*** johannesV has joined #gnuenterprise
*** reinhard has joined #gnuenterprise
*** kilo has joined #gnuenterprise
<kilo> good morning
<dimas> hello
<johannesV> good morning
<reinhard> good morning
*** lxf has quit IRC
*** holycow has quit IRC
*** btami has quit IRC
*** btami has joined #gnuenterprise
*** holycow has joined #gnuenterprise
*** ajmitch has quit IRC
*** ajmitch has joined #gnuenterprise
*** kilo_ has joined #gnuenterprise
*** kilo_ has quit IRC
*** holycow has quit IRC
*** SachaS has joined #gnuenterprise
*** mnemoc_ has joined #gnuenterprise
*** mnemoc has quit IRC
*** btami has quit IRC
*** kilo has quit IRC
*** ajmitch has quit IRC
*** ajmitch has joined #gnuenterprise
*** ajmitch has quit IRC
*** ajmitch_ has joined #gnuenterprise
*** elwis has joined #gnuenterprise
*** elwis has left #gnuenterprise
*** lekma has joined #gnuenterprise
<lekma> hola los amigos
<lekma> :)
<wt> hola
*** ajmitch_ has quit IRC
*** jamest has joined #gnuenterprise
*** johannesV_ has joined #gnuenterprise
*** johannesV has quit IRC
*** wendall911 has joined #gnuenterprise
*** elwis has joined #gnuenterprise
*** jcater has joined #gnuenterprise
*** lekma has quit IRC
*** jcater has quit IRC
*** jcater has joined #gnuenterprise
*** ajmitch has joined #gnuenterprise
*** wt has quit IRC
<elwis> DB000: Traceback (most recent call last):
<elwis> DB000:    File "/usr/lib/gnue/python/gnue/designer/Designer.py", line 196, in RunWizard
<elwis> DB000:     style = templateInformation['ProductStyle']
<elwis> DB000:  KeyError: 'ProductStyle'
<elwis> DB000
<elwis> Bad Bad Designer :(
<jcater> what wizard are you running?
<elwis> errhhmm.. I run the wizard to create a nice multiple block form
<elwis> I'll start up gEdit instead :)
<elwis> Have I missed something here cause I get a huge error message and can't read half of it
<elwis>   <datasource name="dtsARTKAT" connection="pycash" requery="True"
<elwis>               table="artikelkategori"/>
<elwis>   <datasource name="dtsARTIKEL" connection="pycash"
<elwis>               detaillink="artikelkat" master="dtsARTKAT" masterlink="id" requery="True"
<elwis>               table="artikel"/>
<elwis> Suppose I could have been sloppy creating my tables :(
<elwis> no worries, it worked when the designer decided to play cool with me :)
<elwis> Nopes!
<elwis> What does "ValueError: invalid literal for int():" mean?? I haven't done anything yet :(
*** elwis has left #gnuenterprise
*** elwis has joined #gnuenterprise
<elwis> Is it only kilo that administers the wiki? I'm still unable to upload files?
<jamest> elwis: i don't know anything about the wiki
<elwis> ok, thought I should put up my manuals but another day .. :)
*** btami has joined #gnuenterprise
<jamest> elwis: what's it doing to you?
<elwis> sorry.. for being late,., trying to understan dthe lousty MySQL before I finally leave it for good..
<elwis> It says During save of file Main.UserMANUALs an error was found by the version control system. Please notify your TWiki administrator. /usr/bin/ci -q -l -m'Gnue intro' -t-none -w'StefanElwesthal' /var/www/twiki/pub/Main/UserMANUALs/GNUe_for_the_helpless.pdf 2>&1
<elwis> ci: /var/www/twiki/pub/Main/UserMANUALs/GNUe_for_the_helpless.pdf,v: Permission denied
<elwis> Go back in your browser and save your changes locally.
<jamest> elwis: just looking around
<jamest> can any edits to the wiki be saved?
<jcater> hmm
<jamest> the perms look wrong at the filesystem level ( to these non twiki eyes)
<jcater> hmm
<jcater> I think we may have scripts that chown -R /var/www
<jcater> maybe that needs to be tweaked
<jcater> scripts = cron scripts
<jamest> jcater: something else that's odd
<jamest> the epydocs don't regen via cron
<jamest> but work from web user command line
<jcater> but I swore twiki files were somewhere in /var/lib
<jcater> hmm
<jamest> look to be in var/www
*** kilo has joined #gnuenterprise
<SachaS> hi kilo
<SachaS> i revived my gnue todo application. and it is working.
<kilo> great news!
<kilo> was it great work to revive it?
<SachaS> no it was not! i was surprised how easy it was.
<SachaS> took me maybe 3 hours
<kilo> great
<SachaS> the form is still not very good but useable.
<kilo> what was it like to gnue again?
* kilo invents new verb: to gnue
<SachaS> i do not know
<SachaS> focus on simple things first ... gnue probably will do it
<dimas> kilo: :)
*** elwis has left #gnuenterprise
<kilo> like 'to google' is considered valid, 'to gnue' should be also
<dimas> kilo: see, elwis have gnued some doc but wiki reject it :)
<kilo> ah, and he has just eft
<kilo> left
<SachaS> i am not sure how gnue will behave for complex projects. eg will there be a boolean history attribute for a class? so appserver deals with history out of the box.
<SachaS> so a question will be: is the .gcd expressive enough?
<SachaS> kilo: i had a look at this: http://shark.objectweb.org/
*** sjc has joined #gnuenterprise
<SachaS> i am intersted how to setup a workflow engine or business process engine with gnue appserver
<SachaS> and then still the old question of gnue-appserver 2 gnue-appserver or gnue-appserver 2 other-system
<SachaS> but it feels good to gnue again (even though it might one or two days only)
<kilo> latter question is no problem imho through some xml interface
<SachaS> sure :)
<kilo> and in fact separate gnue apps couls interact through tehm also
<kilo> but now my main goal is to be able to lift one of my arms... after training you know
<SachaS> and maybe another thought: maybe one has to dig into the code to get the best out of it
<kilo> reaching out for the mouse is quite a task...
<SachaS> guess it will feel easier for your drink.
<kilo> in fact main problem is that noone seems to be able or committed enough to really dig into some packages
<SachaS> why do think this is the case?
<kilo> give me one name who came here for this purpose and stayed for more then 2 days...
<SachaS> kilo ???
<SachaS> :)
* reinhard !
<SachaS> reinhard
<kilo> reinhard: you came for this purpose? never knew that 8-)))
<jcater> I came for the beer and hot women
<jcater> then jamest and derek wouldn't let me leave
<jamest> lol
<kilo> rotfl jcater
<jcater> something got lost in the translation though
<jcater> I think they were trying to say something about losing hair and lots of squeeming
<reinhard> kilo: just searched the mailing list archives, but they don't go back that far
<reinhard> *actually* I came here to localize GNUe packages for Austria
<reinhard> I didn't know that there are no packages
<reinhard> and at the time when I came there weren't even tools :)
<kilo> not even a screwdriver?
<reinhard> :)
*** sjc has quit IRC
<kilo> ecco e il grande cuppole della duomo di milano
<btami> night
*** btami has quit IRC
*** SachaS has quit IRC
<jamest> reinhard: i think the move to python 2.3 is a go.  what do you think? :)
<reinhard> I agree
*** johannesV_ has quit IRC
<reinhard> jamest: done :)
<jamest> lol
<jamest> already optimized for 2.3 now eh?
<jamest> you're quick
<jamest> does that mean we can drop the 2.2 locale(?) hacks
<reinhard> jamest: file, line number?
<jamest> damn man, that was last week
<jamest> hmmmmm
<jamest> end of Designer.py
<jamest> in gnue-designer/src
<reinhard> huh
<reinhard> never saw that at all
<reinhard> and no idea what it means
<reinhard> but I agree with
<reinhard> # TODO: I don't think this is working!
<jcater> Python 2.2 had a bug in StringIO.StringIO
<jcater> that caused it to break with unicode
<jcater> so you had to use cStringIO instead
<jcater> which doesn't support unicode at all
<jcater> but wouldn't crash
<jcater> might've been fixed in 2.2.1 though
<reinhard> but that whole file doesn't *use* StringIO at all?
<jcater> well, it did at one point :)
<reinhard> :)
<jcater> I guess the hack at the bottom was missed
<jcater> when stringio was removed
<reinhard> well, should be safe to remove it then anyway :)
<jamest> lol
<jamest> reinhard: about the accessors
<jamest> are you talking wrapping all vars in accessors?
<reinhard> no
<reinhard> of course not
<reinhard> but it will make sense quite some times
* jamest puts down the bat he reserves for java coders
<jamest> then i'm ok with it
* jamest was worried about function call overhead
<reinhard> jamest: be assured johannes will beat me if I kill his performance gains ;-)
<reinhard> wow
<reinhard> generator functions are *way* cool
<jcater> hmm
<jcater> so basically they are just a simplification of __getattr__ and __setattr__ ?
<jcater> and we'd use generators when we, in the past, would've used __getattr__ and __setattr__ ?
<reinhard> jcater: you mean properties?
<reinhard> (generators are a different new feature available starting with 2.2)
<jamest> properties let you replace
<jamest> direct var access w/ a function w/o breaking the code
<chillywilly> I love python
<jamest> generators IIRC generate sequences
<jamest> reinhard: is that correct?
<kilo> yes
<jcater> ok, then properties
<jamest> reinhard: thanks
<reinhard> generators generate objects that behave like sequences
<jamest> :)
<jcater> iirc, we use generators in the ResultSet code
<jamest> only if runnin on 2.2 thouhg
<jcater> ok, then what I said about properties :)
<jcater> jamest: yeah
<reinhard> jcater: we use *iterators* in ResultSet code
<jcater> ok
<jcater> well, regardless
<reinhard> generators are kinda special fom of iterators
<jcater> s/generators/properties/g
<jcater> then reread my stuff
<reinhard> properties let you more or less define a special __getattr__ and __setattr__ that only gets used for a single name
<jcater> right
<jcater> and we wouldn't use them for simple stuff, right?
<reinhard> (that's how I would see the relation between properties and old __getattr__/__setattr__)
<jamest> i feel they are more for something like
<jamest> foo.name = 'Bar'
<jamest> now we suddenly want to do a .upper() every time a name is set
<jamest> we can use properties so that a function myFunction() is called whenever our code does a foo.name = 'Bar'
<jamest> to set the now hidden name
<reinhard> 2 examples where I would use a property instead of what we have now:
<reinhard> (provided it doesn't hit performance too bad)
<reinhard> for XML parsed trees, every object has a _parent
<reinhard> that causes circular references throughout GNUe causing memory leaks wherever you look
<reinhard> best solution would be to make _parent a weak reference but that means it can't be used like it was before
<reinhard> so I'd make a __parent be the weak reference and _parent being a property setting and getting the weak reference properly
<reinhard> and all existing code accessing _parent will still work
<reinhard> second example is ResultSet.current which should actually always be self._cachedRecords [self._currentRecord]
<jcater> ok, those make sense
<jcater> just as long as we don't go the java route :)
<reinhard> now as current is a normal attribute, all code has to take care that current is kept in sync with _currentRecord
<reinhard> and if it gets out of sync the ResultSet starts to behave *very* weird :)
<jcater> hmm
<jcater> I wonder if that's my oracle issue
<reinhard> no
<jcater> I bet that might silently fix my oracle problem
<reinhard> I don't think so
<reinhard> I am 99% positive there is *no* way you can get curent out of sync with _currentRecord
<jcater> hmm
<jcater> maybe so
<reinhard> but I had some cases temporarily while working on ResultSet
<jcater> that would just coincide with what I see
<jcater> but I haven't found the actual bug
<jcater> so can't know either way
<reinhard> well however
<reinhard> it shows that it could be a good approach to make the code more stable and secure against bugs
<jcater> yip
<jcater> l8r
*** jcater has quit IRC
<reinhard> actually I think it could even be great for debugging
<reinhard> e.g. if object foo has a variable bar and bar gets set to None at some point
<reinhard> and you don't find who sets it to None
<reinhard> (you know that problem?)
<jamest> what? no. never.
<jamest> look! something shiny!
<reinhard> just temporarly make bar a property and add some debugging output in the set method
<jamest> in any case, i'm still OK with it :)
* reinhard is so excited about the new possibilities :)
<jamest> now I have to go and run a jackhammer till it gets too dark to work
<jamest> home ownership rocks
<reinhard> brrr
<jamest> l8r
<reinhard> time to go to bed for me
<reinhard> l8r all
<reinhard> night all
*** jamest has left #gnuenterprise
*** reinhard has quit IRC
*** wendall_away has quit IRC
*** kilo has quit IRC
*** ajmitch__ has joined #gnuenterprise
*** tiredbones has quit IRC
*** ajmitch has quit IRC
*** jcater has joined #gnuenterprise
*** havoc has quit IRC
*** havoc has joined #gnuenterprise
*** havoc has joined #gnuenterprise
*** havoc has quit IRC
*** havoc has joined #gnuenterprise
*** dcmwai has joined #gnuenterprise
*** jamest has joined #gnuenterprise
*** jcater has quit IRC
*** johannesV has joined #gnuenterprise