*** jcater_ has joined #gnuenterprise
*** jcater has quit IRC
*** reinhard has joined #gnuenterprise
*** reinhard_ has joined #gnuenterprise
*** johannesV has joined #gnuenterprise
*** reinhard has quit IRC
*** dneighbo has joined #gnuenterprise
*** dneighbo_ has quit IRC
*** btami has joined #gnuenterprise
<reinhard> hello btami
*** holycow has quit IRC
<btami> hello reinhard
*** Morphous has joined #gnuenterprise
*** dimas has joined #gnuenterprise
*** Amorphous has quit IRC
*** rsb has quit IRC
*** kilo has joined #gnuenterprise
<kilo> morning
<reinhard> hi kilo
*** dcmwai has quit IRC
*** btami has quit IRC
*** dimas has quit IRC
*** SachaS has joined #gnuenterprise
<SachaS> hi everyone
*** dimas has joined #gnuenterprise
<kilo> hi SachaS
<SachaS> hi kilo
<SachaS> see you later
<kilo> you busy writing the interview?
*** SachaS has quit IRC
*** lekma has quit IRC
*** jcater_ has quit IRC
*** lekma has joined #gnuenterprise
*** dsmith has joined #gnuenterprise
*** lekma has quit IRC
*** sjc has joined #gnuenterprise
*** jamest has joined #gnuenterprise
<reinhard> hello jamest
<reinhard> if i have a datasource
<reinhard> with no attached block
<reinhard> what is the correct way of testing whether the current resultset of the datasource is empty?
*** dwho has joined #gnuenterprise
<dwho> Hi in http://www.gnuenterprise.org/downloads/current.php under Debian GNU/Linux Packages (*.deb format)....there are .exe files.There should be .deb files right?
<reinhard> hello dwho
<reinhard> i think this is a bug in the html
<reinhard> we do not have debs yet
<reinhard> and for some reason it lists the exes there
<reinhard> istead of leaving the complete section out
<dwho> ok...coz I was to install o.53 but just checked back to see if there is anything new:)
<dwho> Also where can I check out the common errors page ....
<reinhard> the common errors page?
<dwho> I mean say when one is installing and encounter an error.....maybe he/she can refer to a list of commonly encountered errors
<dwho> I was trying to install GNUe-Common and when I typed ./setup.py install...I got an error saying there was some problem with my python nstallation : running install
<dwho> error: invalid Python installation: unable to open /usr/lib/python2.3/config/Makefile (No such file or directory)
<dwho> so i trying to figure that out 
<dwho> :)
*** dwho has quit IRC
<reinhard> i can confirm that this is *no* commonly encountered error
<reinhard> err
<reinhard> why are people always so impatient when i'm talking on the phone
<reinhard> :((
*** kilo has quit IRC
<jamest> reinhard: is the datasource being used to query stuff?
<jamest> i think len(datasource) may work
<jamest> but I'd have to check the code 
*** johannesV_ has joined #gnuenterprise
<jamest> ah
<jamest> maybe not
*** lekma has joined #gnuenterprise
*** lekma has joined #gnuenterprise
<jamest> seems like a datasource could just define a __len__ that returns the length of it's result set
<jamest> er, a datasource trigger function
*** johannesV has quit IRC
<reinhard> jamest: that would be good, yes
<jamest> would datasource.count() be acceptable?
<jamest> or does len() make for sense?
<jamest> and could you try len(datasourceName)
<jamest> i don't think the latter works but it's worth a shot :)
<reinhard> jamest: i don't mind how it would be done
<reinhard> len(datasource) doesn't seem too intuitive at first sight
<jamest> ok
<jamest> another option is datasource.recordCount
<jamest> not as a function but as a property
<reinhard> also works for me
<jamest> you're being too damn easy to get along with
<jamest> :)
<jamest> I'm commit something completely untested
<jamest> that I think will work
*** reinhard_ has joined #gnuenterprise
<jamest> [09:01:53 am] <jamest>you're being too damn easy to get along with
<jamest>  [09:01:56 am] <jamest>
<jamest>  [09:04:26 am] <jamest>I'm commit something completely untested
<jamest>  [09:04:33 am] <jamest>that I think will work
<jamest> for you reinhard_
<reinhard_> jamest: thanks, i got that
*** dsmith has quit IRC
*** marcello has joined #gnuenterprise
<jamest> marcello: hello
<marcello> hi jamest
<jamest> reinhard: we both know that jcater will show up in 30 minutes and tell us something else was already in there to do this :)
<marcello> ive seen the new look of the gnue http site, it's much more professional :) i haven't been looking at gnue for a while, but in a few days I will be able to dig it again :)
<jamest> marcello: glad to hear it
<jamest> feel free to ask any questions you may have
<marcello> ok, what's the best package to look at for an example on how to make gnue apps (other than the ones in the example folders i mean)
*** reinhard has quit IRC
<jamest> marcello: what level you wanting to work?
<jamest> just creating forms?  writing python apps of your own?
<marcello> creating forms with table and subtables and fairly simple triggers, and a couple python routines
<marcello> i have no big problem with python yet, i like it
<marcello> subtables = related tables
<marcello> i have an app developed with aci 4d and i would like to use gnue instead, its for a small shop
<jamest> Well, in the gnue-contrib repository Sachas has been creating a babyerp
<marcello> oh ok i will check it out
<jamest> it uses appserver instead of directly to tables but the general layout is the same
<jamest> there is also gnue-pointofsale but that's not using forms at all
<jamest> it does provide interface to a cash drawer i believe
<jamest> but for those relationships you describe gnue-designer provides a wizard to build your forms directly from the tables
<marcello> ok i wonder what protocols italian cash register will have (anyway there are italians developing an open source app server that interfaces with cash registers so maybe i can try to interface gnue app server to that one... we'll see
<marcello> the wizard function is very handy, 4d had one, too
<dimas> jamest: how can i get gnue-contrib from svn?
<dimas> just updated and do not see it
*** dsmith has joined #gnuenterprise
<jamest> ah
<jamest> svn co svn://www.gnuenterprise.org/var/svn/gnue-contrib
<jamest> i guess we don't have that on our website
<jamest> it has 3 projects in it hotline, babyeerp, and web-po
<reinhard> jamest: i have another problem
<reinhard> i have three tables
<reinhard> 1. customer
<reinhard> 2. contact
<reinhard> 3. ticket
<reinhard> every customer has n contacts
<reinhard> every contact has n tickets
<reinhard> i want to query all tickets for a given *customer*
<reinhard> regardless of the contact
<reinhard> how can i do that in forms?
<dimas> jamest: got it, thanks
*** jcater has joined #gnuenterprise
<reinhard> jamest: is there any chance to have working dropdowns in the forseeable future?
<reinhard> it's quite hard to debug non-trivial forms when you don't always see the real content of a field :(
<jamest> what!?!
<jamest> i thought they did work again, only  mouse was wack
<jamest> reinhard: what are the tickets tied to?
<marcello> goodbye for now
*** marcello has quit IRC
<reinhard> contacts are a detail for customers
<reinhard> tickets are a detail for contacts
<reinhard> i want all tickets for a customer
<reinhard> re dropdowns
<reinhard> it's hard to say exactly what is broken
<jamest> the event processing in dropdowns is wack
<reinhard> but there are quite some cases where a more or less random value is displayed
<reinhard> it's hard to reproduce in an isolated way
<reinhard> because the form is quite complex already
<jamest> i don't understand how the dropdowns can even work as they are
<jamest> but was told they worked ok on windows and seemed good enough on linux that I left alone for release
<jamest> but I can probably get back in and redo them next week
<jamest> i know I won't have time before then :(
<reinhard> i'm sure they work good enough for non-complex forms
<jamest> as for your detail of a detail
<jamest> i don't have anything like that
<reinhard> well that's even easy
<jamest> i would think you would have to unlink the ticket detail
<reinhard> there is a "action" table that is a detail for the ticket :)
<reinhard> jamest: and then?
<jamest> populate it via a trigger
<reinhard> how would i make the query?
<jamest> not sure yet
<jamest> trying to think of a way thru this :)
<reinhard> please remember i can't use sql
<jamest> that's what I'm trying to think around
<jamest> i imagine you need to edit this block as well?
<reinhard> no
<reinhard> just list them up in a dialog and let the user select one
<reinhard> and before i pop up the dialog
<reinhard> check if there is more than 0
<reinhard> rows
<jamest> ah
<jamest> this may get a little ugly
<jamest> but i imagine you can unlink that block from the table
<jamest> er datasource
<jamest> then build the list via a trigger
<jamest> using the datasources simpleQuery
<jamest> and a for loop :)
<jamest> so you'd simplyQuery on for a list of contacts for the top master
<jamest> then loop thru them simplyQuerying for tickets on each contact
<reinhard> and how would i merge these results into a single block?
<reinhard> IOW
<reinhard> how would i display the final list in a dialog and let the user select one?
* reinhard starts to understand why people want appservers they can put the logic in :)
<jamest> sigh
<jamest> you want it all dont you? :)
<jamest> I'd do it inside a self contained dialog form
<jamest> you'd have to build the list on the on activation trigger
<jamest> display/selection how I imagine you want it to wrk is going to be impossible in the current UI
<jamest> i've been thinking about adding a new option to entry
<jamest> style=list
<jamest> which would make it doable
*** jamest has quit IRC
*** jamest has joined #gnuenterprise
<lekma> hello everybody
<jamest> hello
<lekma> i have another of my infamous stupid questions
<lekma> in the form developper guide runForm() is advertise as
*** SachaS has joined #gnuenterprise
<lekma> a trigger one can call as follow: form.runForm( 'myform.gfd', { 'division' : 101 } )
<lekma> when i do that i get 
<lekma>  Traceback (most recent call last):
<lekma>   File "<unknown_executioncontext>", line 9, in __main______gnue_forms_GFObjects_GFLayout_GFLayout_instance_at_0x4175728c____pg_vars_btn_edit_ON_ACTION_trg_edit
<lekma>   File "/home/malek/gnue-cvs/gnue/gnue-common/.cvsdevelbase/gnue/common/logic/NamespaceCore.py", line 315, in __call__
<lekma>     return self._objectFunction(*args)
<lekma> TypeError: triggerRunForm() takes exactly 2 arguments (3 given)
<jamest> lekma: bug!
<lekma> :)
<jamest> i can't look atm as I have to help user
<reinhard> jamest: i think i have to solve stuff like that in appserver
<reinhard> my understanding is that such things are outside the scope of forms
<jamest> nah
<jamest> you can do most if it in forms
<jcater> reinhard: keep in mind we do all our logic w/in forms
<jamest> but at the moment that falls into the section "requires effort"
<jcater> so I doubt it's fair to say anything is outside the scope
<jamest> the thing that forms is missing is a list widget
<jcater> just maybe not as easy or complete as it should be :)
<jamest> which I see as the same as a dropdown only with > 1 displayed
<jcater> jamest: I'd like to see a generic "block" layout component
<jcater> oh, nevermind... you are describing something different
<jamest> :)
<jamest> i'm all for changes
<jcater> I've never actually needed a list widget
<jamest> once we get what we having a little for feature complete
<jcater> never thought about it
<jamest> i've got a single case where it'd make it easier to replace a java app
<jamest> as when they save their work it pops up a dialog that lets them choose a state to save in
<jamest> state meaning == pending this, ready for that, i give up
<jamest> types of selections
<jcater> basically just a dropdown that's not a dropdown :)
<jamest> lazy sobs used to have this as a dropdown years ago
<jamest> but they ignored it
<jamest> so the replacement java app I've grown to love made it mandatory 
<jamest> so now they can just always click ok and get the default select
* jamest beats the users w/ a stick
<reinhard> another question
<reinhard> can i do joins with datasources?
<jamest>  /msg jcater man, he's sure trying to make gnue do usefull things, how do we stop him?
<reinhard> lol
<reinhard> i mean something like
<jamest> i think you can readonly
<reinhard> have a datasource with more than one table
<jamest> but I don't think we've ever officailly supported it
<jamest> table="foo,bar"
<jamest> IIRC
<reinhard> ok
<jamest> with <condition>s to enforce the join
<jamest> i **think**
<reinhard> how much effort would you estimate to "officially" support it?
<jcater> I think it's about as official as you can make it
<jcater> most of the dbsig2's support it, at least
<reinhard> :)
<jcater> don't know what it'd take to get the appserver dbdriver to support it
<jcater> I'd like to see an overhaul of datasources
<jcater> i.e., the supported <datasource> definitions
<jcater> not the underlying drivers :)
<jcater> to expand our definition to support a lot more usage cases
<jamest> i didn't think we ever really came up w/ a clean conditional 
<jcater> well, it's lunch time
<jcater> and jcater's food is far more important than datasources
* jcater ducks
<jamest> sigh
<jamest>  /msg reinhard damn cater, can't we just get it working once before we change things again
<jamest> :)
<jamest> on a more serious note
<jamest> can/should any of these things wait till after 1.0
<jamest> as I think forms/designer beats the crap out the last pgaccess i used
<jamest> but we're never mentioned in pgsql MLs
<reinhard> hmmm
<reinhard> wel
<reinhard> well
<reinhard> there are some minor things i'd consider before 1.0 if i were you
<reinhard> reinhard@london:~/prj/gnue/gnue-forms/src$ find . -name "*.py" | xargs grep -i "todo" | wc --lines
<reinhard>     212
<reinhard> reinhard@london:~/prj/gnue/gnue-forms/src$ find . -name "*.py" | xargs grep -i "hack" | wc --lines
<reinhard>      21
<jamest> that's kind of my point :)
<jamest> is what on the todo list should wait till after a 1.0
<reinhard> reinhard@london:~/prj/gnue/gnue-common/src$ find . -name "*.py" | xargs grep -i "todo" | wc --lines
<reinhard>     323
<reinhard> reinhard@london:~/prj/gnue/gnue-common/src$ find . -name "*.py" | xargs grep -i "hack" | wc --lines
<reinhard>      45
<jamest> as compared to lots of software we're post 1.0 state in a lot of areas
<reinhard> i have not looked at forms that much
<reinhard> just in common i see quite some things that are kinda "unfinished"
<jamest> yes
<jamest> maybe I'm not coming across correctly
<jamest> I'm trying to say
<jamest>   before we take on rewriting things that are working "ok" but not perfect (like the datasource definition rewrite)
<jamest>   shouldn't we push some of these "nice" ideas out past 1.0 and focus on getting the lacking/rough/unfinished stuff done
<jamest> I'm as guilty as anyone on wanting change
<jamest> but i'm starting to question if said change is really impacting a 1.0 release
<reinhard> well i don't care about version numbers that much
<jamest> meaning have we gotten to the mindset "1.0 happens when gnue is perfect"
<jamest> i don't either
<reinhard> the main problem i see is common
<jamest> but on a PR front and a stability front it would matter to the end users
<reinhard> for all version numbers < 1.0 we somehow reserve the right to break the api
<jamest> right
<reinhard> and i remember quite some changes *i* did in common (and i didn't do too much) over the last 2 months
<reinhard> where i *had* to break the api to make it *somewhat* clean
<reinhard> and logical
<jamest> that's fine
<reinhard> not sure what changes in common *you* plan to make
<reinhard> after 1.0
<jamest> i'm not saying release 1.0 now
<jamest> i'm saying do we need to stop ourselves from forever spiralling away from the magic 1.0 number
<jamest> the 0.5.x series lifespan is measure in years
<jamest> with several major changes in that timeframe
<lekma> --- ./gnue/gnue-forms/src/GFForm.py     2004-04-15 18:48:35.000000000 +0200
<lekma> +++ ./gnue-touched/gnue-forms/src/GFForm.py     2004-04-15 18:49:29.000000000 +0200
<lekma> @@ -918,11 +912,11 @@
<lekma>    # Allows the trigger to launch a new instance
<lekma>    # of gnue-forms running a different form.
<lekma>    #
<lekma> -  def triggerRunForm(self, fileName):
<lekma> +  def triggerRunForm(self, fileName, parameters):
<lekma>      from GFInstance import GFInstance
<lekma>      instance = GFInstance(self._instance.manager,
<lekma>                            self._instance.connections,
<lekma> -                          self._instance._uimodule,  1)
<lekma> +                          self._instance._uimodule,  1, parameters)
<lekma>      instance.addFormFromFile(fileName)
<lekma>      instance.activate()
<lekma> jamest: above is a quick fix from a colleague of mine (ghoulseine is the name)
<lekma> for the runForm() parameters problem i told you a few lines before
<lekma> it works for me (tm)
<lekma> :)
<jamest> lekma: thank you
<jcater> I wasn't in any way talking about a datasource  rewrite
<jcater> fwiw
<jcater> more, standardize its feature set
<jcater> set a direction for it
<jcater> that sort of thing
<lekma> is it possible to set the value of a form parameter from a form trigger??
<jcater> form.setParameter(name, value) iirc
<lekma> thx
<jamest> a direction for gnue?
<jamest> argh!!!!!!!!!!!
* jamest runs in ph34r
<reinhard> bbl
*** SachaTV has quit IRC
*** johannesV_ has quit IRC
*** bamse has joined #gnuenterprise
*** kilo has joined #gnuenterprise
*** lekma has quit IRC
*** Suit has joined #gnuenterprise
<Suit> Hi
<Suit> been a while since I was here
<jcater> welcome back
<Suit> :) yeah well
<Suit> I see you have a lot of docs with package proposals on-line
<Suit> How far you recon they are? Still under review? (The proposals)
<jcater> basically, yeah
<Suit> Hmm must read them and see if they have or haven't got what I found missing or good with others :) Maybe I can bring something new
*** jcoruna has joined #gnuenterprise
<Suit> I moved to other job
<Suit> from logistics (container transport) to manufacturing and trade
* Suit wonders what the difference is between "lost" and "inventory adjustment"
<Suit> ok the latter could also mean "we found more"
*** jcoruna has quit IRC
<Suit> Well, quite a read already.
<Suit> Btw I saw also firebird support was added
<Suit> which is currently the db-flavour-of-choice?
<Suit> iirc it was sapdb
<Suit> it still is?
<jcater> um
<jcater> I think postgres is still the db of choice
<jcater> has been since the project began
<jcater> but we do recommend firebird
<jcater> esp. for people on windows
<jcater> SapDB is now a MySQL project, so it has been renamed to MySQL MaxDB
<Suit> Heh, hey my nick does not mean I run windows :)
<Suit> btw postgresql is also available for windows.
<jcater> yes
<jcater> but our windows developers primarily use firebird
<jcater> so for people wanting to test, we recommend using what the developers use :)
<chillywilly> postgres shouldn't behave any differently under winders though
<chillywilly> ;P
<jcater> chillywilly: yes
<jcater> but
<jcater> we have people complain about not getting demos up and running
<jcater> so it's a lot easier if we have people on board
<jcater> who can say "do x, then do y, then click z"
<jcater> etc
<Suit> Well, I have looked at a competitor a few days ago (sql-ledger)
* jcater isn't discouraging anyone from doing postgres on windows
<Suit> ok I still have to recuperate from that one
<jcater> or doing sapdb
<chillywilly> haha
<jcater> actually, only one I discourage people from using is MySQL =)
<Suit> It is better on linux
<Suit> So well, I am still doubting, step in on what already is there or build myself from scratch to get what I need
<Suit> time is always a pita
<jcater> tough to say
<jcater> I would personally encourage either, if you plan to use the GNUe tools to do it :)
<Suit> :)
*** SachaS has joined #gnuenterprise
<chillywilly> hmmm, one can only add users to groups in *nix right?
<Suit> But serious, is there already some basic coding/datamodel for manufacturing/assembly manufacturing and so?
<chillywilly> no groups belonging to groups?
<Suit> on the os/fs level?
<Suit> then no
<chillywilly> by hacking /etc/group
<Suit> Ah hum no
<chillywilly> hmmm, yea that makes no sense
<Suit> would be new to me at leat :)
*** sjc has quit IRC
<Suit> ok well I will ponder on
<Suit> cya
*** Suit has quit IRC
*** jamest has left #gnuenterprise
<dsmith> chillywilly: Groups and users are separate namesspaces.  You can have a group name that is the same as a user name.
<dsmith> In fact, most modern distros automaticaly add a group that the same name as the user.
*** dsmith has quit IRC
*** jcater has quit IRC
<reinhard> night all
*** reinhard has quit IRC
*** kilo has quit IRC
<SachaS> night
*** SachaS has left #gnuenterprise
<chillywilly> yea
<chillywilly> bleh, dsmith ran away
*** newbie has joined #gnuenterprise
*** newbie has left #gnuenterprise
*** newbie has joined #gnuenterprise
*** newbie has left #gnuenterprise
*** ogger has quit IRC
<bamse> is there anyone here who could help me a little with triggers on forms ?
*** jcater has joined #gnuenterprise
<chillywilly> hi jcater 
<jcater> word
<bamse> jcater: do you know if I can setup a "PRE-FOCUSIN" trigger for a comboBox on a form ? when I create one, I get a traceback ending with :
<bamse> File ".../gnue/forms/GFObjects/GFEntry.py", line 162, in triggerSetValue
<bamse> DB000:     return self._field.triggerSetValue(*args, **parms)
<bamse> DB000: AttributeError: GFEntry instance has no attribute '_field'
*** jamest has joined #gnuenterprise
<jcater> hmm
<jcater> is field="..." set on that <entry>?
<jcater> and block=".."
<bamse> well, in fact, I am trying to hack the classrepository a bit
<bamse> the entry is type entry in the property page
<bamse> I take the sample form, which runs ok
<bamse> and just add a trigger the says "print("hello")
<bamse> then, I get the problem
<bamse> well, I see in GFEntry.py that only get and set functions are recognised as trigger functions, could it be the source of my problem ?
<bamse> ok, in fact, it seems that adding the trigger to the Entry fails, but adding it to the field works fine
<bamse> I must be forgetting something
<bamse> nevermind, I'll go on with field trigger
<bamse> thanks for the help 
*** rsb has joined #gnuenterprise
*** test has joined #gnuenterprise
*** bamse has quit IRC
*** rsb has quit IRC
*** jamest has quit IRC
*** rsb has joined #gnuenterprise
*** rsb has quit IRC
*** rsb has joined #gnuenterprise
*** holycow has joined #gnuenterprise
*** dneighbo_ has joined #gnuenterprise
*** dneighbo has quit IRC