancement) closed by david 18:02 < AgaviBot> 3) http://trac.agavi.org/trac.cgi/changeset/872: Changeset [872]: use spl_autoload_register() instead of __autoload() if possible, closes ... 18:02 < AgaviBot> 4) http://trac.agavi.org/trac.cgi/changeset/871: Changeset [871]: made it impossible to unlock the Request without knowing the secret key :) 18:02 < AgaviBot> 5) http://trac.agavi.org/trac.cgi/changeset/870: Changeset [870]: WTF WENT WRONG WITH THESE MERGES.... ARRGH 18:02 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has quit [Excess Flood] 18:02 < Wombert> haha 18:03 < Wombert> yeah 18:08 < impl> :x 18:09 -!- shoan_ [n=shoan@61.95.206.107] has quit [Read error: 110 (Connection timed out)] 18:10 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 19:15 < CIA-6> david * r874 /trunk/ (28 files in 14 dirs): changed core.webapp_dir to core.app_dir 19:17 < v-dogg> evening 19:18 < Wombert> hi brother 19:19 < v-dogg> hows it going 19:23 < raidman|Work> evening guys :) 19:24 < raidman|Work> s/guys/developers of agavi/ 19:24 < raidman|Work> :P 19:24 < raidman|Work> the best framework I ever seen befor 19:28 < CIA-6> v-dogg * r875 /trunk/etc/generate_package.php: Removed generate_package.php. It's all done in AgaviPackageTask now. Good riddance. 19:34 < v-dogg> guys... Warning: Channel validator warning: field "version" - A version number should have 3 decimals (x.y.z) 19:35 < impl> 0.11.0? 19:35 < Wombert> who are these morons to tell me how many digits my version number should have 19:36 < Wombert> if there was something else than pear I swear I'd be using it 19:36 < eremit> lol ... 19:36 < v-dogg> to ignore or not to ignore... 19:42 < v-dogg> why do I see closed tickets when I choose "Active tickets: ##" from milestone view? 19:43 < eremit> recently closed tickets ? 19:43 < Wombert> you reloaded 19:43 < Wombert> press the update button and they are gone 19:43 < v-dogg> roger 19:43 < Wombert> bold ones are new, italic ones are changed 19:45 < v-dogg> is #34 still valid? I mean didn't you Wombert get rid of most the &'s already? 19:45 < Wombert> some 19:45 < Wombert> not all 19:45 < v-dogg> ok 19:45 < Wombert> that's one of the post-beta1 tickets 19:46 < v-dogg> what about 184? 19:46 < Wombert> one minute 19:46 < Wombert> let me just quickly finish this checkin 19:46 < v-dogg> sure 19:46 < v-dogg> I'm in no hurry here 19:47 < v-dogg> maybe I'll get something to eat and a beer while you finish up that :) 19:47 < Wombert> do you have time later to work on #293 with me? 19:48 < Wombert> we'll then also fix the rewriting for your veikkomakinen.com box 19:49 < v-dogg> sure 19:49 < Wombert> you got plenty of time on your hands lately it seems 19:49 < Wombert> how come? 19:50 < v-dogg> no school work I suppose 19:51 < v-dogg> and less other things to do at work so I've had time to work on my webapp project 19:51 < Wombert> :) 19:51 < v-dogg> but this is about to change very soon :) 19:52 < Wombert> that's unfortunate 19:52 < Wombert> not tooooo soon I hope ;) 19:52 < v-dogg> I finally have to get first release of this webapp project ready 19:52 < v-dogg> +the 19:52 < v-dogg> and start prototyping the next stage 19:54 < CIA-6> david * r876 /trunk/ (20 files in 14 dirs): renamed webapp to app, www to pub. this breaks your app. I warned you. 19:59 < CIA-6> david * r877 /trunk/ (5 files in 3 dirs): last renames, that should close #298 19:59 < eremit> are there any plans using __Set and __get as generic setters ... to get rid of $this->getController()->getRequest() stuff and favor $this->Controller->Request ? 19:59 < Wombert> no 19:59 < Wombert> that sucks 19:59 < Wombert> come on 19:59 < eremit> just a question :P ... that was a thing i liked on prado ... 19:59 < eremit> less type work 20:00 < v-dogg> no no no! 20:00 < v-dogg> that'll screw up autocomplete 20:00 < Wombert> yes 20:00 < Wombert> for instance 20:00 < eremit> at least my autocomplete doesn't work anyway ... 20:00 < Wombert> we're not prado 20:00 < v-dogg> your autocomplete sucks then :) 20:00 < Wombert> prado, like many other frameworks, sacrificed reason for ease 20:00 < Wombert> which is a bad idea 20:00 < v-dogg> zend studio uses phpdoc @return 20:01 < eremit> :P ... i enforced using this framework in favor of prado ... because agavi != prado ... but still thats a pt i liked ... 20:01 < eremit> v-dogg my zend studio doesn't get the autocompletion ... :/ 20:01 < v-dogg> and even quite nicely, even $this->getContext()->getRequest()->getA ... works 20:02 < Wombert> eremit: maybe you have to enable it 20:02 < eremit> it is enabled ... 20:02 < Wombert> getters are getters for a reason 20:02 < Wombert> also 20:02 < eremit> :P ... sometimes it even works ... 20:02 < Wombert> now that we have the new structure 20:02 < Wombert> you usually don't need request etc anyway 20:02 < Wombert> did you get that eremit... how attribs are now set for the view? 20:03 < Wombert> no more setting attribs in the request and then importing them in the view 20:03 < eremit> $this->setAttribute afaik 20:03 < Wombert> in the action, do $this->setAttribute 20:03 < Wombert> and they are auto-imported in the view 20:03 < eremit> yeah got that ... 20:03 < Wombert> good 20:03 < Wombert> we discussed having $this->getUser() or so in actions but... well 20:03 < eremit> i only need the whole request stuff because of my fancy validators ... 20:04 < Wombert> there is no clean way of doing it, so we don't do it at all 20:04 < Wombert> instead of having a stupid compromise 20:04 < Wombert> $r = $this->getContext()->getRequest(); 20:04 < eremit> yes ^^ ... 20:04 < Wombert> or how about doing YourBaseAction extends Action 20:04 < Wombert> that sets $this->r in initialize() 20:04 < Wombert> that's even less to type than prado then 20:04 < Wombert> see 20:05 < Wombert> this is one of the areas where it's pretty much up to the taste of the individual developer 20:05 < eremit> that's what i'm thinking about ... but i don't wanna do that until i get a look at the validators ... 20:05 < Wombert> so 20:05 < Wombert> do it yourself ;) 20:05 < Wombert> it's easy 20:05 < eremit> i think most of my validation can be done with em too 20:05 < Wombert> yep 20:05 < Wombert> it will ;) 20:05 < Wombert> trust me 20:05 < Wombert> we're porting a pretty complicated project to 0.11 20:05 < eremit> i read the code in the repos ... looks really really nice ... 20:05 < Wombert> and dominik won't stop until everything works with agavi's validation 20:06 < eremit> and i started a pretty complicated project at least partwise in .11 20:07 < eremit> sadly there's a part in the project which cannot be written using agavi :/ 20:09 < Wombert> hm? 20:09 < Wombert> which 20:09 < Wombert> never say never 20:09 < eremit> it's a daemon written in php ... 20:09 < Wombert> and why not agavi? 20:09 < eremit> it would not make any sense to write a daemon application in agavi ... 20:09 < Wombert> the irc bot is written in agavi, too ;) 20:09 < Wombert> what does the daemon do 20:09 < eremit> have you ever tried to fork in php ? ... 20:10 < eremit> it checks every second a mysql table ... and if there are entries to process forks itself to proccess 50 entries per fork ... 20:10 < eremit> we approx will have about 20k entries /min ... 20:10 < Wombert> hm 20:10 < eremit> do you know what a bunch of several forked agavi cli apps means to the ram ? :P 20:11 < Wombert> well 20:11 < Wombert> the bot takes up about 15MB ;) 20:11 < Wombert> but hey, it doesn't leak at all! 20:11 < Wombert> http://trac.agavi.org/trac.cgi/roadmap 20:11 < Wombert> NINETY PERCENT 20:11 < Wombert> WOOOOOOOT 20:11 < Wombert> WOOOOOOOOOOT 20:11 < eremit> ^^ -- your action description is wrong in the source ... 20:12 < v-dogg> Wombert: :) 20:12 < eremit> and gratz Wombert 20:12 < Wombert> eremit: 20:12 < Wombert> ? 20:12 < eremit> The execution of these methods is not dependent ... bla ... validateWrite() will be run even if there is no executeWrite() 20:13 < Wombert> and that is wrong? 20:13 < Wombert> oO 20:13 < eremit> wrong ... i just sat here and wondered about 20mins why my validateWrite doesn't get executed 20:13 < Wombert> oh 20:16 < eremit> further the usage of the {debug} macro of smarty kills you repop filter 20:16 < eremit> but most likely it's the debug which isn't xhtml compat ... 20:16 < Wombert> yep 20:18 < eremit> w00 ... you have changed so much ... 20:18 < eremit> i detect myself thinking the old way of .10 all the time ... oO 20:18 -!- shoan [n=shoan@59.92.145.138] has joined #agavi 20:21 < v-dogg> wombiieee 20:22 < Wombert> y0 mate 20:22 < CIA-6> david * r878 /trunk/src/exception/templates/shiny.php: fixed shiny.php exception template where file path shortening was foobar'd if any of the constants used wasn't yet defined (e.g. when something really bad happened in bootstrap() or so 20:22 < Wombert> what's up 20:23 < v-dogg> wanna start fixing Teh Routing 20:23 * impl is still determined to figure out a better solution than decorators/slots 20:24 < CIA-6> david * r879 /trunk/src/core/ (Agavi.class.php AgaviContext.class.php): throwing exceptions instead of triggering errors when no env/context name given. it just looks nicer :D 20:24 < impl> I almost had it the other day, but the pieces just wouldn't fit into place 20:24 < Wombert> impl: you still got some time until 2.0 ;) 20:24 < impl> pssh :) 20:24 < v-dogg> what's up with uwe btw? 20:24 < Wombert> no clue 20:24 < Wombert> he just disappeared 20:24 < impl> http://trac.agavi.org/trac.cgi/ticket/287 hehe 20:26 < v-dogg> did he contribute something to validation? 20:26 < Wombert> we need 15 or 16 commits a day if we want our growth to continue ;) 20:26 < Wombert> (like http://blog.bitxtender.com/2006/08/01/agavi-project-activity/ ) 20:26 < Wombert> v-dogg: yes, you bet 20:26 < Wombert> he had a lot of excellent ideas 20:26 < Wombert> and implemented large parts 20:27 < Wombert> well there was stuff we didn't like and that didn't make sense, but all in all, it wasn't too bad 20:27 < Wombert> dominik rewrote quite a bit of it 20:27 < Wombert> dominik-validation_mods is a branch of uwe-validation ;) 20:27 < Wombert> anyways, we'll credit him first for the new validation 20:29 < Wombert> sighs 20:29 < raidman|Work> 90% yeah :) 20:29 < Wombert> let's have another look at the roadmap shall we 20:29 < v-dogg> let's 20:30 < v-dogg> I'm fine with leaving #65, #183, #78, #107, #137, #184 and #34 open untill post-beta1 era 20:31 < Wombert> I concur 20:31 < v-dogg> what do you want to do to sample app? it already uses a master template and slots, doesn't it? 20:32 < Wombert> I want to add german and finnish to it 20:32 < impl> What happened to kludgebox? 20:32 < Wombert> bob? 20:32 < v-dogg> too busy working 20:32 < Wombert> he doesn't have time anymore 20:32 < Wombert> neither does mike 20:32 < impl> Oh :( 20:33 < Wombert> bob declared me "captain of the ship" quite some time ago ;) 20:33 < v-dogg> Wombert: german and finnish? 20:33 < impl> I was looking at some old tickets and I realized that I hadn't seen him or agile around at all for a really long time 20:33 < impl> hmmph 20:33 < digitarald> routing question 20:33 < v-dogg> shoot 20:33 < digitarald> can i set the default values? 20:34 < v-dogg> sure 20:34 < v-dogg> http://trac.agavi.org/trac.cgi/wiki/Routing 20:34 < digitarald> in the application :) 20:34 < impl> digitarald: Take a look at the sample application 20:34 < v-dogg> at least if you have a callback class 20:34 < impl> 20:34 < digitarald> i have 20:34 < digitarald> e 20:35 < v-dogg> but what excatly do you need? 20:35 < Wombert> digitarald: do you want defaults for generating? or for incoming routes 20:35 < Wombert> v-dogg: german and finnish language versions 20:35 < v-dogg> ok :) 20:35 < v-dogg> swedish too? 20:35 < digitarald> 20:35 < digitarald> 20:35 < digitarald> 20:35 < digitarald> 20:35 < digitarald> 20:35 < Wombert> if you know swedish... sure 20:35 < v-dogg> ok :) 20:35 < Wombert> raidman speaks farsi and... what was the other one? 20:36 < v-dogg> (missus can help with it :) 20:36 < impl> digitarald: en ? 20:36 < raidman|Work> Wombert armenian :) 20:36 < Wombert> right 20:36 < Wombert> :) 20:36 < digitarald> no ... then its always /en/ .... 20:36 < impl> Wombert: How about Latin? ;p 20:36 < Wombert> omg 20:36 < eremit> old greek could i provide too ^^ 20:36 < Wombert> oO 20:36 < v-dogg> haha 20:37 < digitarald> i want to set lang-default not in the xml, i want to set it during the callback 20:37 < v-dogg> sample app is getting quite international 20:37 < Wombert> digitarald: then do it! 20:37 < eremit> sorry wombert had to make my latinum and small graecum at school :P 20:37 < digitarald> how can i access the default value in onNotMatched? 20:37 < Wombert> my condolences 20:37 < impl> Latin is a cool language 20:37 < impl> :P 20:37 < Wombert> digitarald: I wouldn't do it like that 20:38 < Wombert> store the language in a global place 20:38 < Wombert> e.g. as a request attribute 20:38 < Wombert> and read it in onGenerate 20:38 < raidman|Work> Wombert, farsi is a right to left language, I can help in i18n for farsi 20:38 < eremit> impl: latin is really cool, you can read nearly any roman based language newspaper :P 20:38 < digitarald> languages are stored in database ;) 20:38 < Wombert> in onMatched and onNotMatched, set it there 20:38 < Wombert> yes but the current language... 20:38 < impl> eremit: Indeed :D 20:38 < raidman|Work> Wombert, hyeren(armenian) is realy like english 20:38 < Wombert> really? 20:38 < Wombert> oO 20:38 < Wombert> yeah I knew farsi is rtl 20:39 < Wombert> shouldn't be a problem with utf8 :) 20:39 < impl> A lot of languages are a lot like English 20:39 < impl> Like... Estonian? I think 20:39 < impl> No 20:39 < impl> Romanian 20:39 < digitarald> how can i access it the default values? 20:39 < eremit> i could also provide platt german (doen't know the translation of platt :P ) 20:39 < eremit> frisian ? 20:39 < digitarald> nordic german ;) 20:39 < impl> http://www.php.net/manual/ro/introduction.php 20:40 < impl> Very similar to English :P 20:40 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has quit ["Bye Bye"] 20:40 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has joined #agavi 20:40 < digitarald> solution? 20:40 < digitarald> have to go ... cya tomorrow 20:40 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has quit [Client Quit] 20:41 < Wombert> uh 20:41 < Wombert> well that was quick 20:41 < Wombert> :D 20:41 < Wombert> impl: yeah well I understand most of the romanian text there 20:42 < Wombert> I do, to an extent, also understand dutch, danish, swedish 20:42 < Wombert> all pretty similar to german 20:42 < eremit> ^^ if you wait a year .. i'm currently learning swedish and norwegian ... 20:42 < impl> A lot of Swedish looks like English too 20:43 < Wombert> but understanding and writing/speaking are pretty different things ;) 20:43 < Wombert> the only thing where I really don't understand jack schitt is finnish 20:43 * Wombert hugs horros 20:43 < impl> haha 20:43 < Wombert> (who isn't there) 20:43 * Wombert hugs v-dogg 20:43 < Wombert> (who is there) 20:43 < impl> I speak a little Swedish... very little 20:46 < impl> Jag ar en Svenska person 20:46 < impl> or somesuch :P 20:49 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has quit ["Bye Wombert!"] 20:50 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 20:51 * Wombert scratches his head 20:51 < Wombert> this is gonna get pretty nasty 20:51 < Wombert> with the way output types work right now 20:51 < Wombert> because you can't have parents 20:51 < Wombert> and we already have xhtml and html 20:51 < Wombert> and 20:51 * Wombert thinks 20:51 < Wombert> no, this is not gonna work 20:52 * Wombert concludes that we need i18n support inside the framework 20:52 < Wombert> well it _can_ work 20:52 < Wombert> but it's not really cool if we want per-language templates 21:01 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 21:01 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/164: Ticket #164 (task) closed by david 21:01 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/wiki/ConfigurationDirectives: ConfigurationDirectives edited by david 21:11 < v-dogg> what sort of i18n support? 21:11 < v-dogg> $this->getContext->getCulture() =P 21:13 < Wombert> I'd say from the controller 21:13 < Wombert> but yes, something along these lines 21:13 < Wombert> maybe with support for per-language templates 21:13 < Wombert> and some other stuff 21:14 < Wombert> translation service that can translate validation messages etc 21:38 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit ["#killall raidman"] 21:46 -!- [1]eremit [n=eremit@p5490D874.dip.t-dialin.net] has joined #agavi 22:03 -!- eremit [n=eremit@p5490D3B4.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 22:03 -!- [1]eremit is now known as eremit 22:06 < eremit> n8 all 22:06 -!- eremit [n=eremit@p5490D874.dip.t-dialin.net] has quit [" Try HydraIRC -> http://www.hydrairc.com <-"] 22:27 < CIA-6> david * r880 /trunk/src/routing/AgaviRouting.class.php: routing generator options are now passed as nested parameters to the 'generator' parameter 22:29 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 22:29 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/880: Changeset [880]: routing generator options are now passed as nested parameters to the ... 23:08 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 23:08 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/300: Ticket #300 (enhancement) created by david 23:32 < CIA-6> david * r881 /trunk/ (4 files in 2 dirs): 23:32 < CIA-6> cleaned up action code, removed some references nonsense, updated phpdocs, made 23:32 < CIA-6> private vars protected where feasible, changed ActionStackEntry::getMicrotime() 23:32 < CIA-6> to work with and return a proper float value. adapted tests. refs #65, #281, 23:32 < CIA-6> #107, #137 and #34; closes #300 23:33 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 23:33 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/300: Ticket #300 (enhancement) closed by david 23:33 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/881: Changeset [881]: cleaned up action code, removed some references nonsense, updated phpdocs, ... 23:36 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 23:36 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/302: Ticket #302 (enhancement) created by david 23:36 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/ticket/301: Ticket #301 (enhancement) created by david 23:39 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 23:39 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/303: Ticket #303 (enhancement) created by david --- Day changed Wed Aug 02 2006 00:02 < CIA-6> david * r882 /trunk/src/database/ (9 files): cleaned up database code, removed some references nonsense, updated phpdocs. refs #107 and #137 00:05 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 00:05 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/882: Changeset [882]: cleaned up database code, removed some references nonsense, updated ... 00:05 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has quit ["Bye Wombert!"] 00:28 * Macca waves 01:01 < Macca> what'd you change??? 01:20 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has left #agavi [] 02:03 -!- benny`work [n=benny@p54AC3CD4.dip0.t-ipconnect.de] has quit ["unset($this);"] 02:19 -!- Wombert [n=Wombert@dslb-088-064-013-061.pools.arcor-ip.net] has quit [] 02:28 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 03:50 < Macca> [Wed Aug 02 10:50:11 2006] [error] [client 203.206.205.212] PHP Fatal error: Allowed memory size of 8388608 bytes exhausted (tried to allocate 1204948 bytes) in /var/www/agavi/src/renderer/AgaviPhpRenderer.class.php on line 77, referer: https://development.maccasoft.com.au/lld/direct/www/client/find-contractor 03:50 < Macca> doesn't sound good 03:53 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 03:54 < Macca> ... 04:46 < Macca> someone fix it!!! 04:49 < Macca> it occurs when i do: $this->getContext()->getRequest()->setAttribute('result', $ret); 04:49 < Macca> $ret -> $ret[0] fixed it 04:49 < Macca> so nm :D 06:16 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 06:22 -!- shoan_ [n=shoan@59.92.153.132] has joined #agavi 06:25 -!- shoan [n=shoan@59.92.145.138] has quit [Read error: 60 (Operation timed out)] 08:07 -!- impl [n=impl@httpcraft/php/nforbes] has quit ["Leaving"] 08:14 < v-dogg> huomenta 08:21 -!- raidman|Work [n=raidman@62.60.194.3] has joined #agavi 08:42 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 08:52 < ttj> Huomen'. 08:54 < v-dogg> huoment 08:59 < v-dogg> gaahh... we really have to do something to agavi.org 09:00 < v-dogg> it's driving people away 09:00 < ttj> :-) 09:03 < ttj> So, what do you want to do about it? 09:06 < v-dogg> first, remove that video and update api docs 09:29 -!- shoan_ [n=shoan@59.92.153.132] has quit [] 09:35 -!- shoan [n=shoan@59.92.153.132] has joined #agavi 10:54 < v-dogg> "the icons are still in PNG format, the SVG files will be relased once the support for the format improves in KDE" 10:55 < v-dogg> wtf? how stupid is that 10:55 < v-dogg> (Crystal SVG) 11:02 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has joined #agavi 11:06 < digitarald> morning 11:11 -!- shoan [n=shoan@59.92.153.132] has quit [] 11:15 -!- Wombert [n=Wombert@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 11:15 < raidman|Work> huoment, morning 11:35 -!- eremit [n=eremit@p5490D874.dip.t-dialin.net] has joined #agavi 11:35 < eremit> heyja 11:57 -!- shoan [n=shoan@61.95.206.107] has joined #agavi 12:03 < v-dogg> Wombert: do you have any good _svg_ icon sets? 12:03 < v-dogg> crystal icons are released only in png 12:03 < v-dogg> "the SVG files will be relased once the support for the format improves in KDE" 12:03 < eremit> is it intended handleWriteError() is always executed, even if there is no error in set in the request ... 12:04 < eremit> v-dogg ... try kde-look and the gnome-look pages ^^ i used to get some free svg's from there 12:04 < v-dogg> ... and that's the most ridiculous reason I've ever heard 12:08 < eremit> uhm ... i don't get the idea of current trunk ... 12:18 < digitarald> did u try tango icon set? 12:19 < v-dogg> yes 12:19 < v-dogg> http://commons.wikimedia.org/wiki/Category:Nuvola_SVG_icons 12:23 < Wombert> have have either nuvola or crystal as svg 12:23 < Wombert> can't remember which 12:23 < Wombert> tango.... 12:23 < Wombert> besides that... none 12:24 < digitarald> just a question ... why is webapp now app? because of clear seperation from context? 12:24 < Wombert> because it doesn't have to be a web app 12:24 < v-dogg> because agavi irc bot is not a webapp 12:24 < Wombert> could be a deamon, an irc bot, a console application, whatever 12:24 < v-dogg> and renaming webapp to ircapp would have been... silly ;) 12:24 < digitarald> my speech ;) 12:26 < v-dogg> hey Wombert, I'm updating pear packaging 12:27 < v-dogg> who do we want to list as devs? 12:27 < v-dogg> currently there are bob, mike, you and I 12:28 < Wombert> one second 12:37 < Wombert> [11:04] eremit: is it intended handleWriteError() is always executed, even if there is no error in set in the request ... 12:37 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 12:37 < Wombert> sure? 12:37 < Wombert> :p 12:39 < Wombert> eremit: are you returning true from validate()? 12:39 < Wombert> also, there was another issue yesterday with validateWrite() and such, right? 12:39 < Wombert> I will fix these then 12:48 < digitarald> the loggers.xml from the buildtools is not up to date ... i have no sample for changing my invalide working copy :( 12:48 < CIA-6> dominik * r883 /trunk/src/routing/AgaviWebRouting.class.php: 12:48 < CIA-6> fix missing initialization of a variable 12:48 < CIA-6> closes #304 12:48 < Wombert> oh yeah and v-dogg the tests are broken for logging 12:49 < v-dogg> woot? 12:49 < v-dogg> how? 12:49 < Wombert> dunno 12:49 < Wombert> run it 12:49 < Wombert> bbiab 12:49 < v-dogg> php AllTests.php, like this? 12:50 < Wombert> phpunit AllTests 12:50 < Wombert> inside tests2/ 12:50 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 12:50 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/ticket/304: Ticket #304 (defect) closed by dominik 12:50 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/883: Changeset [883]: fix missing initialization of a variable closes #304 12:54 < v-dogg> what the hell has happened! 12:55 < eremit> anybody knows a good smtp enabled mail class besides the pear ones ? 12:55 < v-dogg> that php mailer or something 12:57 * v-dogg slaps Wombert around a bit with a huge pink dildo 12:58 < v-dogg> http://trac.agavi.org/trac.cgi/changeset/874#file20 12:58 < v-dogg> you broke it 12:59 < v-dogg> LoggingConfigHandlerTest.php must be reverted back to http://trac.agavi.org/trac.cgi/changeset/863 13:17 < eremit> the xsd of config_handlers.xsd doesn't match the example in the config_handlers.xml in code_templates ... fyi 13:18 < digitarald> patch day :( 13:20 < v-dogg> eremit: what's wrong? 13:21 < eremit> the whole example ... isn't allowed to have childs ... 13:22 < eremit> it must be 13:31 < eremit> uh ... is there any way to get a custom config file being parsed and inserted into AgaviConfig automatically ? 13:33 < v-dogg> not officially 13:33 < eremit> :/ 13:33 < v-dogg> I wrote a parser for that 13:34 < eremit> uh? - my problem is the config_handlers only get parsed after dispatch ... 13:34 < eremit> the only solution i'm aware of would be a customized filter but a bit heavy overhead only for some additional app specific config files 13:35 < v-dogg> my parser parses a simple xml config and sets the values into AgaviConfig 13:36 < v-dogg> I use it between bootstrap() and dispatch() 13:36 < eremit> yeah that's where i tried to use AgaviConfigCache ... but .... 13:39 < v-dogg> http://pastebin.ca/111482 13:40 < v-dogg> and a sample config: http://pastebin.ca/111483 13:40 < v-dogg> (note that there's a raw content download after the listing) 13:41 < eremit> ty^^ 13:49 < CIA-6> v-dogg * r884 /trunk/ (2 files in 2 dirs): Default logging configs should now work almost out-of-the-box (you only need to mkdir %core.app_dir%/log/ and enable logging in settings.xml). 13:52 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 13:52 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/884: Changeset [884]: Default logging configs should now work almost out-of-the-box (you only ... 13:54 < Wombert> [12:34] eremit: uh? - my problem is the config_handlers only get parsed after dispatch ... 13:54 < Wombert> wrong 13:54 < eremit> hm? 13:56 < eremit> i get a AgaviConfigurationException '.../application.xml' does not have a registered handler. 13:56 < eremit> and it is surely registered ... 13:56 < v-dogg> no it isn't :) 13:57 < eremit> ok it is in the config_handlers.xml :P 13:57 < v-dogg> 13:58 < eremit> ah i'm answering wombert^^ ... 13:59 < eremit> you thinggie is working nicely ... 13:59 < v-dogg> ok 13:59 < v-dogg> so what gives you AgaviConfigurationException '.../application.xml'... ? 13:59 < eremit> if i have the following: 14:00 < eremit> 14:00 < eremit> and do a AgaviConfigCache::import(AgaviConfig::get('core.config_dir').'/application.xml'); between bootstrap and dispatch 14:01 < eremit> oh .. anyhow i just see i have misunderstood the usage of import havent i ? 14:02 < Wombert> yes 14:02 < v-dogg> $array = include(AgaviConfigCache::checkConfig(AgaviConfig::get('core.config_dir') . '/application.xml')); 14:02 < Wombert> $config = include(AgaviConfigCache::checkConfig( 14:02 < Wombert> he was faster ;) 14:02 < Wombert> just "application.xml" _should_ work too 14:02 < v-dogg> copy-paste, you see :) 14:02 < Wombert> not sure tho 14:02 < eremit> yes...ty ... 14:03 < v-dogg> Wombert: did you get my dildo-slapping? 14:03 < Wombert> is enough, too 14:03 < Wombert> v-dogg: still busy 14:03 < Wombert> sorry 14:03 < v-dogg> np 14:03 < Wombert> will have a look at the dildos you bought later 14:03 < Wombert> harhar 14:03 < Wombert> :> 14:05 < eremit> nooOOOo...pls kill my boss ... after having the solution to the config problem he came in ... saw xml and just said ... we dont wanna use xml for configuration ... 14:05 * eremit gets his gun having to clear some things up. 14:07 < v-dogg> I sure would like to hear why not xml 14:07 < eremit> his reason is quite simple ... our support and installation team will never understand it ... rofl ... 14:08 < v-dogg> yeah, go ahead and get the gun 14:08 < v-dogg> your support and installation team is going to need to be put down 14:08 < eremit> yep ... 14:08 < eremit> but that's not my problem :PP its another division ... 14:09 < Wombert> eremit: show him http://trac.agavi.org/trac.cgi/wiki/ConfigurationFormatsCompared 14:09 < Wombert> (outdated tho!) 14:10 < eremit> yeah but it's nice to compare all of them ... 14:10 < eremit> anyhow ... now i simply do a AgaviConfig::Import( parse_ini_file( ... ) ); *shrugs* 14:14 < CIA-6> dominik * r885 /branches/dominik-validation_mods/src/filter/AgaviFormPopulationFilter.class.php: fix bug in fpf when a checkbox didn't have a value set but the given value in the populate array was false (it would have set the chechbox to true then) 14:14 < CIA-6> dominik * r886 /branches/dominik-validation_mods/src/ (2 files in 2 dirs): new convenience functions 14:15 < CIA-6> dominik * r887 /branches/dominik-validation_mods/src/request/AgaviRequest.class.php: and another convenience function 14:17 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 14:17 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/887: Changeset [887]: and another convenience function 14:17 < AgaviBot> 2) http://trac.agavi.org/trac.cgi/changeset/886: Changeset [886]: new convenience functions 14:17 < AgaviBot> 3) http://trac.agavi.org/trac.cgi/changeset/885: Changeset [885]: fix bug in fpf when a checkbox didn't have a value set but the given value ... 14:28 < CIA-6> david * r888 /trunk/src/config/AgaviConfig.class.php: phpdoc fix 14:29 < AgaviBot> News from Agavi: Timeline at http://trac.agavi.org/trac.cgi/timeline: 14:29 < AgaviBot> 1) http://trac.agavi.org/trac.cgi/changeset/888: Changeset [888]: phpdoc fix 14:55 -!- shoan [n=shoan@61.95.206.107] has quit [] 14:56 -!- shoan [n=shoan@61.95.206.107] has joined #agavi 14:58 -!- [1]eremit [n=eremit@p5490D874.dip.t-dialin.net] has joined #agavi 14:59 -!- AgaviBot [n=dzuelke@dslb-088-064-013-061.pools.arcor-ip.net] has quit ["Bye Wombert!"] 14:59 -!- Wombert [n=Wombert@dslb-088-064-013-061.pools.arcor-ip.net] has quit [] 15:00 -!- eremit [n=eremit@p5490D874.dip.t-dialin.net] has quit [Nick collision from services.] 15:01 -!- [1]eremit is now known as eremit 15:05 -!- Wombert [n=Wombert@dslb-088-064-013-061.pools.arcor-ip.net] has joined #agavi 15:07 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit ["#killall raidman"] 15:16 < Wombert> re 15:51 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 16:08 -!- brujahRg [n=brj@212.175.111.100] has joined #agavi 16:10 -!- benny`work [n=benny@p54AC42F7.dip0.t-ipconnect.de] has joined #agavi 16:38 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 16:44 < eremit> wombert you want the users to prevent having logic in getDefaultView()? 16:44 < Wombert> yes 16:44 < Wombert> does 16:44 < Wombert> not 16:44 < Wombert> belong 16:44 < Wombert> there 16:44 < Wombert> that's why it's called getDefaultViewName() now 16:44 < eremit> why not remove the function? and add a constant ? 16:44 < v-dogg> ? 16:44 < eremit> to the class 16:45 < Wombert> you can't overwrite constants in child classes 16:45 < v-dogg> how would that work? 16:45 < v-dogg> yeah 16:45 < Wombert> plus, getDefaultViewName() _could_ return array('module', 'viewname'); 16:45 < eremit> and a public var ? s.t. which cannot have logic ? 16:46 < eremit> it's senseless imho to have a function which does not is allowed to have logic ... 16:46 * Wombert sighs 16:46 < Wombert> if you need logic on the first request, either put it into the view if you like 16:46 < Wombert> or add an executeRead() 16:47 < eremit> sure ... i'm only complaining about a function without the usage of a function ... 16:48 < Wombert> it is a function 16:48 < Wombert> it could even include logic to determine what the default view is 16:48 < Wombert> the problem is 16:49 < Wombert> many people had their actions serve POST 16:49 < Wombert> and then in getDefaultView 16:49 < Wombert> they implemented logic to read stuff from the database etc 16:49 < eremit> yes but WHY do you use a function if YOU dont want to have logic there ... there are logic less solutions ... 16:50 < Wombert> maybe you need logic to determine what view to use 16:51 < eremit> which could be done in executeRead 16:51 < eremit> at least i'm doing it there ... 16:54 -!- shoan [n=shoan@61.95.206.107] has quit [] 16:57 < v-dogg> fair point 17:12 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has joined #agavi 17:24 < CIA-6> v-dogg * r889 /trunk/etc/phing/AgaviPackageTask.php: Updated pear package info 17:35 -!- Wombert_ [n=Wombert@dslb-088-064-012-112.pools.arcor-ip.net] has joined #agavi 17:35 -!- Wombert [n=Wombert@dslb-088-064-013-061.pools.arcor-ip.net] has quit [Nick collision from services.] 17:36 -!- Wombert_ is now known as Wombert 18:13 -!- impl [n=impl@httpcraft/php/nforbes] has joined #agavi 18:42 < Wombert> eremit: are you there 18:48 < eremit> Wombert yes 18:48 < Wombert> you said handlError is always executed 18:48 < Wombert> I can't reproduce that 18:48 < Wombert> also, you said validateWrite() isn't executed if there is no executeWrite() 18:48 < Wombert> can't reproduce that either 18:48 < eremit> i will try to produce that again ... 18:49 < eremit> and send you the exact code ... 18:49 < eremit> handleError is not always executed ... i recognized it as my fault ... 18:49 < Wombert> you are aware that if there is no execute method at all, it won't run validation, right? 18:49 < eremit> no ... 18:49 < Wombert> i.e. if your action has only executeWrite, it won't run validateRead 18:49 < eremit> maybe that was my fault 18:50 < Wombert> because the action doesn't serve the READ method 18:50 < Wombert> so it doesn't validate either 18:50 < Wombert> has always been like that, by the way 18:50 < Wombert> okay looks like we solved these problems then ;) 18:51 < eremit> yeah ... 18:51 < eremit> oh that late ... brb must do some shopping .oO 18:55 -!- brujahRg [n=brj@212.175.111.100] has quit [] 19:15 < v-dogg> can we make some sort of a policy what class variables are by default private and what are protected 19:16 < v-dogg> or is there any good reason to keep something private? 19:16 < v-dogg> context perhaps? 19:16 < v-dogg> I wanna close http://trac.agavi.org/trac.cgi/ticket/107 19:20 < impl> I've found fairly little use for private 19:20 * impl looks through SecurityUser 19:26 < digitarald> short question ... why can i set my modpub/ directory in config and its ot used anywhere? would be cool to have a function like getBaseHref() ... getModpub() ... 19:27 < Wombert> diy 19:27 < Wombert> I never used modpub 19:27 < v-dogg> AgaviConfig::get('core.modpub'); (or something) 19:27 < digitarald> its redundant? 19:28 < Wombert> pretty much yeah 19:28 < Wombert> at least I think so 19:28 < digitarald> agavi never used it but its in the main settings file ... ? 19:28 < Wombert> because modules aren't really modular 19:28 < digitarald> yes 19:28 < digitarald> i would not use it too, the getBaseHref is all i need 19:29 < digitarald> why should i put my images/css/js in modpub ... 19:29 < Wombert> modpub/Modulename/blah 19:29 < Wombert> but nonsense imho 19:30 < Wombert> should we throw away modpub? 19:30 < Wombert> v-dogg? 19:30 < Wombert> devosc? 19:30 < Wombert> digitarald? 19:30 < Wombert> impl? 19:30 < Wombert> Macca? 19:30 < impl> I always figured modpub was for static content not related to the internal workings of Agavi 19:30 < impl> That's what I used it for in M2 19:31 < impl> If you were to have your content on some other server modpub might be useful 19:31 < impl> e.g. modpub = http://content.mydomain.com/ 19:32 < digitarald> static content in an extra subdomain ... not the usual way ... the entry is redundant 19:33 < digitarald> static content is in my htdocs ... which is 'pub' in the sample app 19:33 < impl> If you have a lot of content, maybe you'd put it in a subdomain 19:33 < impl> Lots and lots of large sites use e.g. http://images.site.com/ 19:34 < v-dogg> or you might have some general css files there 19:34 < v-dogg> I can see it not beeing impossible :) 19:35 < digitarald> but i don't want to access AgaviConfig::get('core.modpub') ... in my templates??? 19:36 < digitarald> then there has to be another implementation like $r->getModpub() 19:36 < Wombert> impl: good example 19:37 < Wombert> however, you don't have all of your content there typically 19:37 < Wombert> so I still think it's redundant 19:37 < Wombert> v-dogg: there are occasions where private makes sense 19:37 < Wombert> e.g. all view and action members should be private, imho 19:38 < v-dogg> agreed 19:38 < Wombert> I already cleaned up action and database completely yesterday 19:39 < v-dogg> I'm cleaning up everything I can find :) 19:39 < v-dogg> but phpdocs only 19:40 < Wombert> hey should I make FPF an action filter again 19:40 < Wombert> dominik suggested that 19:40 < Wombert> he argued that you can't have it per module otherwise 19:44 < v-dogg> why did you make it global? 19:45 < CIA-6> david * r890 /trunk/src/filter/ (3 files): really really minor cosmetic things 19:45 < Wombert> main reason was 19:45 < Wombert> uh 19:45 < Wombert> two ^^ 19:45 < Wombert> a) can't run it twice 19:45 < Wombert> b) you're guaanteed output is complete by then 19:45 < Wombert> I realize that this should be the casae for an action filter, too, if it's the first one but... well 19:46 < Wombert> how about a method that indicates whether or not a filter has run already? 19:46 < Wombert> or, even better, how about execute() and executeOnce() 19:46 < Wombert> ? 19:46 < Wombert> you know this 19:46 < Wombert> static $run 19:46 < Wombert> blah is nonsense 19:46 < Wombert> and dangerous, too, if you run two contexts 19:47 < v-dogg> hey btw, I got strict warnins from.... uhm... 19:48 < v-dogg> somewhere :) 19:48 < v-dogg> storage::write 19:48 < v-dogg> write ($key, &$data) 19:48 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has quit ["Bye Bye"] 19:49 < v-dogg> is that & really a good idea? 19:50 < Wombert> only very few &s are necessary 19:51 < Wombert> (cleaning up loggers right now) 19:54 < Wombert> ehm 19:54 < Wombert> what's up with FileAppender? 19:55 < Wombert> _getHandle() 19:55 < Wombert> wtf 19:57 < v-dogg> ? 19:59 < v-dogg> is something else wrong other that that underscore? 20:00 < v-dogg> have you touched something else than logging now? 20:00 < v-dogg> I'm about to commit phpdoc fixes into config, renderer and storage 20:01 < Wombert> go go go 20:02 < impl> goooo! 20:02 < Wombert> public function &format($message) 20:02 < Wombert> { 20:02 < Wombert> $return = sprintf('%s', $message->__toString()); 20:02 < Wombert> return $return; 20:02 < Wombert> } 20:02 < Wombert> this is _such_ a nonsense 20:02 < Wombert> I mean 20:02 < CIA-6> v-dogg * r891 /trunk/src/ (9 files in 3 dirs): phpdoc fixes. refs #137. 20:02 < Wombert> wtf 20:02 < Wombert> seriously 20:02 < Wombert> wtf 20:03 < v-dogg> just a redundant sprintf :) 20:03 < Wombert> I mean &format 20:03 < Wombert> yeah and the sprintf 20:03 < Wombert> lol 20:04 < Wombert> didn't pay attention to that ^^ 20:04 < v-dogg> & is also... uhm... not needed :) 20:04 < Wombert> the rotating appender is yours, right? 20:04 < Wombert> because the phpdoc says bob 20:05 < v-dogg> ah, I've missed that 20:05 < Wombert> np 20:05 < Wombert> fixed it 20:05 < v-dogg> thanks 20:05 < Wombert> don't mention it! 20:06 < v-dogg> already did! 20:06 < Wombert> even though I suffered freat privations and all 20:06 < Wombert> great even 20:06 -!- shoan [n=shoan@59.92.154.81] has joined #agavi 20:06 < Wombert> changing an @author phpdoc isn't easy, I tell ya 20:06 < Wombert> you need loads of skill and experience and 20:06 < Wombert> okay I'll shut up already 20:06 * impl grins 20:07 < impl> Wombert: I think the solution to decorators is nested actionstacks 20:07 < Wombert> and chaim 'em? 20:07 < impl> Well, let me show you what I was thinking 20:08 < v-dogg> action tree 20:08 < Wombert> go 20:08 < Wombert> anyone ever measured file_put_contents vs fwrite? 20:08 < Wombert> because in FileAppender you could just file_put_contents with FILE_APPEND flag 20:08 < Wombert> probably slower tho 20:08 < Wombert> open the file each time etc 20:16 < Wombert> we don't need &format, do we? 20:22 -!- raidman [i=armen@unaffiliated/raidman] has joined #agavi 20:22 < Wombert> arrrrggghhhhh 20:22 * Wombert bangs head against wall 20:23 < Wombert> these tests 20:23 < Wombert> they have to go 20:23 < Wombert> they suck 20:23 < Wombert> suck 20:23 < Wombert> suck 20:27 < raidman> evening guys 20:27 < Wombert> hi 20:29 < raidman> hi :) 20:29 < shoan> what sucks? I thought it rocked ;) 20:33 * impl can't stand unit testing 20:34 < Wombert> it's much needed 20:34 < raidman> impl, but how you test your apps? 20:34 < Wombert> but it gets on my nerves 20:35 < Wombert> big time 20:35 < Wombert> I awear 20:35 < Wombert> swear 20:35 < Wombert> and I swear when I must use them ^^ 20:35 < impl> Wombert: Sorry it took me so long to make this (I had to go eat lunch): http://noelism.org/flow.txt 20:35 < Wombert> impl: hmmh 20:35 < Wombert> you got mad ascii art skillz brotha 20:35 < Wombert> :D 20:36 < impl> haha 20:36 < impl> :p 20:36 < Wombert> the thing is 20:36 < Wombert> if you forward in/after the action 20:36 < Wombert> you a) have the problem that the forwarded content comes before the action's 20:36 < Wombert> b) uh 20:36 < Wombert> what was that again 20:36 < Wombert> ah yeah 20:36 < Wombert> it's presentational 20:36 < Wombert> you know? 20:36 < Wombert> slots and such 20:37 < Wombert> ah I see a view there 20:37 < Wombert> hmmh 20:37 < impl> Yeah, the view would have to treat the result of the ActionStack->getLastEntry()->getPresentation() as a variable 20:40 < raidman> Wombert, that is why you want throw away current execution flow? 20:41 < raidman> you want do that in queue against stack? 20:41 < Wombert> there are some things 20:41 < Wombert> like the fact that you can't forward in slots 20:42 < Wombert> and, due to that, probably not use a decorator inside a decorator etc 20:42 < Wombert> so the current model has weaknesses 20:42 < Wombert> so we need a new one 20:42 < Wombert> many people don't like slots much either 20:42 < Wombert> which would be another reason 20:42 < Wombert> I guess we need some major out-of-the-box-thinking to come up with something better 20:42 < impl> The major problem I see with adding a whole nested ActionStack dealie is that it would probably seriously slow down execution time 20:43 < impl> Decorators are bad enough as is 20:43 < impl> Plus it sounds complicated :\ 20:44 * Wombert ponders what to do about this test bullshit 20:44 * Wombert concludes that if the test refuses to work, he's gonna throw it away 20:44 < impl> hehe 20:44 < raidman> you can decorate your components in a module, every components coulde be somthing like modules 20:44 < raidman> this is wath i'm thinking about 20:45 < Wombert> see 20:45 < impl> What is a "component" 20:45 < Wombert> I'd also rework the entire concept of modules, actions, components a little 20:46 < Wombert> yeah exactly 20:46 < Wombert> I think modules as they are now are pretty redundant 20:46 < Wombert> not entirely, but pretty 20:46 < impl> They're less a module and more of a group 20:47 < impl> So maybe we should just scratch modules altogether and go with a more flexible grouping system 20:47 < impl> these actions belong to this group, this group, and this group 20:47 < Wombert> consider a forum 20:47 < Wombert> a forum has a public interface 20:47 < impl> Sort of like UNIX users/groups 20:47 < Wombert> and an administration area 20:47 < Wombert> and a news engine 20:47 < Wombert> too has public interface and admin area 20:48 < Wombert> now you want to group both admin interfaces together 20:48 < Wombert> and both public interfaces 20:48 < Wombert> at least in a way 20:48 < Wombert> I dare to say 20:48 < Wombert> you can't achieve that 20:48 < impl> Not with the current system 20:48 < Wombert> you can achieve the separation 20:48 < Wombert> but not the drop-in functionality 20:48 < Wombert> where you take a forum... component, drop it into an existing app, and it works 20:48 < Wombert> there'll always be customization involved but... well 20:48 * raidman is wondered really what's a module? 20:49 < Wombert> we can work towards a structure to support it 20:49 < Wombert> raidman: that really depends 20:49 < raidman> Wombert, in a CMS case? 20:49 < Wombert> for example, my modules typically are "frontend" and "backend" for public part and administration part 20:49 < Wombert> and I use sub-actions to group stuff together 20:49 < raidman> Wombert, me too 20:49 < impl> Wombert: We could do it very much like UNIX permissions actually... for the grouping I mean 20:49 < Wombert> yeah in a cms I'd have "admin" or so and "rendering" modules 20:49 < Wombert> but see 20:49 < Wombert> that's just me and my taste 20:50 < Wombert> no idea what others do 20:50 < impl> It could have a primary group 20:50 < impl> and belong to other groups 20:50 < raidman> Wombert, somthing like News/actions/Admin/Blah 20:50 < raidman> Wombert, somthing like News/actions/Public/Blah 20:50 < impl> That way we could organize the filesystem by primary group but still allow participation/interaction in other groups 20:50 < impl> Would that provide drop-in functionality? 20:53 < raidman> Wombert, hmm, you have just two modules? 20:54 < Wombert> impl: we'll never have drop-in functionality 20:54 < Wombert> because you have to customize output 20:54 < impl> I don't see why not 20:54 < impl> hm 20:54 < Wombert> well 20:54 < Wombert> you at leaast have to define your decorators in all the forum's views etc 20:54 < Wombert> which makes it nasty to begin with 20:54 < Wombert> and then there are other things 20:54 < Wombert> such as routes 20:54 < Wombert> etc etc etc 20:54 < Wombert> that's whx the milestone is 2.0 ;) 20:55 < Wombert> but I was thinking about something where at least an entire package could be dropped in in one piece, and it provides functionality for the different parts, and... ah I don't know :| 20:55 < Wombert> afaik, zimba (from switzerland, disappeared long ago) had worked on an implementation of dependency injection or something to replace decorators, slots, and solve quite a lot of problems 20:55 < Wombert> don't have a clue what he did tho 20:56 < raidman> Wombert, check this out please: http://osc-dev.opensourceclub.org/browser/trunk/webapp/modules 20:57 < Wombert> raidman: yes I believe that's the way it is supposed to work 20:57 < Wombert> but 20:57 < Wombert> you then have the problem that you can't separate things, you know? 20:57 < Wombert> you have createnewsaction and displaynewsaction in the same place 20:58 < Wombert> that's why I suggested we change this completely in agavi 2 20:58 < Wombert> I just don't know how yet ;) 20:58 < raidman> hmm 20:58 < raidman> you mean it could be in different plase, but what's is benefit? 20:59 < raidman> place* 21:00 < eremit> Are you realling thinking about 2.0 already ? oO 21:00 < raidman> afaik, zimba implement dependency injection for agavi? 21:01 < impl> eremit: They have open tickets for it and everything 21:02 < eremit> yeah, but if you start thinking about 2.0 maybe i should write down a wishlist ... :P 21:03 < eremit> Wombert, how difficult would it be to create ONE agavi.php file which has all agavi classes included already ? 21:03 < Wombert> what for 21:03 < eremit> at least the elemental ones 21:03 < Wombert> that is massive overhead 21:03 < eremit> performance 21:03 < eremit> i dont care about ram ... :P ... ram is cheap and fast ... but io access is slow ... 21:04 < impl> Wombert: That's what M2 did 21:04 < eremit> i just thinking about where to get the last bit of performance out of the app ... 21:04 < impl> It had all the core classes in mojavi-classes.php or somesuch 21:04 < impl> And you just included it 21:05 < Wombert> right 21:05 < eremit> prado did the same ^^ they have a package with all core classes in one file ... 21:05 < Wombert> do it yourself? :p 21:05 < eremit> sure i will, i only ask if it's complicated ... are there more requires than the autoload and bootstrap requires 21:06 < Wombert> no 21:06 < Wombert> well compile.xml 21:06 < Wombert> but 21:06 < Wombert> that doesn't really matter much 21:06 < Wombert> it's just another file with many includes in it 21:10 < eremit> ty i will check it tomorrow and make a profiling about it ... 21:16 < raidman> Wombert, impl, shoan, eremit I have to go guys, night ;) 21:16 -!- raidman [i=armen@unaffiliated/raidman] has quit ["#killall raidman"] 21:22 < v-dogg> doesn't compile.xml do just that? 21:22 < v-dogg> it parses all the core code into one file 21:23 < v-dogg> just add all classes you want 21:27 < Wombert> not all of them 21:27 < Wombert> some are included in Agavi::bootstrap(); 21:27 < v-dogg> but just a few 21:27 < Wombert> yeah 21:28 < Wombert> I don't get the point of it either 21:28 < v-dogg> so that's shouldn't be a problem 21:29 < v-dogg> I counted 6 21:30 < v-dogg> ah there's more 21:30 < CIA-6> david * r892 /trunk/ (17 files in 7 dirs): cleaned up logging and fixed the pile of stuff I broke in r874. Can't be bothered to write down the tickets this refs. 21:31 < Wombert> right 21:31 < Wombert> I deserve my pizza now 21:31 < Wombert> brb 21:45 -!- [1]eremit [n=eremit@p5490C758.dip.t-dialin.net] has joined #agavi 22:03 -!- eremit [n=eremit@p5490D874.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 22:03 -!- [1]eremit is now known as eremit 22:09 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has left #agavi [] 22:51 -!- shoan [n=shoan@59.92.154.81] has quit [] 22:52 -!- shoan [n=shoan@59.92.154.81] has joined #agavi --- Day changed Thu Aug 03 2006 00:35 < Macca> Wombert? 00:37 < Wombert> eah? 01:19 < Macca> hi :D 01:19 < Macca> [Wed Aug 02 10:50:11 2006] [error] [client 203.206.205.212] PHP Fatal error: Allowed memory size of 8388608 bytes exhausted (tried to allocate 1204948 bytes) in /var/www/agavi/src/renderer/AgaviPhpRenderer.class.php on line 77, referer: https://development.maccasoft.com.au/lld/direct/www/client/find-contractor 01:19 < Macca> got that last night 01:19 < Macca> err yesterday 01:23 < Macca> hmm.. i don't like how you have config.php and config/ in the same directory. 01:23 < Macca> auto complete always messes up 01:23 < Macca> can't you rename one to settings or something? 01:26 < Macca> Configuration directive "core.app_dir" not defined, terminating... 01:26 < Macca> ??? 01:27 * Macca pokes Wombert 01:28 < Wombert> 8MB is not quite enough nowadays 01:28 < Macca> is that a php.ini bit? 01:28 < Wombert> memory_limit 01:29 < Wombert> core.webapp_dir => core.app_dir 01:29 < Macca> right 01:30 < Macca> so that's just updating config.php ? 01:30 < Wombert> maybe configs, too 01:30 < Wombert> if you have autolaods etc defined 01:31 < Macca> nah 01:31 < Macca> so yeah, can we get config.php or config/ renamed? :) 01:32 < Wombert> rename it yourself ;) 01:32 < Wombert> init.php or so 01:32 < Macca> no :( 01:33 < Macca> hmm. what's the best way to write a database wrapper/layer? 01:33 < Macca> so i don't have to write the SQL throughout my views/actions ? 01:34 < Wombert> use propel 01:34 < Wombert> and/or put your stuff into models 01:34 < Wombert> that's what they are for 01:35 < Macca> models? 01:36 < Macca> i haven't seen any demos of models :( 01:36 < Macca> and propel is too hard :( 01:40 < Macca> oh man 01:40 < Macca> i have to write an XML describing my database 01:40 < Macca> ffs! 01:42 < Macca> my database is massive. (in a sense..) i cbf doing the XML fo rit 01:43 < Macca> surely there's another way? 02:13 < Wombert> propel ;) 02:13 < Wombert> it's not hard 02:13 < Wombert> plus you're a developer 02:13 < Wombert> an engineer 02:14 < Wombert> you can figure that out! 02:14 < Wombert> or use models 02:14 < Wombert> they encapsulate business logic 02:14 < Wombert> anyways, I really gotta go, sorry man 02:19 < Macca> i don't understand models :( 02:21 < Wombert> figure out the best way to use them 02:21 < Wombert> $usr = $this->getContext()->getModel('User')->loadById(134); 02:21 < Wombert> or so 02:21 < Wombert> ->create(); 02:21 < Wombert> ->save(), 02:21 < Wombert> whatever 02:21 < Wombert> I have no idea either, I don't use them 02:21 < Wombert> anyways, I really gotta go to bed now 02:32 -!- benny`work [n=benny@p54AC42F7.dip0.t-ipconnect.de] has quit ["unset($this);"] 02:37 -!- Wombert [n=Wombert@dslb-088-064-012-112.pools.arcor-ip.net] has quit [] 02:39 < Macca> he has no idea how to use models.. but he created them? :S 02:40 < Macca> how do i even make models? :S 03:16 < Macca> v-dogg? 03:31 < Macca> IMPL 04:21 < Macca> how do i make route->gen() give a full URL (with domain) ? 05:35 < Macca> hmm 05:35 < Macca> it'd be so awesome if i could use quickform in agavi 07:02 < Macca> ................. 07:29 < v-dogg> huomenta 07:29 < v-dogg> Macca: gen('name', $params, array('relative'=>false) ); 07:30 < Macca> huomenta? 07:31 < Macca> can i start writing the docs for again? :D 07:31 < v-dogg> http://www.canuckabroad.com/language/finnish.shtml 07:32 < Macca> ? 07:32 < Macca> oh 07:32 < Macca> morning 07:32 < Macca> you're finnish? 07:33 < v-dogg> yep 07:34 < Macca> cool 07:34 < Macca> have you seen the pear HTML_QuickForm thingy? 07:35 < v-dogg> no 07:35 < Macca> it's pretty cool 07:35 < Macca> like a form generator 07:35 < v-dogg> but if it's pear, I probably don't want to either :) 07:35 < Macca> and there's HTML_QuickForm_Controller.. which handles multiple page forms 07:35 < Macca> and validation and stuff 07:35 < Macca> it's quite handy. it would be cool if something similar were available in agavi? :) 07:36 < v-dogg> is there a reason you cannot use it with agavi? 07:36 < Macca> i'm scared :( 07:37 < v-dogg> I need to head to work 07:38 < Macca> aww :( 07:38 < v-dogg> I'll be there within 20 mins 07:38 < Macca> woohoo 07:38 < Macca> see ya then :D 07:38 < v-dogg> see ya 07:42 < Macca> there yet? 07:42 < Macca> how bout now? 07:45 < Macca> wtf 07:45 < Macca> did someone break the exception page? 07:45 < Macca> some images aren't being drawn properlly :S 07:56 < Macca> back now 07:56 * Macca shrugs 07:56 < Macca> prob ff screwing up 07:56 < Macca> what's wrong with returning an array in an action's execute to redirect to another view? 07:59 < Macca> i gotta go to tafe :( 07:59 < Macca> cya 08:00 < Macca> meh 08:00 < Macca> no point 08:00 < Macca> still here :D 08:04 < v-dogg> tafe? 08:04 < v-dogg> < Macca> what's wrong with returning an array in an action's execute to redirect to another view? 08:04 < v-dogg> that doesn't work? 08:05 < Macca> it does 08:05 < Macca> but wombert always says "dont do that" 08:05 < Macca> he says to do it in a view 08:05 < Macca> why? 08:05 -!- Wombert [n=Wombert@dslb-088-064-012-112.pools.arcor-ip.net] has joined #agavi 08:05 * Macca hides 08:06 < v-dogg> it's a separation-of-logic thing 08:06 < v-dogg> I'm not sure if redirection to another _view_ is that bad 08:07 < v-dogg> are you sure Wombert wasn't talking about redirecting to another action? 08:08 < v-dogg> but yeah, maybe you shouldn't redirect to another actions views from an action 08:08 < Macca> not sure. 08:08 < Macca> :) 08:09 < v-dogg> but what's that HTML_QuickForm 08:10 < Macca> what about it? 08:10 < Macca> http://user.maccasoft.com.au/~brendanm/test/examples/qf/ 08:10 < Macca> it generates forms 08:10 < Macca> have a look at those examples 08:10 < Macca> (some don't work btw :) 08:10 < Macca> http://user.maccasoft.com.au/~brendanm/test/examples/qf/rules-builtin.php 08:13 < v-dogg> I can't see no reason why you couldn't use that (generation and client-side validation) with agavi 08:13 < Macca> yeah i know 08:13 < Macca> but it's the controller side 08:13 < Macca> sec 08:13 < Macca> http://user.maccasoft.com.au/~brendanm/test/examples/regWizard.php 08:13 < Macca> like that 08:14 < v-dogg> hmm... 08:14 < v-dogg> I still see no problem :) 08:14 < Macca> mmm 08:18 < Macca> and how do i create models? 08:19 < eremit> morning ... 08:19 < Macca> how's the rsi? 08:20 < eremit> better ^^ ... but working on the keyboard only with bandages with steel inlets on the hands :/ 08:20 < v-dogg> Macca: create as in code, or create at runtime? 08:20 < Macca> as in code 08:20 < Macca> wonderful. 08:21 -!- raidman|Work [n=raidman@62.60.194.3] has joined #agavi 08:23 < v-dogg> Macca: just code... you know... morning 08:24 < v-dogg> huomenta 08:24 < Macca> then what? 08:24 < Macca> how do i use it? 08:25 < v-dogg> just like a normal class or via $this->getContext()->getModel('MyModel'); 08:26 < Macca> how do i make sure it's loaded? 08:26 < Wombert> ? 08:27 < v-dogg> morn wombie 08:28 < Wombert> cheers man 08:28 < Macca> guten morgen wombert. 08:28 < eremit> morning raidman, morning wombert :P 08:30 < Macca> it's afternoon actually. 08:30 < Macca> Thursday, August 03rd 2006 15:30:58 GMT+10 08:36 < raidman|Work> eremit, :) 08:39 < v-dogg> Macca: in what part of Australia you live? 08:39 < v-dogg> you are +7 h from finland 08:39 < Macca> victoria 08:39 < v-dogg> ok 08:39 < Macca> i'm GMT+10 08:40 < Macca> AEST 08:40 < v-dogg> I was in NZ '98 08:41 < v-dogg> exchange student 08:41 < Macca> cool 08:41 < Macca> NZ != AU :D 08:42 < v-dogg> guess what, I knew that :) 08:42 < Macca> you'd be amazed how many people don't :S 08:43 < v-dogg> people are ignorant 08:43 < v-dogg> mostly just americans though 08:43 < Macca> yep 08:44 < Wombert> hey 08:44 < Wombert> did I mention 08:44 < Wombert> that I could become an australian citizen 08:44 < Wombert> I found that out the other day 08:44 < Macca> YAY 08:44 < Macca> GOOOOOOO WOMBERT 08:44 < Wombert> I just need the time to talk to a lawyer and figure out how NOT to lose the german one when I apply for the australian 08:45 < Macca> my brother gave up his US citizenship when he went to germany 08:45 < Wombert> this really sucks if you ask me 08:45 < Macca> what does? 08:45 < Wombert> germany and well... bulgaria or so are the only countries that actively ban you from having a dual citizenship 08:46 < Wombert> most countries just don't care at all 08:46 < Macca> one thing that shits me with PHP is if you have a function that returns an array you can't do my_func()[0] 08:46 < Macca> aww :( 08:46 < Wombert> list($foo) = my_func(); 08:46 < Macca> we had a german exchange student once 08:47 < eremit> isn't {my_func()}[0] possible ? 08:47 * Macca shrugs 08:47 < Wombert> nope 08:47 < Wombert> php can't do this kind of dereferencing 08:47 -!- impl [n=impl@httpcraft/php/nforbes] has quit [Read error: 60 (Operation timed out)] 08:47 < Wombert> in fact, if a method returns an object, you can only dereference that since php5 08:48 < Macca> wow. 08:48 < Wombert> $this->getContext()->getRequest(); etc 08:48 < Macca> man you would die if you saw the code behind this, wombert. 08:48 < Wombert> if you say so ;) 08:49 < Macca> i'm ashamed of it :( 08:55 < eremit> gna ... anybody knows a good javascript calendar picker ? ... 08:58 < Wombert> you bet 08:58 < Wombert> http://www.dynarch.com/projects/calendar/ 09:00 < eremit> ty ... i'll try it ... 09:00 < eremit> :) 09:00 < Wombert> the best there is hands down 09:03 < Macca> can classes exist inside classes? 09:06 < Macca> i don't get it.. 09:06 < Macca> are templates part of a class? 09:09 < v-dogg> kinda 09:09 < v-dogg> they are included inside a renderer 09:09 < v-dogg> so you can use $this->getContext() for example 09:10 < Macca> yeah. 09:10 < Macca> which causes problems if i wanted a class inside my template :S 09:10 < v-dogg> you want to declare a class inside a template? 09:10 < v-dogg> wanna tell us why?-) 09:10 < Macca> .. quickform :) 09:11 < Macca> http://user.maccasoft.com.au/~brendanm/test/examples/regWizard.phps 09:11 < Macca> i need classes like that 09:12 < v-dogg> ok... 09:13 < Macca> =) 09:13 < Wombert> why are you using quickform again? 09:14 < Macca> makes forms easier? 09:14 < Wombert> forms _are_ easy with agavi! 09:14 < Macca> ... 09:14 < Macca> i still have to do the HTML, etc.. 09:14 < Macca> plus going back on forms 09:14 < Macca> changing stuff etc 09:14 < Macca> vlidation 09:14 < Macca> validation* 09:15 < Wombert> uh 09:15 < Macca> i know what you're going to say! 09:15 < Wombert> form population filter will mark erroneous forms and their labels 09:15 < Wombert> and insert values again 09:15 < Wombert> check checkboxes 09:15 < Macca> but it's too hard :( 09:15 < Wombert> select select fields etc etc etc 09:15 < Wombert> ?????? 09:16 < Wombert> it is _not_ too hard 09:16 < Wombert> v-dogg: wanna help him make fpf work? 09:16 < Wombert> I'm in a hurry 09:16 < Macca> hehe 09:23 < Macca> i need pages with states. 09:24 < Macca> for this one in particular i need: page1) input search criteria, page2) results from search, with option to select result and continue to page3 with that result, or return back to page1 (restoring defalt values), page3) action to take with selected result, leading on to page4a/4b, etc... 09:24 < Macca> quickforms using HTML_QuickForm_Controller seems ideal for this. 09:27 < Macca> feel free to convince me otherwise... 09:28 < Macca> ... 09:30 < Macca> ? 09:30 < Macca> fine :( 09:31 -!- v-dogg [n=vmakinen@130.232.39.52] has quit [Read error: 110 (Connection timed out)] 09:33 < Wombert> no time 09:33 < Wombert> sorry 09:33 < Wombert> ask v-dogg when he is bag 09:33 < Wombert> back 09:33 < Wombert> lol 09:33 < Wombert> cu tomorrow 09:33 -!- Wombert [n=Wombert@dslb-088-064-012-112.pools.arcor-ip.net] has quit [] 09:41 < Macca> :( 09:42 < Macca> this is to fkn hard! 09:42 * Macca cries 09:50 < Macca> someone make v-dogg come back! 10:02 -!- shoan [n=shoan@59.92.154.81] has quit [] 10:09 < Macca> nnnooooo 10:09 < Macca> come baaack 10:36 -!- shoan [n=shoan@61.95.206.107] has joined #agavi 10:44 < ttj> A slightly off topic question, but anyone competent with XML Schemas here? 10:44 < ttj> http://phpfi.com/136840 12:01 < Macca> where's v-dogg ? :( 12:37 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit [Read error: 104 (Connection reset by peer)] 13:01 -!- eremit [n=eremit@p5490C758.dip.t-dialin.net] has quit [" HydraIRC -> http://www.hydrairc.com <- 100,000+ downloads can't be wrong"] 13:01 -!- eremit [n=eremit@p5490C758.dip.t-dialin.net] has joined #agavi 13:38 -!- shoan [n=shoan@61.95.206.107] has quit [] 14:17 < eremit> oO it'S getting empty here ... 14:19 < Macca> uhoh :( 14:20 < eremit> gna ... where's wombert ... 14:20 < eremit> $this->getContext()->getController()->forward() throws an exception because of this bloody parameter locking 14:22 < Macca> parameter locking? 14:23 < Macca> where are you calling that? 14:23 < eremit> yes the $request parameters are locked ... so nobody can change them ... 14:23 < eremit> it's called automatically ... try a $this->getContext()->getRequest()->setParameter('module','foobar'); in an action 14:23 < eremit> and you see the exception :P 14:40 -!- shoan [n=shoan@59.92.154.81] has joined #agavi 15:28 -!- shoan_ [n=shoan@59.92.170.117] has joined #agavi 15:29 < shoan_> is this room quite or am I not connected? 15:29 < eremit> quiet 15:29 -!- raidman|Work [n=raidman@unaffiliated/raidman] has joined #agavi 15:29 < shoan_> damn typo :) 15:29 < eremit> heh ... i guessed right ;) 15:30 < eremit> shoan_, are you actively using agavi atm ? 15:30 < shoan_> actively used 0.9/0.10 15:30 < eremit> hmm dmn ... 15:30 < shoan_> not the new one 15:30 < shoan_> not *yet* 15:30 < eremit> i have some problems using the forward() in the new one ... 15:30 < shoan_> starting a 2 projects next week using the new one 15:31 < shoan_> not sure if I can help you there :( 15:31 < eremit> oO ... nice ... started my current one last week and have to be finished with the first prototype at saturday :/ 15:32 < eremit> i don't think you can help with the redirect... i guess its a bug because of womberts lock request parameter paranoia 15:32 < shoan_> i don't use new technology for tight deadlines ;) 15:32 < eremit> heh ... i only had the choice of _new_ technology ... had an outtime of nearly 4 month ... 15:32 < eremit> why refreshing old knowledge when learning new is not really slower 15:32 < shoan_> point :) 15:33 -!- shoan [n=shoan@59.92.154.81] has quit [Read error: 60 (Operation timed out)] 15:33 -!- shoan_ is now known as shoan 15:33 < shoan> finally :) 17:12 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit ["#killall raidman"] 17:13 -!- raidman|Work [n=raidman@62.60.194.3] has joined #agavi 17:26 -!- impl [n=impl@httpcraft/php/nforbes] has joined #agavi 18:30 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has joined #agavi 19:31 < raidman|Work> good night shoan impl eremit :) 19:31 < raidman|Work> good night agavi guys :) 19:31 -!- raidman|Work [n=raidman@unaffiliated/raidman] has quit ["#killall raidman"] 19:43 -!- v-dogg [n=vmakinen@mcinen-dsl.utu.fi] has joined #agavi 19:50 < eremit> hey v-dogg 19:50 < v-dogg> heya 19:50 < eremit> are you involved in the parameter lock toggle part of request ? 19:51 < eremit> there's a serious bug or an api change without any notification ... 19:51 < v-dogg> not really 19:51 < v-dogg> what's wrong? 19:51 < eremit> [13:18] $this->getContext()->getController()->forward() throws an exception because of this bloody parameter locking 19:51 < impl> I don't understand the purpose of parameter locking 19:51 < v-dogg> yep, that's intentional :) 19:51 < impl> What is it supposed to do? Prevent race conditions? 19:52 < eremit> and how should i forward ? 19:52 < impl> (But Agavi isn't threaded?) 19:52 < v-dogg> eremit: return array('module', 'action') from view::execute 19:52 < eremit> i want to change action in action ... 19:52 * impl blinks 19:52 < eremit> not in view 19:53 < eremit> impl, i dont understand it either, and i really dislike it ... 19:53 < v-dogg> impl: it prevents you from using un-validated parameters and forwarding from action 19:53 < impl> Why wouldn't you want to forward from the action? 19:53 < eremit> i want the freedom to use my data and change my data wherever i like ... 19:53 < v-dogg> because forwarding from action is where chaos lies 19:53 < v-dogg> :P 19:53 < impl> I forwarded from the action all the time in Mojavi 19:53 < eremit> and how should i do the following ... : 19:53 < impl> then I just returned VIEW_NONE 19:54 < impl> I don't want to make a whole separate view just to forward :\ 19:54 < eremit> ^^ exactly ... 19:54 < v-dogg> impl: it might be frustrating at times but forwarding is not part of actions responsibilities 19:55 < eremit> and i would to have business logic in the view 19:55 < eremit> huh? 19:55 < impl> Forwarding seems like action logic to me 19:55 < v-dogg> it's an extreme point of view but forwarding is view's job 19:55 < eremit> sorry, that's nonsense ... you want to enforce devs to write a view only to forward ? ... that's needless performance burning ... 19:56 < v-dogg> consider an AddProduct action 19:56 < eremit> consider an Index page different for User and Guests ... 19:56 < v-dogg> you make it for web use and on Success you forward to ShowPrdLIsting 19:56 < v-dogg> but then you'll add a, say, xmlrpc context 19:57 < v-dogg> and add products from a desktop app 19:57 < CIA-6> dominik * r893 /branches/dominik-validation_mods/src/ (3 files in 3 dirs): add file upload validator 19:57 < v-dogg> you don't want to forward to ShowPrdListing anymore 19:57 < v-dogg> because it's a presentational thing 19:57 < eremit> yes, but shouldn't it the developers choice and not the framework devs choice ? 19:58 < eremit> for me atm, it's like enforcing special version names :PP 19:58 < impl> v-dogg: I have to ponder that 19:59 < eremit> the problem should be mentioned in a documentation but please not be enforced ^^ 19:59 < v-dogg> I admit that it's an extremist's point of view (and I was against it myself first :) 19:59 < impl> It sort-of makes sense 20:00 < impl> However 20:00 < v-dogg> maybe it has to be discussed more 20:00 < eremit> sure, in some cases ... 20:00 < impl> Making View::execute() return an array to another module/action 20:00 < impl> That makes no sense to me 20:00 < v-dogg> that I have to agree with 20:00 < eremit> but at least there should be a way to unlock the parameters, because there are cases where forward in views in needless overhead 20:00 < eremit> or impl solution ... another nice way ^^ 20:00 < v-dogg> it's ambiguous 20:01 < impl> Yeah 20:01 < eremit> maybe returning a special actionforward object s.t. like that 20:01 < impl> How do actions work with different output types again? 20:02 < v-dogg> we'll have to discuss this when wombie gets back 20:02 < impl> Is there like an executeHTML() and executeXMLRPC()? 20:02 < v-dogg> no 20:02 < v-dogg> actions are supposed to be output-type-indifferent 20:02 < v-dogg> hence no forwarding too ;) 20:03 < eremit> hehe ... surely i will discuss it with him ... at the moment the whole locking is disabled in my installation 20:03 < impl> hmmmmf 20:04 < eremit> and the routing doesn't like complex regex too atm ^^ 20:04 < eremit> another point which took me a hour today *grin* 20:04 < impl> You have to complex regex inside parameter names 20:04 < impl> So it can figure out how to regenerate it 20:05 < eremit> i wanted something like this: ^/LostPassword(/(uid:[0-9]*)/(code:[0-9a-fA-F]*))?$ 20:05 < impl> Yeah but see then it doesn't know how to regenerate it 20:06 < eremit> i havent taken a deep look into the routines yet ... 20:07 < impl> You should make two routes probably 20:07 < impl> Or rather a route and a subroute 20:07 < v-dogg> that should be possible eremit 20:08 < v-dogg> but not with that syntax 20:08 < eremit> how then ? 20:08 < eremit> the problem was if i have LostPassword without params gen() produced LostPassword// which simply looked ugly 20:08 < eremit> impl, ty didn't knew subrouting is possible ... 20:09 < impl> :) 20:09 < impl> Take a look at the sample app routing config 20:11 < v-dogg> ^/LostPassword{/(uid:[0-9]+)} and subroute {/(code:... )} 20:11 < eremit> nice ... ty i'll try it ... 20:12 < eremit> would redunce another 6 actions down to 3 ... 20:13 < eremit> i have a bad wish for agavi 2.0 ... 20:13 < eremit> please make the view finding configurable ... 20:13 < impl> What do you mean? 20:13 < v-dogg> what do you mean? 20:13 < impl> hehe 20:13 < v-dogg> echo off 20:14 < eremit> hmm ... currently all views have to be in views/ ... i would be nice, if you can configure it more flexible ... like views// 20:14 < eremit> because this: LostPasswordConfirmErrorTimeoutView.class.php is really really painfull to type ;) 20:15 * impl blinks 20:15 < eremit> action: LostPasswordConfirm view: ErrorTimeout 20:15 < v-dogg> hmm... 20:16 < v-dogg> I have max 3 views per action and all named Input, Success or Error 20:16 < eremit> understand the problem ? ... you can really easily produce looong filenames if you have actions with 10 or 15 views ... 20:16 < v-dogg> are you sure you need all those views? 20:17 < eremit> hmm ... i could reduce them surely... because 7 of my 10 views are error views, but this would mean to have high logic in the ErrorView 20:17 < eremit> which i usually keep really low ... and do only output related conversions etc in there ... no consideration which template to load etc ... 20:18 < impl> Why don't you just have one error template that supports different messages as sent by $request? 20:19 < eremit> because our designer would never understand that ^^ .... 20:19 < impl> Just tell him not to touch it! 20:19 < eremit> he want's to open a smarty template and write the problem in there ... the possible solutions etc ... 20:19 < impl> :\ 20:20 < eremit> surely i could load the whole text from an external source ... and tell him ... but your text, description and solutions in there ... 20:20 < eremit> but that'S another file to load, another template to parse ... etc ... 20:21 < impl> Maybe your designer needs to learn some PHP ;) 20:21 < eremit> no ... 20:22 < eremit> i have bad experience with designers and php ... they tend to mess up the whole application ... 20:22 < eremit> they don't even have write access to php files :P only to .html and .tpl 20:22 < impl> haha 20:24 < eremit> that'S why i have so many views ... and there will be more as soon as i enable the xmlrpc and xul parts of the app 20:24 < impl> O_o 20:26 < eremit> :P yeah it's a complex app ... and i really hope agavi can manage the load it will produce ... 20:35 < eremit> and i really miss some basic scaffolding and crud support in agavi :P 20:35 < eremit> maybe i should write a code generator for it *thinking loud* 20:35 < impl> Scaffolding is for the weak! 20:36 < eremit> sure but if you boss comes in and tells you you only have of weeks to work it out ... it comes in handy 20:40 < v-dogg> what sort of scaffolding? 20:40 < impl> v-dogg: Like rails has 20:40 < v-dogg> I'm not that familiar with RoR 20:41 < impl> Basically, you tell it how your DB works and it automatically makes list/edit/add actions 20:41 < v-dogg> like 'agavi action|project|module' but better? 20:41 < v-dogg> ah 20:41 < v-dogg> you need propel :) 20:41 < eremit> you know symfony ? cake ? 20:41 < impl> Symfony is pathetic 20:42 < impl> Never used cake though 20:42 < eremit> propel is the right direction yes ... but the whole frontend part is missing ... 20:42 < eremit> cake is nice, but sticked too much with naming conventions and php4 support ... 20:42 < impl> Cake's naming conventions are _weird_ 20:42 < eremit> symfony is bloody overloaded but has some nice ideas for rad development ... 20:42 < impl> I'm looking at its class list right now and wtf'ing 20:43 < impl> http://api.cakephp.org/hierarchy.html 20:43 < v-dogg> eremit: I use visual db designer to create tables and just create a propel schema from and existing db 20:43 < eremit> yes they are ... really weird ... 20:43 < eremit> v-dogg: me too :P ... but i'm missing the frontend part for the crud operations ... 20:43 < v-dogg> crud operations beeing? 20:44 < eremit> basically creating the source for Create, Read, Update, Delete operations on every table ... 20:44 < impl> v-dogg: This is all WEB TWO POINT OH terminology ;) 20:44 < eremit> atm you have to create every action, view etc by yourself ... i assume you use propel for the models like i do ... 20:45 < v-dogg> yes 20:45 < eremit> if you really want to make rapid development the ror way is nice ... you just tell him, to create actions and views for these four operations on every table ... 20:45 < v-dogg> creating actions with agavi script is easy and manipulating data with propel is easy 20:45 < eremit> afterwards you only need to change the parts you really need ... 20:46 < v-dogg> easy + easy = very easy ;) 20:46 < eremit> yes but agavi can only create basic actions ... not full flavoured ... 20:46 < eremit> loool 20:46 < v-dogg> yeah, I know what you mean :) 20:47 < eremit> i had a podcast showing it in effect ... ;) ... nice if you boss stands on your tiptoes which usually every dev knows ... 21:00 < eremit> symfony shows in their admin generator screencast what i meant v-dogg ... 21:01 < eremit> and ruby in their 15-min blog tutorial ... but that's rather RoR specific 21:01 < impl> I can't stand the RoR guy 21:01 < impl> God 21:01 < impl> If he says "oops" one more time 21:01 < impl> I will stab him 21:01 < eremit> loool ... sorry i don'T have sound enabled for movies here :P 21:02 < eremit> would disturb my audio book :P 21:02 < impl> eremit: Think 19-year-old college geek (high-pitched voice, gay accent) 21:02 < eremit> i think it's the micro... and they have speed up the movie to get the 15min mark ... 21:05 < eremit> but anyhow the ror thing is impressive 21:11 < v-dogg> that sort of building is never going to be a part of agavi core 21:13 < v-dogg> but hopefully we'll have some agaviforge someday to share some agavi add-ons 21:13 < eremit> would be really nice ... the autogenerate code shouldn't be that problem using phing 21:14 < v-dogg> the whole thing should be fairly easy with propel and phing 21:15 < v-dogg> but I've had enough of computers for one day now 21:15 < eremit> yeah, currently i have no clue about how to parse the xml of propel within phing 21:15 < v-dogg> gonna go watch telly :) 21:15 < eremit> nice evening ^^ ... 21:15 < v-dogg> eremit: you have to write a task to do it 21:15 < eremit> :( ... 21:16 < v-dogg> that's simple 21:47 -!- [1]eremit [n=eremit@p5490DDFE.dip.t-dialin.net] has joined #agavi 22:04 -!- eremit [n=eremit@p5490C758.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 22:04 -!- [1]eremit is now known as eremit 22:36 -!- Wombert [n=Wombert@dslb-088-064-004-049.pools.arcor-ip.net] has joined #agavi 22:47 < v-dogg> heyyy Wombert 22:51 < v-dogg> I installed a small agavi app to another server with cgi/fastcgi setup 22:51 < v-dogg> and routing didn't work :) 22:52 < v-dogg> but that was easily fixed 22:53 < v-dogg> because routing just took reques uri ($ru) from the wrong variable 22:53 < v-dogg> but I'm off to bed now. we'll talk more tomorrow 22:53 < v-dogg> night 22:54 < v-dogg> (AgaviWebRouting.class.php:69 was to blame) 22:54 < v-dogg> night #2 22:55 * impl pokes Wombert 22:55 < impl> Wombert: I don't think I like how forwarding has changed :\ 22:55 < impl> With the parameter locking thingies --- Day changed Fri Aug 04 2006 01:01 -!- Macca` [i=brendanm@maccasoft.com.au] has joined #agavi 01:13 -!- Macca [i=brendanm@maccasoft.com.au] has quit [Connection timed out] 01:15 < Wombert> re 01:15 < Wombert> hi impl 01:15 < impl> Hello Wombert 01:26 < Wombert> y000 01:26 < Wombert> what exactly don't you like about the new forwarding? :) 01:37 -!- devosc [n=devosc@rrcs-24-172-199-238.central.biz.rr.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"] 01:45 < impl> Wombert: I don't think it's necessary to require a whole view for it 02:13 < Wombert> it is! 02:13 < Wombert> forwarding is a presentational thing 02:23 -!- Wombert [n=Wombert@dslb-088-064-004-049.pools.arcor-ip.net] has quit [] 03:07 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 03:13 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 05:01 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 05:16 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 05:21 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 05:25 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 05:51 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has joined #agavi 07:32 < v-dogg> huoemtna 07:41 -!- shoan_ [n=shoan@59.92.155.45] has joined #agavi 07:54 -!- shoan [n=shoan@59.92.170.117] has quit [Read error: 110 (Connection timed out)] 08:31 < eremit> morning 08:32 < eremit> ah as i see, the forward thing has been discussed ... 08:33 < eremit> v-dogg there ? 08:33 < v-dogg> _o/ 08:33 < eremit> [21:49] I installed a small agavi app to another server with cgi/fastcgi setup 08:33 < eremit> [21:49] and routing didn't work :) 08:33 < eremit> ^^ for me it works ... 08:33 < eremit> using fcgid here ... 08:33 < v-dogg> ok 08:34 < eremit> maybe it's cgi implementation dependant ? ... suPHP works too .. 08:37 < v-dogg> it depends on what env/server variables are used and what values they get 08:37 < v-dogg> but why and how that happens I have no clue 08:38 < eremit> which fastcgi setup have you used ..? 08:39 < eremit> there are several solutions out ... 08:40 < v-dogg> Server API: CGI/FastCGI 08:41 < eremit> ahaha ... :P the solution on the system level is interesting .. there are several api implementation out ... 08:41 < v-dogg> a few php.ini directives also make a difference 08:41 < eremit> fast-cgi fcgid and f-cgi are the ones i'm aware of ... 08:42 < eremit> and some of them behave special in php context ... 08:43 < v-dogg> but like I said, this time fixing was easy. agavirouting just used a wrong env variable 08:43 < eremit> hehe ... 08:44 < v-dogg> this just complicates things even more 08:45 < v-dogg> we have to find a way to decide what route parsing method to use 08:45 < v-dogg> route parsing itself is not difficult when you know what variables to use 08:45 < eremit> cant you sanity check the vars ? 08:46 < v-dogg> some sort of sanity check is probably necessary 08:47 -!- impl [n=impl@httpcraft/php/nforbes] has quit ["Leaving"] 09:04 -!- raidman [i=armen@unaffiliated/raidman] has joined #agavi 09:12 -!- shoan_ [n=shoan@59.92.155.45] has quit [] 09:22 < CIA-6> v-dogg * r894 /trunk/src/ (2 files in 2 dirs): phpdoc embellishments 09:40 -!- raidman [i=armen@unaffiliated/raidman] has quit [Nick collision from services.] 09:40 -!- raidman [i=armen@unaffiliated/raidman] has joined #agavi 09:43 < v-dogg> mornign raidman 09:44 < eremit> morning raidman ^^ 09:45 -!- raidman_ [i=armen@unaffiliated/raidman] has joined #agavi 09:46 -!- raidman [i=armen@unaffiliated/raidman] has quit [Nick collision from services.] 09:46 -!- raidman_ is now known as raidman 09:47 < v-dogg> israel is bombing your network connections?-p 09:47 < raidman> lol 09:47 < eremit> O_o 09:48 < raidman> hi guys :) 09:48 < raidman> what's up? 09:49 < v-dogg> lazy friday at the office 09:49 < eremit> busy friday at the office ... tomorrow the first prototype of my current app has to be released :/ 09:50 < v-dogg> who releases prototypes on saturday?-) 09:50 < v-dogg> you weirdos ;) 09:50 < raidman> and normal holiday at home for me 09:50 < raidman> :) 09:50 < v-dogg> raidman: where were you again? Iran? 09:51 < raidman> yeah :) 09:51 < eremit> v-dogg: a company which works tightly on timeline ... and monday is the conference ... so i have sunday for fast debugging 09:51 < v-dogg> debuggin is for customers :) 09:52 < eremit> loool ... tell that the customers ... sadly there much money ingame ... and the most interesting parts are the follow up jobs ... 09:52 < eremit> that's why the first one has to be as clean as possible 09:53 < v-dogg> raidman: does that Israel vs. Hizbollah/Lebanon mess affect your daily life at Iran at all? 09:54 < eremit> O_o political discussions ... 09:54 < v-dogg> no, not political at all 09:54 < raidman> no here only one who interested about that is government, here no one interested about that 09:54 < eremit> you are at the edge ^^ v-dogg ;) 09:55 < v-dogg> but I'm not taking any sides :) 09:55 < eremit> hehe ... 09:55 < raidman> ahmadinejad (our stupid president) advertised about, but really no one interested about that :) 09:55 < v-dogg> just interested how the daily life goes on there 09:55 < v-dogg> you president is surely taking sides :) 09:56 < raidman> very normal :) 10:12 -!- devosc [n=devosc@user-0cdv7qr.cable.mindspring.com] has left #agavi [] 10:22 < raidman> I have to go guys, bye :) 10:22 -!- raidman [i=armen@unaffiliated/raidman] has quit ["#killall raidman"] 10:48 -!- shoan [n=shoan@61.95.206.107] has joined #agavi 11:30 -!- digitarald [n=Miranda@244.Red-80-59-74.staticIP.rima-tde.net] has joined #agavi 11:30 < digitarald> morning 11:32 < eremit> heyja digitarald 11:33 < v-dogg> mornink 11:37 -!- kaos [n=kaos@ppp-82-135-69-182.dynamic.mnet-online.de] has joined #agavi 11:38 < digitarald> btw ... shiny.php uses still core.webapp_dir configuration enrtries 11:42 < CIA-6> dominik * r895 /trunk/src/exception/templates/shiny.php: 11:42 < kaos> thx 11:42 < CIA-6> webapp_dir => app_dir 11:42 < CIA-6> refs #298 11:44 < v-dogg> I was just about to... 11:44 < kaos> hihi 11:45 < kaos> btw, i am playing with the init of webrequest (once again) 11:45 < kaos> your cgi config is really a problem 11:45 < v-dogg> webrouting, you mean? 11:45 < kaos> i can either fix it in a way that it works with rewrite off 11:45 < kaos> yeah 11:45 < kaos> ofc 11:45 < kaos> or that it works with rewrite on 11:46 < digitarald> the logging xml from the trunk ... is it valid?? 11:46 < kaos> i couldn't find a single way to differentiate between them tho :< (so the user would have to supply the info whether the url was rewritten or not) 11:47 < digitarald> ... The attribute 'priority' is not allowed. 11:47 < v-dogg> wtf... I changed that... 11:49 < digitarald> in the xsd oder the xml? 11:50 < v-dogg> xml 11:50 < v-dogg> just a sec 11:50 < digitarald> put the priority to the parameters? 11:52 < CIA-6> v-dogg * r896 /trunk/ (3 files in 3 dirs): priority => level, somehow missed these last time, sorry for the inconvenience 11:54 < v-dogg> kaos: I just wrote a custom WebRouting that has a different parsing methods for different server setups 11:54 < v-dogg> it tries to guess the setup or you can force it via config parameter 11:55 < kaos> cool 11:55 < kaos> does it fix the tests ? 11:55 < v-dogg> haven't tried yet 11:56 < v-dogg> but it fixes my acute need for a routing system for veikkomakinen.com :) 11:58 < kaos> thats great :) 11:59 < v-dogg> time to get lunch 11:59 < v-dogg> bbl 12:22 -!- Wombert [n=Wombert@dslb-088-064-004-049.pools.arcor-ip.net] has joined #agavi 12:42 < eremit> hey wombert 12:43 < Wombert> houmenta! 12:43 < Wombert> 'sup 12:43 < Wombert> hey v-dogg 12:51 < eremit> Wombert is there any way to convince you to allow forward() in actions again ? 12:51 < Wombert> of course! 12:51 < Wombert> if the arguments are reasonable ;) 12:52 < Wombert> look, guys 12:52 < eremit> lool ... is performance a reasonable argument ? 12:52 < Wombert> I realize that it's slightly less convenient than the old behavior 12:52 < Wombert> however, I guarantee you that you will, one day, be glad we made that decision 12:52 < Wombert> because actions are your application's most valuable part 12:53 < eremit> the convenience is not that problem ... but creating a view which basically only implements logic to decide whereto forward() is senseless performance burning ... 12:53 < Wombert> you can't afford brokedness in them 12:53 < Wombert> therefor, we have to move as much stuff out of them as possible 12:53 < eremit> maybe we can find a way, to enforce some kind of unlocking ... so a dev needs to be aware what he is doing ... 12:54 < Wombert> eremit: the performance drawback is pretty much negligible 12:54 < Wombert> and just forward()ing somewhere else may not be the only thing you want to do there 12:54 < eremit> maybe it is ... but why should i accept any kind of drawback ... 12:54 < eremit> in my case it is the only thing ... 12:55 < Wombert> disallowing forward() wasn't intentional by the way 12:55 < eremit> if ( !$user->isAuthenticated() ) forward() else forward() 12:55 < Wombert> it was just an unforseen side effect of this request parameter locking 12:55 < Wombert> and then it occured to me it's actually a pretty good idea 12:55 < Wombert> may I ask where you are forwarding to there? 12:56 < eremit> ok some basics about the app ... it consists of three modules ... common, guest, user ... common implements action accessible by everyone, guest only by guest 12:56 < eremit> and user only by user ... 12:56 < eremit> there should be an action Index ... which decided which action to use ... the guests or the users one ... 12:57 < eremit> maybe i missed s.t. but afaik there's currently no other possibility ... 12:58 < eremit> than puttin it into common ... 12:58 < eremit> except the routing allows logic like authenticed checks ... 12:58 < eremit> authenticated* 13:08 < kaos> you could do that 13:09 < kaos> by adding the auth info to the sources array 13:09 < eremit> pardon? 13:09 < Wombert> or 13:09 < Wombert> by using two routes 13:09 < Wombert> and a callback 13:09 < kaos> yeah, but i think the sources array is way nicer 13:09 < Wombert> that decides whether or not to execute 13:09 < Wombert> yep 13:09 < Wombert> but maybe not as simple for him to implement ;) 13:10 < eremit> using two routes would force me two have logic wherever i have a link to /index wouldn't it ? 13:11 < Wombert> uh 13:11 < Wombert> well 13:11 < Wombert> what you want is two entry pages 13:11 < Wombert> right? 13:11 < Wombert> one for the user, one for the admin 13:11 < eremit> one for the user and one for the guests ... 13:11 < Wombert> well whatever 13:11 < eremit> but basically yes ... 13:12 < Wombert> 13:12 < Wombert> 13:12 < Wombert> to the first, add the action name for the logged in user 13:12 < Wombert> to the second, add the regular index 13:13 < Wombert> or, as dominik suggested, you could add the auth info to a routing source 13:13 < Wombert> however, that would mean you have no more control over the incoming url 13:13 < Wombert> so I suggest you use the callback 13:13 < kaos> you do 13:13 < Wombert> well actually, you COULD 13:14 < Wombert> yeah, nested routes 13:14 < eremit> tyvm ... 13:14 < eremit> now even i dont need forwarding in action anymore :P 13:15 < kaos> 13:15 < kaos> 13:15 < kaos> 13:15 < kaos> 13:15 < kaos> 13:15 < kaos> 13:15 < v-dogg> BUT it still has to be discussed! 13:15 < v-dogg> :) 13:15 < Wombert> 13:15 < Wombert> 13:15 < Wombert> 13:15 < Wombert> 13:15 < Wombert> 13:15 * eremit turns logging on 13:15 < Wombert> oh wait the $ is wrong there 13:15 < Wombert> or is it? 13:15 < Wombert> no it ain't 13:15 < Wombert> would that work, dominik? 13:16 < kaos> yes 13:16 < kaos> at least it should 13:16 < Wombert> I'll try that later ;) 13:16 < eremit> v-dogg: i think what should have been discussed is not the forwarding itself ... 13:16 < Wombert> v-dogg has a point tho 13:16 < kaos> well, maybe we should add a strict mode :p 13:16 < eremit> whoever wrote this routing ... godly ... 13:16 < kaos> where all that stuff is enforced 13:16 < Wombert> eremit: it basically worked like this 13:17 < Wombert> I called dominik every five minuted 13:17 < Wombert> with a new idea 13:17 < Wombert> he said "uuuuuh that's difficult" 13:17 < Wombert> and then he wrote it 13:17 < Wombert> <3 kaos 13:17 < kaos> :> 13:17 < v-dogg> haha 13:17 < shoan> :) 13:17 < eremit> O_o ... 13:17 < kaos> well, i think the nested routes were my idea (actually first thought out mainly for performance ^^^) 13:18 < eremit> v-dogg: we should more discuss the idea of enforcing developers to think the way the framework devs do ^^ 13:18 < v-dogg> eremit: but if you think differently, you are most likely wrong :p 13:18 < eremit> the nested routes part was a part i havent understood in the source :P ... but it seems to give most flexibility as possible ... 13:18 < Wombert> the question is 13:19 < Wombert> does it make sense 13:19 < Wombert> to enforce certain types of behavior 13:19 < Wombert> I say: yes, absolutely 13:19 < eremit> v-dogg: not that different kind of thinking ... as i said yesterday ... preventing me from accessing my data wherever i want is like enforcing a special kind of version string :PP 13:19 < v-dogg> hihi :) 13:19 < Wombert> eremit: we aren't 13:19 < Wombert> you can access your data, and the very same 13:19 < Wombert> it's just the way you access it 13:19 < Wombert> that we enforce 13:20 < v-dogg> no analogy there ;) 13:20 < Wombert> and by doing that, we achieve something quite impressive IMO: we can guarantee that an application is safe if you have validation on 13:20 < Wombert> and that's pretty cool if you ask me 13:20 < Wombert> well not 100% safe 13:21 < Wombert> but it's kind of a tainted mode that's guaranteed to work 13:21 < Wombert> and the best thing: you can enable it, but no action code must be changed 13:21 < eremit> sure that's really cool, but still i dislike the "enforcing" idea ... without giving the developer the possibility to circumvent it... 13:21 < Wombert> why!? 13:21 < Wombert> it makes no difference if you use $request->getParameter(