*** reinhard has joined #gnuenterprise good morning all morning reinhard *** btami has joined #gnuenterprise *** kilo has joined #gnuenterprise good morning kilo: did I miss the hu.po commit mails? I didn't see anything, but an svn up gave me new versions of hu.po is hu translation ready? good morning translation is ready i havent received any commit mails recently for some days now http://www.gnuenterprise.org/cgi-bin/viewcvs.cgi/trunk/?root=gnue show the updates though s/show/shows btw please check german translations, last night after a clean svn checkout setup-cvs.py showed 6 errors for which package? dunno, was quite sleepy... ok, will have a look thanks all i remember is at the end of the setup process it showed the errors *** johannesV has joined #gnuenterprise good morning hi johannesV gauntlet... doesnt work for me Traceback (most recent call last): File "", line 11, in __main___ON_STARTUP_trgStartup File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/logic/NamespaceCore.py", line 324, in __call__ return self._objectFunction(*args, **kwargs) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/forms/GFForm.py", line 241, in getParameter return self._parameters [parameter] AttributeError: GFForm instance has no attribute '_parameters' *** lekma has joined #gnuenterprise morning everyone hi lekma do any of you have one of those new mac/intel laptops ? i'm considering bying 2 of those... kilo: can you please try again? hmmm.... commit messages in #gnue-commits don't work either? kilo, ah, i thought jamest has removed that bug (he introduced accidentally) reinhard: you mean the gauntlet? * reinhard introduced that bug, FWIW kilo: yes lol no, it was jamest (look at a svn blame :) btw, i've just added a default-collation function to our pysqlite2-driver (which is activated automatically if the connection object supports it) and yes, #gnue-commit seems to hang (as well as the commit-mails) nice, works fine. thx reinhard johannesV, reinhard: did you guys found the time to talk about null boolean? we didn't talk, but FWIW, I think at least for postgres NULL == False for booleans erm that is of course nonsense if I think about it twice, because postgres has no booleans well, you could issue a create table foobar (somebool boolean not null) .... pushing out the releases now reinhard, have you updated svn before ? yes ok as my last commit is about 10 minutes ago (and even if I hadn't, the release script would do so automatically) :-) reinhard erm that is of course nonsense if I think about it twice, because postgres has no booleans * lekma is confused there are no booleans in postgres? erm after svn up my appserver seems to not work... File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/appserver/language/Object.py", line 87, in __getattr__ [name]) [0][0] File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/rpc/drivers/xmlrpc/ClientAdapter.py", line 330, in __call__ return self._adapter._callMethod_ ('_call', *nargs, **params) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/rpc/drivers/xmlrpc/ClientAdapter.py", line 201, in _callMethod_ raise errors.RemoteErr where's the rest of that traceback ? lekma: you're right, there *is* a boolean in postgres kilo, what about your traceback ... ? what were you doin ? a session.find userId = session.find('AAA_User', {'name':userName, 'pass':password}, [],[])[0] (what a nice module name i have...) and what about the traceback ? this is the traceback, just no linebreaks somehow... File "Main.py", line 141, in evalLogin user = IF.evalLoginPage(userName=userName, password=password) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/appserver/language/Object.py", line 87, in __getattr__ [name]) [0][0] File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/rpc/drivers/xmlrpc/ClientAdapter.py", line 330, in __call__ return self._adapter._callMethod_ ('_call', *nargs, **params) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/rpc/drivers/xmlrpc/ClientAdapter.py", line 201, in _callMethod_ raise errors.RemoteError, (exType, exName, exMessage, exDetail) RemoteError: Procedure instance has no attribute 'repository' hah got them \o/ hm looks like it's one of mine, doesn't it ... damn, we really need unittests * johannesV starting database-server ... well the moment I see something like 'ClientAdapter' I immediately think of Austria... :) hm, strange, calling procedures works fine here ... ok, now i can reproduce it (although) rev 8247 worked ok erm, although it's a bit different yeah, it's all cause of that findItem stuff .... so must be something since then ah, it's a typeo i see it \o/ ok, now for another one ... :) ? since i've fixed that one, anotherone seems to appear ... reinhard, please wait for another 15 minutes with the release ... (maybe we could do it on afternoon, so kilo could do some more testing) kilo, please svn up and retry ok seems to work fine now thx for the fix at least it was caught before release :) so I should wait with release until afternoon? should I wait with forms release, too, or only appserver? well... green lamp from me doing forms release now CIA returned from coma... ok, if kilo is fine with appserver you can do that release too johannesV: i can't seem to find, in source, a reference to "isnull" condition you showed me yesterday, where is it? * lekma must be blind or is it ["null" ["Field", "bla_bla"]]? lekma, let me have a look yeah that could be :) so add another ["not", ["null" ... no there is a "notnull" condition "notnull" seem to be here as well :) yes, just look into gnue-common/src/datasources/GConditions.py ok, doing appserver release now, too yay!! \o/ hi reinhard. on the gnuenterprise website if I check for the versions of appserver i see the last 0.4.3 september 2005 reinhard: must be the site. sacha: no, that's correct I'm doing the release right now, so it can't show on the site already :-) so big release then yep 0.5 sweet i will be working for my own company again will see if I can use gnue again eventhough I must admit i did some ruby on rails work. maybe once I am closer to the Austria gnue powerhouse things will change again hi kilo hi sacha how are you? wow 2 sachas :) hi kilo :) your secret twin? :D if I need to blame someone, I can blame sacha if I need to blame someone, I can blame SachaS lol sacha: SachaS: hello guys derek needs a twin then too :) hi dimas_ ok, so we can now go for a branch of gnue-forms then ? yes *** Shufla has joined #gnuenterprise hello *** yure has joined #gnuenterprise hi Shufla reinhard: huh :) We've been trying to convince Fabien to share his code & API docs withour success :( *** dimas_ has quit IRC the max for a number(4,1) is 999,9? correct? yes thx for reassuring me, i got strange result from Math.pow(10, length - scale) in javascript... so there must be a bug somewhere *** btami has quit IRC wow I wasn't aware that we have a 82 page developer's guide for forms lol weren't you??? *** sacha_ has joined #gnuenterprise *** sacha has quit IRC *** kilo has quit IRC *** kilo has joined #gnuenterprise *** kilo has left #gnuenterprise *** sjc has joined #gnuenterprise *** jamest has joined #gnuenterprise reinhard: congrats on the release today *** jamest has left #gnuenterprise *** jamest has joined #gnuenterprise *** siesel has joined #gnuenterprise *** jcater has quit IRC ok, so about a year ago I talked about expanding navigator now I'd like to actually start working on it *** johannesV_ has joined #gnuenterprise my initial goal was a framework the other gui apps could build from and netbeans does this quite well *** jcater has joined #gnuenterprise for a quick look at what I'm trying http://www.javalobby.org/eps/netbeans_platform/ a flash presentation building custom apps using netbeans *** derek has quit IRC *** johannesV has quit IRC this has moved a bit from just navigator now though so I figure I'll spit out some stuff here and see if people scream *argh!!!!!* just a sec ok i use gclientapp a lot but most of the time I don't need all it has to offer the since we've stuck a lot of crap in the constructor it comes along for the ride my rough notes from friday not complete but to give a mind set Features apps need: Non-GUI configuration debugging profiling file storage authentication GUI menus toolbars status bars windows which I started to break down into App * Should require no arguments to it's constructor * No functionality beyond the registry * Configuration via code for simple apps * Configutation via an xml stream - so we can create a base GUI app and developers can remove things from it and add things to it. Like netbeans Registry * Manages the loading of components * Should also provide a service to reload components upon demand (optional auto look for updates?) Components * Components provide services to the application * They should not require args to their constructors, instead an api should exist to let them register services. * Constructors should be able to list dependencies of constructor types * Every component should have a primary interface defined via zope.interface I've called this thing in my notes gap (GNUe application platform) but the name is whatever so gap would end up being somehting like gap.foundation # The primary application gap.service.registry # Manages components (always loaded) then components like *** Shufla has quit IRC hmmmm gap.utility.logging # Loging support gap.utility.config # Configuration processing gap.utility.debug # Debug support gap.utility.profile # Profiler gap.utility.clp # Command line processor gap.utility.datasources # Datasources indeed this sounds interesting and useful as a development platform etc I think I understand what you mean so, and this isn't "real" code class GoatHerder(gnue.appcliation.foundation): addComponent('gap.utility.config') addComponent('gap.utility.datasources') etc or that could be populated by some type of configuration file linked to the app (maybe in a python egg) so that we could distribute a complete framework and then you could pull the parts you don't require i could also see the base UI framwork being part of gap so then navigator, forms, designer, appserver would all derive from gap and only load the components they need am I right in this doesn't require a complete rewrite of everything, as much as a reorganization? jcater: yes what exactly would be the advantage of "addComponent('gap.utility.datasources')" against "from gnue.common import datasources" ? just a sec reinhard: I was wondering the same thing glad you asked :) seriously I always thought we have one or two abstraction levels too much, not too little ;-) so I'm wondering where the benefits are ok several you'll note those are strings and honestly I see the addComponent() option as only for simple scripts the from foo import is hardcoded into your app the latter can be loaded from a config file at runtime *** derek has joined #gnuenterprise so you could replace components in the app without changing the app at all or lets say I want to add something to forms specific for ajrs i could extend my app with forms "add-ons" without changing the code components would also be registered as to loading at runtime from string, we do the same with our plugin loader yes please elaborate on "add something to forms specific for ajrs" what do you mean? a new control? oh, I think I see i'm trying to think of some concrete samples a new valid XML tag for the gfd? so addComponent('gap.utility.datasources') may or may not coincide with a gap.utility.datasources? it could be any class registering itself as providing that "service" jcater: right or am I off base? right then the other components of forms wouldn't be looking for say gDebug specifically they'd say "give me whatever provides gap.application.debug" and use that if it's loaded you'd no longer have hard coded dependency between the various sections of gnue actually, i was looking at zope3 interfaces and they are really close to this now when you add in their components system oooooo they give some examples too I think hang on http://svn.zope.org/Zope3/trunk/src/zope/ *** dimas has joined #gnuenterprise everything in the app subdir there is zope3 specific, the other dirs have stuff that you can slap into non zope apps http://svn.zope.org/Zope3/trunk/src/zope/interface/README.txt?rev=65911&view=markup and http://svn.zope.org/Zope3/trunk/src/zope/component/socketexample.txt?rev=39510&view=markup so where we'd end up is a system made of components with a well defined API via an interface loaded components would register with some type of manager components could ask for the component that provided the "Foo Service" then use that service without hard coded links between components and with enough thought into the UI part we'd end up close to what that netbeans demo does for those without flash create new netbeans project module run it and by default it *is* the netbeans editor open the loaded components dialog and uncheck all the crap you don't want add a few new components and you have a new app using most the underlying framework of netbeans then also demo adding a google style search bar to the existing applications toolbar by creating a new component i know there is a lot to digest in those links above I'm still trying to understand how we would use this in gnue-forms I can't imagine a situation where I would want to, say, replace the GDebug system with something different (not to think about the possible performance implications when replacing GDebug with something highly dynamical) I could really imagine something like this useful for adding new GF* elements to forms ok like we have no tree view, so I can write my own and plug it in actually, that is one of my questions... class MyFoo(): addComponent('gap.utility.datasources') but I think that could be done with python's abilities to dynamically load modules how does MyFoo then reference whatever gap.utilitiy.datasources provides? jcater: def initOne(): # I need the connection mgr conMgr = getComponent('gap.utility.datasource').getConnectionMgr() ah so the object would request links to instances that already exist via a global pool of components then if we want to get fance fany fancy refreshComponent('gap.foo.bar') your first example made it look like class foo: addComponent('foo.bar') somehow had addComponent creating a dynamic class i'm making up examples as I type them :) reinhard: so foo was somehow being extended (like with python's support for property() things ok, before I try to example forms lets pull back gap eventually forms, designer, appserver, navigator all use this system so instead of saying "what can I do in forms" we pull back and say, what can I do in the app here I have lots of GClientApp stuff lots but in app1 I don't use logging, or debugging, or command line options, or configuration options but I have to load them as GCLientApp does that then in the game i was working on I didn't need datasources but it has to load them now lets say I really, really like forms but I want all logins to any form monitored in real time i would create that component, add it to forms configuration, and as part of it's startup it would ask for a handle to the whatever datasouce mgr was loaded and register with it to get certain events, i would change a bit of the forms code but now all login username/passwords are emailed to bigbrother@mycompany.com *** siesel has quit IRC or lets say I want real time monitoring of calls in my call queue here and I want that in the toolbar of every form a user has open I'd create a component that displays the # of calls waiting on a button that launches our call center monitoring app modify my copy of forms configuration file to include that button on the toolbar, and away I go currently everyone in the call center here opens the full monitoring app nagivator would really grow just to be a enhanced component manager of sorts and eventually could load the "forms" component and display it's output in a general window handle provided by the gap generic ui system or it could load jcatersWickedModule2k component that monitors the current inventory levels of the snack machine in the breakroom in real-time *** btami has joined #gnuenterprise reports could be embedded in forms more easily I guess all of this could be done without the component system jamest is describing, going on like we're doing now but I do see true value in this direction fwiw i spent some time digging after I thought about going the netbeans route and zope3, twisted, and peak all do something like it, most based on zope3 interfaces well z3 and twisted use z3 interfaces and components (twisted enhanced it) peak i think is compatible at some level i can see the compents at the app level being useful and interfaces useful anywhere in gnue well I won't stop you anyway but I'm slightly worried about the overhead this adds to gnue reinhard: where do you see it added? i see startup time possibly I see it in complexity and number of abstraction levels which probably means higher learning curve for people wanting to hack gnue more bugs less performance more dependencies and less time to code that stuff that we *actually* want to see in gnue-forms odd, i'd have thought the oposite actually, I got the opposite on most of those points as I said imho, one of the bad things about gnue's code base (which I claim a good bit of responsibility for) is the complexity and number of abstraction levels and the learning curve needed I won't stop you anyway probably a proof of concept code will help decide in which direction to go reinhard: i'm wanted to do proof of concept anyway, but wanted to know if that was ok with you before I started i see navigator as the place I could do this both johannes and me probably won't touch navigator in a year from now so there's no risk of interfering with each other i figured that was the safe place as long as you are ok with us working on forms i'm fine with it FWIW I've committed forms technote 00006.txt with a proposal for changes in gfd format regarding layout management and grids it's not much yet, but I'd like to get some early feedback jcater: did you generate the xml description in the forms developer's guide? yes or did you just copy & paste from the dtd file? *** sacha_ has left #gnuenterprise you directly did generate an openoffice file? there is a script somewhere in there that generates it cool comon/utils iirc ok nice must run for birthday party, will be back tomorrow cu all *** reinhard has quit IRC *** lekma has quit IRC another interesting read from this weekend http://twistedmatrix.com/projects/core/documentation/howto/components.html twisted is broked BAD like REALLY bad we just got contracted to fix a twisted implementation for a major online learning facility twisted is full of memory leaks in the worst kind of way :) just an fyi wha? isn't twisted now the default communications interface of zope3? twisted is used for a LOT of application development these days from apt-proxy to zope (and many others) * jcater wonders if you're replacing a really bad implementation of twisted moreso than an implementation of a bad twisted library that's what I was thinking myself so far in reading the zope3 docs i'm really impressed at what they've done all files pretty much require unit testing, interface.py files most dirs I've hit have a DEPENDENCIES.cfg that list all imports in that dir a README.txt explaining what it's for I can't imagine they'd ditch zserver for twisted without some solid testing jamest: as i see, this component/lego system is similar to Delpy component, and very useful for writing custom gnue based app (different than gnue packages, using standard gnue .gfd .grd) am i right ? yes well i don't know about the Delpy but for custom apps, and gnue apps themselves s/Delpy/Delphi if it proves itself I think we could migrate most the guts of our gui apps to this jamest/jcater : we used twisted at ldc as well it has major issues (again just FYI) it is good for some stuff, business application development isnt its strong suit per say are you sure you didn't just get your derek all over it also in fairness i always thought zope had several production quality issues zope2 or zope3? so zope using it isn't a vote of confidence in my book :) zope2 is a known mess but pretty stable on twisted i was not at ldc when they used it, so that didnt have my "evil touch" zope3 is a from the ground up rewrite using forced good coding practices i'm not saying that it's quality and the group we were asked to consult for that had twisted implemention we had nothing to do with coding at all they actually called us in for "postgres" performance problems but what i've been digging thru is nice we got in and looked around and they had like 8gig database on a monster of a machine and it was crashing because of memory consumption as we all know postgres on a SMP machine with 4gig of RAM and an 8gig data set doesnt die because of memory problems in research all memory consumption was coming from twisted :) again limited exposure.. twisted could be greatest thing since sliced bread i wish active record was written in python as i think active record would be a great thing for gnue i would really consider having especially those working on app server looking at activerecord as a basis for an ORM and doing a python implementation *** dimas has quit IRC *** johannesV_ has quit IRC derek: maybe I'm missing something but the active record tuts I'm looking at are huge tuts? tutorials? tutorials did you see objectWrapper in gnue-contrib? i havent seen it im just saying after using active record a lot its quite mature and does what i wanted app server to always do w/o much fuss it lets me treat sql tables like objects do you have gnue-contrib checked out? but not make them gross sql tables and provides functionality to use sql directly against them w/o much fuss here from gnue.common.contrib.objectWrapper.ObjectWrapper \ import ObjectWrapper, Field, \ _getDecimal, _setDecimal, \ _getDatetime, _setDatetime class Item(ObjectWrapper): _tableName = 'items' _primaryKeys = 'item_id' _properties = ( Field('item_id' , typecheck=int), Field('created' , typecheck=datetime), Field('sku' , typecheck=basestring), Field('make' , typecheck=basestring), Field('model' , typecheck=basestring), Field('description' , typecheck=basestring), Field('unit_price' , typecheck=Decimal), ) created = CommonProperty('_created', _getDatetime, _setDatetime) unit_price = CommonProperty('_unit_price', _getDecimal, _setDecimal) that is a completely usable object i dont have gnue checked out on this machine (mostly because i dont have any non OSX boxes i use regularly, other than servers) using the default gnue connection but do you define that somewhere? or is done for you? what? I wrote that class to access a table created by gsd then if I want to do a join _properties = (Field('invoice_id' , typecheck=int), Join('items', className=InvoiceItem, masterLink ='invoice_id', detailLink='invoice_id', ), then you have load(), post() but all that is in gsd? gfd? ... pure python but where do you define it that is the definition define what? that is the definition of a data aware object in python it's a regular python class that I can do what I want with but where does that code go in my app jamest@hal:~/svn/gnue-contrib/objectWrapper/tests/objects$ ls address.py baseAddress.py __init__.py invoiceAddress.py invoiceItem.py invoice.py item.py state.py currently in active record i can define a model (which is similar to say a gsd) and then that object is available anywhere i instantiate it i have a dir here at work that contains all those models i can have inheritance on tables as well as polymorphic associations inheritance isn't an issue validations can also exist directly in a model though I'd have to se a polymorphic association sample to grok it validations are in that model above created = CommonProperty('_created', _getDatetime, _setDatetime) so instance.created('1/1/2006') converts automatically the typecheck arg is used to verify the end result is what you expect it's not as powerful as appserver http://wiki.rubyonrails.org/rails/pages/HowToUsePolymorphicAssociations but it's a pretty easy to use method of mapping objects to databases is a sample of polymorphic and it'd be fairly simple to build these from the gsd directly looks like foreign keys jamest... which you do jcater: poly is much more than foreign keys foreign keys is : belongs_to, has_many and has_to_belongs_to_many (hatbm) STI is way to inherit from a base an object using the type colum so you could have Person table with a type column and then have Employee << Person Contractor << Person etc poly is much more involved its like a has to and belongs to many on steriods gack What about foreign keys? Because the _id field can reference records in different tables you can never use a foreign key in the database. So you can create a less ?stable? database environment. Answer If you need to guarantee referential integrity you can always use a trigger (if your database supports it). This is obviously non-portable though. jamest? we use FK in database I'll admit I'm an old dog http://wiki.rubyonrails.org/rails/pages/UnderstandingPolymorphicAssociations (see the questions at the bottom) But if they want me to ditch referential integrity I'm not sure this is a trick i'd learn it simplifies code enough it is worth it for reusability.. for us its non issue is we are not concerned with portability of db so we work around with triggers so like gnue appserver, it is only useful for new projects jcater: for a large part yes i think thats somewhat par for the course if a legacy application had structures that dont take advantage of a new framework...its hard for a framework to cope with that however, activerecord has some ways around it true you can "map" things of course, I'm of the old school mentality that a database schema should be designed to outlive the shiny new framework but if they dont exist you can't do it so say STI relies on a column called type if you dont have something to represent that type there isnt much you can do if you have something that represents type but is called mud you could make it use mud * derek is not stating that active record is the end all i just think that it does a LOT of what the goals of appserver were (once upon a time) and does so fairly well the things i dont like about it are the same things i didnt like about appserver being too object minded :) * derek really enjoys rails as a framework, but still think that something like a gfd... has serious merit * derek debates daily about trying to write something to allow gfds to be used as rails applications i used to loathe web applications because they were so limiting, but with all the new javascript and AJAX stuff available its not as true anymore one of our major customers did an IE upgrade nationwide before i would say 5% of applications were suited for web and 95% not it happened to break every usage of pre-AJAX-hype custom javascript I have now i would say its more 40% to 60% seeing as how it works fine in all versions of IE i have access to and how their corp tech support is baffled IE is satan plain and simple my enthusiasm for ajax based apps has been somewhat tempered as I have no way to debug this issue (which is the same hell java based web apps gave me when I was in math) i'm hoping a move to prototype for the com channel will magically make things work again as it's wider spread but I'm still of the mindset that rich client apps are here to stay for the forseeable future with rails we have scriptaculous and protoype all built in so we write very little javascript its all rails method calls and all the xplatform stuff is just magically handled again not end all of end alls we try to stay away from JavaScript/AJAX except where it really brings value as supporting it is suck *** klasstek has joined #gnuenterprise *** btami has quit IRC *** psu has joined #gnuenterprise hey psu did I see you restart the kernel cousins jamest: he did!!! and you guys are giving him tons of new content to put in there *** psu has quit IRC *** psu has joined #gnuenterprise back as is GNUe Traffic - #108 now on line cool where? ah if I checked gnue-announce, I'd know http://www.kernel-traffic.org/GNUe/gnue20060320_108.html or, if you prefer original XML goodness http://www.kernel-traffic.org/GNUe/gnue20060320_108.xml * derek wants to see the firefox plugin :0 * psu wonders if he should use GNUe Designer to do GNUe Traffic as "all" Designer is, is a big XML editor of course, by same logic, "all" that emacs is, is a text editor s/emacs/vi or whatever designer is not currently an xml editor it is currently an elegant way to generate a python exception well, an xml generator anyway (traceback) elegant? don't you mean convoluted? yeah that's what I meant if that's an RSS feed than you can already monitor it with firefox :) psu: where's the RSS feed eh? www.kerneltraffic.org/GNUe/rss.rdf but not working for me at the moment Next question? why does RoR and AcvtiveRecord kick so much ass? :) It just does, OK? Actually, I thought that derek was the Ruby on Rails fanboy around here me to! me too! me tooo! me too :) it's all about the ration of l33t to suck it doesn't "kick so much ass" rails is new and full of l33t GNUe and the python upon which it is based is old and full of suck * derek thinks my comment of "look at AR" to get ideas for appserver ORM turned into "use rails" which was not the intent of the comment at all though reinhard and johannes have tried really, really hard jcater and myself once put code into GNUe and that kind of taint runs deep gnue is doing so much more than something like rails it would be impossible to compare the two To me, there have always been wonder Rapid Deployment Framworks but I just did compare them jamest: sadly that is how a lot of people think new buzzword == better product that are, as well as flavour of month, actually good at some things the thing I like most is the integrated unit and functional testing * psu seems to remember a regular feature of old GNUe Traffics i love unit testing chillywilly: java has a lot of that built in where people would come on line and say "why not use FOO?" wouldnt make me use java :) I know it's nothing new (where FOO = current flavour of month) derek's std reply was i dislike java, but netbeans destroys any editor I've used hands down psu: i think its always important to research whats out there, but that doesnt mean ditch what you are doing and i adopt it "FOO is generic framework, GNUe is explicitly a toolkit for database-driven biz apps" jamest: textmate is good, but you need mac to use it :( derek: agree textmate sucks poo can it let me right click on a say a class var and say go to def er click on a class var, go to source psu: my other famous comment is ... framework X is all the rage right now, but in 2 years gnue will be here and foox.com will be a non resolving url :) then right click, refactor->rename change the name derek: yes james it can pretty much do anything you want it to do and it fixes all references to that var or class name in *all* loaded projects * derek notes that it is a great editor for ruby (rails) not sure what its support for python or such is like oh, wait I tried textwranger, not textmate textmate is non-free i'd kill (or blame derek) for an editor like that for python which is sadly the typical case on mac yes textmate is non free (which is evil) * jcater is learning plus the matise gui builder in netbeans is the shit but it does its job amazingly well.. the guys we have still working with a lot of python like it for python as well jamest: i have heard good things about netbeans it makes java worth coding but visual studio is the bomb to many people as well :) is it time to remind jcater of his pre-GNUe forms client written in java? in fact all C# developers i encounter actually dislike C# the language BUT all love visual studio Or doesn't he deserve a painful flashback? and basically say they like C# because of visual studio not because of c# * derek personally didnt like visual studio too much i did very much like delphi's IDE but then again i like gnue designer so my sense of development tool is skewed ;) * psu though you liked designer *because* it reminded you of delphi prolly mostly i liked it because it made me more productive emacs is all you need in 13 oops textmate actually has eliminated my emacs addiction that is impossible http://wwwf.centos.org/127_story.html?storyid=127 a friend sent me that but i notice it's on /. too is that the moron from Tuttle, OK? yeah yep, it is totally forget CentOS imagine the poor chumps who are the Tuttle, OK IT staff how bad does it have to suck to report to that fellow? I think the CentOS guy really needs to invoice them for his consulting services good thing tehy werent using debian or there would be no one to blame for installing that damn spyware! and that kind of issue would only crop up every 3 years between new releases roflmao i thought it was every 5 years if they are going to speed up that release cycle im switching back to windows where im guaranteed a decade between releases :) * chillywilly is on the Ubuntu bandwagon now * derek slowly using more and more BSD (not even of the OSX variety) not really by choice though we were the first ubuntu affiliate though :) *** psu has quit IRC *** kilo has joined #gnuenterprise *** krizz has joined #gnuenterprise *** yure has quit IRC *** sjc has quit IRC *** klasstek has quit IRC *** jamest has left #gnuenterprise *** kilo has quit IRC *** jamest has joined #gnuenterprise *** jcater has quit IRC *** derek has quit IRC *** jcater has joined #gnuenterprise *** jamest has quit IRC *** jamest has joined #gnuenterprise *** derek has joined #gnuenterprise *** derek has quit IRC *** derek has joined #gnuenterprise *** jamest has quit IRC *** jcater has left #gnuenterprise *** sacha has joined #gnuenterprise *** krizz has joined #gnuenterprise *** krizz has quit IRC