*** johannesV has joined #gnuenterprise
*** sjc has joined #gnuenterprise
*** reinhard has joined #gnuenterprise
*** btami has joined #gnuenterprise
*** kilo has joined #gnuenterprise
*** dcmwai has quit IRC
<kilo> good morning
*** mo_ has joined #gnuenterprise
<mo_> hi there
<reinhard> hi here
<reinhard> :)
<mo_> :)
<mo_> i'm a totally newbie to gnue (i try to help someone to finalize his erp), and i don't know how to use gnue-reports-server, someone could help me pls ?
<reinhard> I am not sure if gnue-reports-server is already functional
<reinhard> IIRC you can use repots only standalone (not as server)
<reinhard> by now
<mo_> ah ok, that's why i couldn't have any results ://
*** sjc has quit IRC
*** holycow has joined #gnuenterprise
<mo_> has someone experienced a memory leak ? it seams that python doesn't free memory (list of objects), i haven't test it till 100% mem, 100% of swap
<holycow> *meh*
<holycow> i'm not liking rekall very much at all
<holycow> and chillywilly is right, kde is just bloody annoying
<holycow> not me mo_ not yet anyway
<holycow> infact, rekall sucks pretty much
<reinhard> mo_: what tool do you suspect leaking?
<mo_> python :)
<reinhard> I use appserver and have it running for weeks without restarting
<reinhard> oh
<mo_> it seems that it will be corrected with 2.5 :/
<reinhard> huh?
<reinhard> so you use python 2.4?
<mo_> é.3
<mo_> 2.3
<reinhard> ok
<reinhard> we all use 2.3 here, too
<reinhard> but i can't remember anybody suspecting a memory leak
<mo_> http://evanjones.ca/python-memory.html
<kilo> hmmm
<kilo> i would not use python to deal with graphs including millions of branches... but it could be an issue, yes, in special cases
<mo_> ok, i'll make some research, but this url was just an example to indicate that there could be a leak ...
<mo_> how many objects do u use in your app ?
<reinhard> mo_: hoenstly, i wouldn't call that a memory leak
<reinhard> as the memory is reused
<reinhard> i mean
<reinhard> when he runs the graph algorithm several times in a loop
<reinhard> python wouldn't suck more memory each time
<reinhard> but it would reuse the same gigabyte all over again
<mo_> i hope that this memory can be reused, in the app in which i work i've got this pb, python use more and more memory after each call
<mo_> and perhaps my friend sucks in python programming too :)
<kilo> i agree with reinhard, it is not a memory leak. as if you exit python, it surely hands memory back to the os. it is a memory management problem only
<mo_> ok, but finally a problem ...
<kilo> @seen lekma
<bigbrother> kilo: lekma was last seen here 4 weeks, 5 days, 1 hour, 36 minutes, and 39 seconds ago saying: <lekma> will one of them call be the one to provide it
<kilo> @weather LHUD
<bigbrother> kilo: The current temperature in Szeged, Hungary is 36°F. Conditions: Partly Cloudy. Humidity: 95%. Dew Point: 35°F. Pressure: 29.93 in (1013 hPa). Visibility: 5.0 miles (8.0 kilometers). UV: 0 out of 12
<kilo> @weather EGPH
<bigbrother> kilo: The current temperature in Edinburgh, United Kingdom is 46°F (10:20 AM GMT on November 10, 2004). Conditions: Partly Cloudy. Humidity: 62%. Dew Point: 34°F. Pressure: 30.15 in (1021 hPa). Visibility: 7.0 miles (10.0 kilometers). UV: 1 out of 12
<kilo> @weather EGPI
<kilo> @weather EGPB
<bigbrother> kilo: The current temperature in Sumburgh, United Kingdom is 46°F (10:20 AM GMT on November 10, 2004). Conditions: Light Rain. Humidity: 93%. Dew Point: 45°F. Pressure: 29.98 in (1015 hPa). Visibility: 4.4 miles (7.0 kilometers). UV: 0 out of 12
<holycow> @seen seisel
<bigbrother> holycow: I have not seen seisel.
<holycow> is that the proper spelling of his name? anyone have his email address?
<kilo> @seen siesel
<bigbrother> kilo: siesel was last seen here 5 weeks, 2 days, 11 hours, 58 minutes, and 2 seconds ago saying: <siesel> night
<holycow> ehe :)
<holycow> danke
<kilo> holycow: http://gnuenterprise.org/community/contrib.php, search for Jan
<holycow> gnue search engine is having a few issues :/
<holycow> ah!
<holycow> danke :)
<kilo> holycow: check out gnue-reports/samples/php_client
<kilo> mo_: afaik gnue report server IS working
<holycow> oh neat, k
<kilo> johannesV: i did not follow, but what is the news about modal forms?
<mo_> kilo : oki cool :) is there some docs on how to use it ?
<kilo> not much, look athe samples. php_client is a good example
* holycow apt-..removes rekal
<holycow> *bleh*
<holycow> :)
<kilo> i love apt-get remove...
<mo_> rm -rf / too :)
<holycow> lol :)
<btami> mo_: http://www.gnuenterprise.org/~btami/gnue-reports-server-test.py
<holycow> rekall is decent, but it's got the faint whiff of commercial development and ms
<kilo> the other day i issued apt-get remove glin* and it wanted to remove almost the whole system...
<holycow> the 'just get something out the door as fast as possible' sorta deal
<btami> mo_: i don't know how outdated that samole...
<btami> it worked, when i wrote it
<mo_> btami : oki thx, i'll test it
<mo_> brb
*** mo_ has quit IRC
*** mo has joined #gnuenterprise
<holycow> hey guys.... with respect to gnue designer
<holycow> would it be possible to make ALL popups and text boxes resizeable?]
<holycow> having a teeeeeny weeeeny little window to select stuff from is ... *ummm*  ... well reminds me of lotus notes :)
<holycow> i know, i know, submit patch
<holycow> :) ehe
<holycow> re: designer... when new data sources, why does it remember l/p?
<holycow> it's convenient from one perspective
<holycow> what if someone wanted to login to a new db with a different account on postgres for example?
<holycow> how would one know which is which in the pulldown?
<kilo> hmm
<johannesV> kilo: you would use a <from style="dialog"> to create a 'modal' form
<kilo> should we not differentiate between a dialog and a modal form?
<johannesV> kilo, you can't :)
<kilo> remember, we CAN do anything, we just need to define the proper environment for it 8-)))
<kilo> so there is no modal form... and is it the proposed way, using dialogs?
<johannesV> a modal form *is* a dialog
<johannesV> there is no mechanism from the ui-driver (gtk/wx) which defines a 'modal form' other than a dialog
<johannesV> one need to differe between a 'form' and a 'dialog'
<johannesV> where a form is just a 'window' (which might be transient for another one)
<johannesV> it get's displayed and the programm continues with it's main loop
<reinhard> in short: a dialog is a modal form
<reinhard> a modal form is a dialog
<reinhard> it's one and the same
<johannesV> a dialog is different, becaus after displaying it's window the programm does *NOT* get back to the main loop
<johannesV> but it wait's until the dialog-window get's closed
<johannesV> kilo, do you see the difference ?
<johannesV> modal like you means always 'dialog' because you want the user to finish the dialog first
<johannesV> and then continue with the main loop
<johannesV> a form beeing 'transient' for another one means: form a should always stay on top of another form b
<johannesV> (but the programm still is in the main loop [not responding to events from form b])
<kilo> imho a dialog is a subclassed modal form, at least iirc from C
<johannesV> well, please refer to the docs of wx and gtk2
<johannesV> btw. for which 'master form' should a 'modal main form' be transient then ?
<reinhard> kilo: that's another definition of "dialog"
<reinhard> we don't subclass forms in gnue-forms
<reinhard> so we don't have that meaning of "dialog"
<holycow> what would happen if i removed --purged postgresql... and reinstalled it... i would haveto run the gnue common setup-cvs.py again?
<reinhard> holycow: nope
<reinhard> just reload the database
<holycow> supposing i deleted the db too i mean
<johannesV> damn, sapdb seems not to be a dbsig2 driver ... :(
<reinhard> johannesV: sure?
<reinhard> johannesV: i was quite sure there is a dbsig2 module for max/sapdb
<kilo> there are remains of source code in forms that show once there was an intention to do modal/unmodal forms. it is only this that i ask this question, i don't want to say if anything is ok/wrong. i just asked with not any intention in my mind 8-)))
<btami> johannesV: it worked before
<johannesV> i've prepared the backend-server, created a new instance
<johannesV> and after importing sapdb
<johannesV> it does not contain all that stuff other dbsig2-modules do
<johannesV> (like TIMESTAMP, parmastyle, ...)
<johannesV> kilo, np, but that modal-stuff in forms-code does not do very much (cause of the underlying uidriver)
<johannesV> ah, there seems to be another python-module then ... (sapdbapi)
<johannesV> will look for that
<johannesV> ok, seems to be a DB API 2.0 conformant module to download from mysql.com
<johannesV> uh, no it's much easier then
<johannesV> it is packed into it's own sub-module of sapdb shipped as deb python-maxdb
<holycow> so in terms of best practices
<holycow> obviously the gnue db is to store all gnue related information
<holycow> i should be creating new db's in postgres and creating users for that db correct?
*** btami has quit IRC
<reinhard> holycow: for non-gnue stuff?
<reinhard> or for your gnue application?
<holycow> yep my gnue app ... i'm just trying to build a big picture understanding of the right way to do things
<holycow> let me rephrase my question
<holycow> what is the gnue db in postgres designed to do? it is specific to the gnue it self and 'how' it functions correct? i.e. one should not be using that at all?
*** kilo has quit IRC
<reinhard> the gnue db will hold both gnue internal data *and* your user data
<holycow> isn't the mixing of data dangerous?
<holycow> sorry
<holycow> mixing of purpose i mean ?
<reinhard> i don't think so
<holycow> okay
<reinhard> i think the overhead for creating several db's would be much more trouble
<reinhard> and we have the gnue db self contained
<holycow> sorry if the questions sound dumb
<reinhard> you only have to backup, archive etc. a single db
<holycow> i admin a lotus domino db... and it has a hierarchical db
<reinhard> no, in fact i had to think a little before answering
<reinhard> because i've never considered this an issue at all
<holycow> i.e. the db and the form are actually the 'same', so you end up creating dbs with individual acls attached to them
<reinhard> and that's *really* dangerous
<reinhard> :)
<holycow> ehe :)
<holycow> well as long as you don't mind me asking... i have a whole bunch of questions, i just want to ask the important ones first
<reinhard> that's ok
<reinhard> i will be choppy in my answers as always
<holycow> my approach here is from an admin/user persepctive
<reinhard> as i'm at work and the phone rings all the time
<holycow> ideally, i want to merge the benefits of the domino system, with gnue which is wonderfull, and rdmbs
<holycow> so my questions are all skewed toward a 'central admin' with tons of 'users' to deliver applications to, prefereably centrally
<holycow> lol
<holycow> you do gnue dev fulltime or part time, if you don't mind me asking?
<reinhard> not full time in any case
<reinhard> i'm not sure about part time or hobby
<reinhard> as i'm self-employed and the borderline is quite thin ;-)
<holycow> ah cool :) *nod*
<holycow> oh totally, same here
<holycow> sounds like we are both looking at gnue from the same perspective then
*** johannesV_ has joined #gnuenterprise
*** johannesV has quit IRC
*** jamest has joined #gnuenterprise
<holycow> reinhard, do you use gnue in your business?
*** jamest_ has joined #gnuenterprise
*** jamest has quit IRC
*** jamest_ has quit IRC
*** jamest has joined #gnuenterprise
<johannesV_> hmmm
<reinhard> holycow: yes, but currently only in a small field
<reinhard> holycow: see gnue-contrib/hotline
<holycow> ah cool.
*** dimas_ has joined #gnuenterprise
*** dimas_ has quit IRC
*** holycow has quit IRC
<reinhard> bbl
*** dcmwai has joined #gnuenterprise
*** jcater has joined #gnuenterprise
*** holycow has joined #gnuenterprise
<johannesV_> jcater, can you tell me which "paramstyle" the oracle-db-drivers use ? (cxoracle/dcoracle)
<jcater> :field
<chillywilly> hi
<johannesV_> field ?
<chillywilly> today is my birthday ppl
<johannesV_> i thought there is only 'qmark', 'format', 'named', 'numeric' and 'pyformat' ..
<johannesV_> chillywilly, and mine is tomorrow :)
<jcater> select * from mytable where column = :name
<chillywilly> cool
<jcater> happy b'day, bi0tch
<johannesV_> jcater, so it is 'named'
<jcater> mine is monday
<jcater> johannesV_: I think that's what they call it
<johannesV_> hmm
<chillywilly> november rules
<jcater> the others may work
<jcater> :1 may work as wel
<jcater> well
<jcater> give me a second to test
<johannesV_> i've re-activated the maxdb/sapdb driver
<jcater> but I use named all the time in my python script
<johannesV_> jcater, you can look at module.paramstyle
<johannesV_> (where module is cxoracle or dc...
<johannesV_> on a cursor.execute (statement, params), do you define params as dictionary or sequence ?
<chillywilly> bleh, this debian mirror is slow
<johannesV_> well, sapdb-driver uses 'named' style but requires the params given as sequence
<johannesV_> and raising an error if it is a dictionary
<johannesV_> so i would like to know if i can change __param_named () in dbsig2.connection to return parameters as sequence
<johannesV_> (instead of dict)
<chillywilly> geez, when is sid going to get firefox 1.0?
* johannesV_ back in a few minutes
* johannesV_ back
*** johannesV_ has quit IRC
*** johannesV has joined #gnuenterprise
<holycow> hey guys
<nickr> sacre vache!
* holycow anoints nickr 
<holycow> chillywilly, good point
<holycow> chillywilly, perhaps they were waiting for 1.0 final to arrive
<holycow> it was definately pointless to roll out the rc candidates consider development was fairly breeze in the last part for mozfox
<nickr> firefaux is for chumps
<holycow> firfaux? rofl :)
<mo> ah ben ca parle francais :)
<mo> bon ben a plus :)
*** mo has quit IRC
<nickr> non, je ne parle pas francais
<holycow> il parle "faux" francais
<holycow> *cough*
<holycow> >_>
<nickr> Yea pretty much.
<holycow> that probably makes only sense reading it as an english person trying to write french
<holycow> <_<
<nickr> hm?
<chillywilly> firefox rules
<nickr> chump
<holycow> lol ;)
<holycow> i love these live cd's
<johannesV> i'm off for today ...
<johannesV> jcater, would you please write the oracle's paramstyle when you've time to; i'll read the logs tomorrow. thanks
*** johannesV has quit IRC
<holycow> knoppix really comes in handy when repairing windows machines
<jcater> johannesV: sorry, I got distracted
<jcater> by lunch
<jcater> :)
<jcater> >>> import cx_Oracle
<jcater> >>> print cx_Oracle.paramstyle
<jcater> named
<jcater> >>>
<jcater> >>> import DCOracle2
<jcater> >>> DCOracle2.paramstyle
<jcater> 'numeric'
<jcater> fwiw, I know named works on both
<jcater> not sure if both support numeric though
<jcater> curs.execute('select * from dual where dummy = :1', 'X');
<jcater> Traceback (most recent call last):
<jcater>   File "<stdin>", line 1, in ?
<jcater> TypeError: function takes exactly 1 argument (2 given)
<jcater> so, I'd say no
<holycow> just out of curiosity, how is work on appserver directed?  is there a gameplan in place featurewise or is it developed organically based on need?
<jcater> have you read their whitepaper?
<jcater> it seemed to address the future plans
<jcater> iirc
<jcater> (it has been a while since I read it)
<holycow> ehe :) obviously i haven't /me browses some more
<holycow> :)
<holycow> oh neat
<holycow> i don't know why i haven't poked deeper into the site
<holycow> before
<holycow> v cool
<reinhard> holycow: there's also the feature plan (i think it should be available on the web, too, otherwise look at the TODO file)
<reinhard> however, we currently expect appserver development to stall a bit
<reinhard> because appserver is quite much there where we want it for now
<reinhard> more or less usable for small scale business applications with no speical security needs
<reinhard> and as we started to work on packages we hit quite some issues with forms and reports
<reinhard> for that reason we plan to put our forces there for the next few weeks
<reinhard> (as you might already have seen if you watch our commits)
<reinhard> I also still hope that jamest and jcater come back to active development
<reinhard> :-)
<jcater> I'm hoping myself
<jcater> been trying to get things in place at work so I can
<jamest> as have i
<holycow> ^5
<holycow> you guys are amazing
<holycow> reinhard, danke, thats very cool
<holycow> there is something to be said for deciding on a 'freeze' where people can develop for the particular freeze number i think, but still awesome
<reinhard> holycow: I think I agree about the freeze thing
<reinhard> actually I would like that kind of freeze in appserver to happen now
<reinhard> but we are still at the very start of developing packages
<reinhard> and somehow we want to be able to fix conceptual bugs the pop up
<reinhard> (like e.g. the question whether non-nullable fields should be checked before or after OnValidate triggers....)
<reinhard> OTOH
<reinhard> I think I would dare to develop against the current state of appserver
<reinhard> because still while adding features I think we haven't broken backwards compatibility for a long time
*** feti has joined #gnuenterprise
<reinhard> jbailey: are you *the* jbailey?
<reinhard> :-P
<reinhard> any chance we can get our deb packages updated?
<reinhard> looks like sarge is getting colder and colder
<jbailey> reinhard: No, I'm just *a* jbailey.  Didn't you hear I come in six packs now?
<reinhard> jbailey: rofl
<jbailey> ajmitch: Got new gnue stuff up on mentors yet?
<reinhard> ... and we should get new packages in before it freezes ...
<reinhard> bbl
*** cilkay has quit IRC
<holycow> holy shit
<holycow> feti!
<holycow> heyd dude :)
<holycow> guys, feti
<holycow> feti, guys
<holycow> :)
<holycow> feti has his own php framework, looked v cool last time i had a chance to take a peek
<jbailey> holycow: Are you trying to scare him?
<holycow> http://mojavi.org/ <-- :)
<holycow> jbailey, hey :) ehe, naw fetis cool
<chillywilly> mojavi looks interesting
<holycow> yeah if your focusing only a web based front end, an mvc like mojavi could save you a lot of time
*** dcmwai has quit IRC
<holycow> feti, msg me if you get a chance
*** cilkay has joined #gnuenterprise
*** sjc has joined #gnuenterprise
*** sjc has quit IRC
*** kilo has joined #gnuenterprise
*** kilo has quit IRC
<ajmitch> morning
<jbailey> ajmitch: reinhard wants updated packages.  Are they on mentors?
<ajmitch> nope, not yet
<jbailey> reinhard: See?  What am I to do?
* jbailey hides.
* ajmitch will only be around for another few min
<ajmitch> bah
<ajmitch> I've got them mostly done :P
<jbailey> ajmitch: It's cdbs dude.  Does it really take more than 5 minutes? =)
<ajmitch> yeah, course it does :)
<ajmitch> > 20 min just to upload packages :)
<jbailey> ajmitch: bah!
* jbailey tosses a nickle a to ajmitch 
<jbailey> Go buy some bandwidth!
<jbailey> =)
* jbailey hides again.
<ajmitch> thanks
<ajmitch> it'll buy me about 100kb of data allocation :)
<jbailey> wow.  That's insane.
<ajmitch> nah
<ajmitch> 20c/MB for full rate DSL :)
<jbailey> For a $50 up front cost, my home connection is going up to a 5MB downlink.
<jbailey> (No additional monthly cost over what I pay now)
<ajmitch> 2Mbit/192Kbit has only just been released
<ajmitch> after years of 128Kbps being the main speed
<ajmitch> sigh, forgot to package forms ;)
<ajmitch> I've got to go, sorry
<ajmitch> will be uploading packages later :)
<jbailey> ajmitch: /msg me when you do.
<jbailey> (or email, as always)
<reinhard> thanks both jbailey and ajmitch
<reinhard> good night all
*** reinhard has quit IRC
*** jamest has quit IRC
*** cilkay has quit IRC
*** cilkay has joined #gnuenterprise
*** cilkay has quit IRC
*** jcater_ has joined #gnuenterprise
*** jcater has quit IRC
*** bluesbaron_ has joined #gnuenterprise
*** bluesbaron_ has left #gnuenterprise
*** holycow has quit IRC
*** cilkay has joined #gnuenterprise
*** jcater has quit IRC
*** wayneg has quit IRC
*** holycow has joined #gnuenterprise
<feti> sup cow