*** johannesV has joined #gnuenterprise
*** btami has joined #gnuenterprise
<btami> good morning
*** johannesV_ has joined #gnuenterprise
<johannesV_> good morning
*** kilo has joined #gnuenterprise
*** reinhard has joined #gnuenterprise
<kilo> good morning
*** johannesV has quit IRC
<reinhard> good morning all
<johannesV_> good morning :)
*** bigbrother` has joined #gnuenterprise
*** bigbrother has quit IRC
*** yure has joined #gnuenterprise
*** btami has quit IRC
*** jamest has joined #gnuenterprise
<jamest> reinhard: you started validating the typecasts in fields?
<jamest> as I'm getting an odd error with svn head
<jamest> XML markup error
<jamest> 'datetime' is not valid for typecast-attribute which allows these values only:
<jamest> 'Default', 'Description', 'ValueSet', 'Typecase', 'Label'
<reinhard> not that I remember
<reinhard> might be however that johannesV_ changed it
<jamest> ok, that error msg is wack
<johannesV_> :)
<johannesV_> that was me, indeed
<reinhard> I'm searching a differnt bug right now
<johannesV_> this is because the options defined in a ValueSet will be checked now
<jamest> johannesV_: i think it's reporting the wrong level though
<jamest> it typechecks right
<johannesV_> sure ?
<jamest> i just switched a typecast='number' to 'numberz'
<jamest> and instead of printing the allowed values it prints these as allowed values
<jamest> 'Default', 'Description', 'ValueSet', 'Typecase', 'Label'
<johannesV_> ah, i think i see it already
<jamest> which look more like gparse dict keys
<reinhard> also I think I found my bug...
<jamest> i'm building a glimpse index on the source to look but it takes a while :)
<reinhard> jamest: if a trigger is defined within a <dialog>, what should "form" point to in your opinion?
<reinhard> the main form, or the dialog?
<jamest> also, fwiw pylint 0.11 is out with the block level disable
<johannesV_> jamest, found and fixed
<jamest> reinhard: it should point to the dialog form
<jamest> all <dialog> is is an alias for <form type="dialog"> or some such
<jamest> so in a dialog you have an on-startup trigger that does
<reinhard> so there would be no way to access the main from from within a dialog?
<jamest> form.setFeature('GUI:MENUBAR:SUPPRESS',1)
<jamest> so you can flip the ui elements of that dialog form off
<jamest> reinhard: you passed values back and forth via the activateDialog() function
<reinhard> yes
<reinhard> ok
<jamest> activateDialog('dialogFormName', parameters)
<reinhard> so we want to keep it that way, right?
<jamest> though I think this broken when parameters became a caseless dict
<reinhard> every dialog would get its own global namespace?
<jamest> i hacked something in to get things working for me but i think it was flaky enough i redid my forms here to not depend on it
<jamest> i don't see any reason to share them
<reinhard> and, for example, blocks and datasources from the main form would also not be visible in a dialog trigger?
<jamest> i had hoped that dialogs would be self contained with a standard API
<jamest> that way if I wanted a custom dialog that asks for the base info on a new customer
<jamest> i could hook that dialog into multiple forms
<reinhard> ok
<reinhard> makes perfect sense
<jamest> via the import we support(ed)
<reinhard> actually it seems like it was like that before
<reinhard> I wonder where I broke it :-)
<jamest> caseless dict for parameters on forms broke stuff there
<jamest> i don't think it was ever really fixed after that
<reinhard> it was fixed as far as it works again for me :)
<jamest> :)
<reinhard> <reinhard> I wonder where I broke it :-)  <--- I meant the separate namespace per dialog
<reinhard> looking at that now
<jamest> hmmmm
<reinhard> found it
<jamest> date, text, number is all we have
<jamest> i thought we did a datetime too
*** kilo has left #gnuenterprise
<reinhard> hmm
<reinhard> implementation of GFField even only checks for "number"
<reinhard> and treats all other values the same
<reinhard> AFAICT
<reinhard> jamest: could you please do me a favour and remove me from the list of commit-gnue mailing list moderators?
*** derek has quit IRC
<jamest> reinhard: they are the same to the field
<jamest> it's the input handler that cares about how they are presented to the UI
<jamest> as text
<jamest> iirc
<reinhard> yes
<reinhard> but that's not dependent on the "typecast" attribute
<reinhard> :)
<jamest> hmmm
<jamest> what about boolean
<jamest> i swore boolean worked
*** johannesV__ has joined #gnuenterprise
<reinhard> I would guess that it was not checked at all in earlier versions
*** johannesV_ has quit IRC
<reinhard> yep
<reinhard> revision 8376, johannes, check for valid attributes if a value-set is defined
<reinhard> 1 week 1 day ago
*** yure has quit IRC
*** derek has joined #gnuenterprise
*** sacha has quit IRC
*** jamest has quit IRC
<reinhard> <reinhard> jamest: could you please do me a favour and remove me from the list of commit-gnue mailing list moderators?   <-- did you see that?
<jcater> reinhard: I removed you
*** jamest has joined #gnuenterprise
*** yure has joined #gnuenterprise
*** sjc has joined #gnuenterprise
<reinhard> jcater: thanks
<jamest> in the old forms namespace
<jamest> you could do a startup trigger
<jamest> on-startup
<jamest> that does something like
*** johannesV__ has quit IRC
<jamest> global FixedPoint
<jamest> from gnue.common.external.fixedpoint import FixedPoint
<jamest> then in other triggers in the form you could just use FixedPoint
<jamest> other examples of this are in the forms tech reference
<jamest> that no longer works
<jamest> which will break a ton of stuff for me :)
<jamest> as you can also define functions in on-startup with a global then use those functions form wide
<reinhard> I tried to make that work but I failed
<reinhard> what is the trick about it?
<reinhard> it always told me that the variable name I had after "global" doesn't exist
<reinhard> hmmmm
<reinhard> let me try something
<reinhard> erm
<reinhard> it actually works for me
<reinhard> I wonder
<reinhard> does your case invove any dialogs?
<reinhard> as it might have changed that dialogs got their own global namespace
<jamest> i'll have to look later, the users figured out my day was going good
<jamest> so took care of that this afternoon
<reinhard> ok
<reinhard> just let me know
<reinhard> because within the same form, globals get shared between triggers OK for me
<jamest> ok
<jamest> the difference for me that jumps out
<jamest> is that this is a from foo import bar
<reinhard> (I was wrong when I said it wouldn't work - it had always worked, but I had a bug in my test..)
<jamest> not that that makes sense
<reinhard> even that works for me AFAICT
<reinhard> in ON-STARTUP I have
<reinhard> global DateTime
<reinhard> from mx import DateTime
<reinhard> and in a button trigger I can use DateTime perfectly
<jamest> ok, thanks, i'll dig in more later to see what's up in that form
*** dimas has quit IRC
*** yure has quit IRC
*** sandroid has joined #gnuenterprise
<derek> http://www.gestix.com/  spammed me toeday
<jamest> and you share with us?
*** lupo__ has joined #gnuenterprise
<reinhard> How would you call this if A is the parent of B, B is the parent of C, and C is the parent of A ?
<SachaS> reinhard: a circular list?
<jcater> reinhard: a design error?
<reinhard> jcater: yes
<reinhard> jcater: I would need a name for this design error to present to the user in an error message :)
<jamest> circular reference?
<jamest> derek issue 00321.4
<reinhard> ok, thanks
*** kilo has joined #gnuenterprise
<derek> jamest: yeah i like to share my spam :)
* derek just wondered if anyone here had looked at it since it claims to be an ERP
<reinhard> never heard of it before
<reinhard> funny thing is it calls itself "open source based" but seems to be 100% proprietary
<derek> it runs on linux
<derek> that makes it open source based right? ;)
<chillywilly> hah
<reinhard> jamest, jcater: just committed the lowest level of the class hierarchy we talked about yesterday. Please have a look whether this matches what you expected
<jamest> reinhard: cool, i'll look at it tonight when I get home
<jcater> I'll try to tonight too
<reinhard> nice
<reinhard> not that it's much though
<reinhard> I've implemented a Node class
<reinhard> I consider a TypedNode subclass that introduces what we now call _type
<reinhard> that can implement find_parent_of_type, find_children_of_type etc
<jamest> heading home now, will read logs
<jcater> I have a wife who unfortunately wants to spend time with me tonight
<jcater> so not sure when I can read it
<reinhard> and maybe a NamedNode subclass that adds the possibility to define a name for each node that can be used in __repr__, debug tree output etc
<reinhard> no hurry though, I'm just hacking around a little :)
<reinhard> good night all
*** reinhard has quit IRC
<SachaS> http://www.debianhelp.co.uk/sugarcrm.htm sugarcrm on debian
*** lupo__ has quit IRC
*** jamest has quit IRC
*** sandroid has quit IRC
*** kilo has left #gnuenterprise
*** jamest has joined #gnuenterprise
*** jcater has quit IRC
*** sjc has quit IRC
*** nickr has quit IRC
*** nickr has joined #gnuenterprise
*** nickr has quit IRC
*** nickr has joined #gnuenterprise
*** derek has quit IRC
*** nickr has quit IRC
*** nickr has joined #gnuenterprise
*** jamest has quit IRC
*** sacha has joined #gnuenterprise
*** bigbrother has joined #gnuenterprise
*** bigbrother_ has quit IRC