*** jcater has quit IRC
*** Morphous has joined #gnuenterprise
*** Amorphous has quit IRC
*** dimas has joined #gnuenterprise
*** johannesV has joined #gnuenterprise
*** derek has quit IRC
*** dimas has quit IRC
*** Vee has quit IRC
*** Vee2d2 has quit IRC
*** icltlfatppl has quit IRC
*** dimas has joined #gnuenterprise
*** derek has joined #gnuenterprise
*** Vee2d2 has joined #gnuenterprise
*** Vee has joined #gnuenterprise
*** icltlfatppl has joined #gnuenterprise
*** LuftHans has joined #gnuenterprise
*** dcmwai has quit IRC
*** kilo has joined #gnuenterprise
*** LuftHans has quit IRC
*** dcmwai has joined #gnuenterprise
*** holycow has quit IRC
*** holycow has joined #gnuenterprise
*** dcmwai_ has joined #gnuenterprise
*** dcmwai has quit IRC
*** dcmwai_ has quit IRC
*** dcmwai has joined #gnuenterprise
*** holycow has quit IRC
*** johannesV has quit IRC
*** btami has joined #gnuenterprise
*** reinhard has joined #gnuenterprise
*** johannesV has joined #gnuenterprise
*** kilo has quit IRC
*** kilo has joined #gnuenterprise
*** dcmwai has quit IRC
*** hda has joined #gnuenterprise
<kilo> SachaS: hi
<SachaS> hi kilo
<SachaS> how are you?
<kilo> fine thx
<kilo> check Wiki please, in the Manuals section I put your conversation with holycow
<SachaS> :)
<SachaS> ok. will check.
<kilo> heavily edited...
<dimas> hi all
<dimas> where is the wiki?
<SachaS> dimas: check on www.gnuenterprise.org -> community wiki
<SachaS> kilo: looks good to me.
<dimas> SachaS: thanks
<kilo> http://www.gnuenterprise.org/cgi-bin/twiki/view
<SachaS> kilo: we might have to think about the proper (secure) settings of the postgresql database.
<kilo> yes, that is thin ice, setting up security like that...
<kilo> most people will leave it as it is
<kilo> and forget about setting it back to secure mode
<SachaS> yeah like me ;)
<kilo> hehe
<kilo> well i think i should give that wiki a real test
<kilo> eg set up postgresql on a machine following the directions in it
<kilo> or someone else up for the test? 8-)
<SachaS> good idea kilo ;)
<reinhard> kilo: FYI
<reinhard> When you start the sample.gfd form and you get errors that end with DB000: TypeError: __init__() takes exactly 2 arguments (1 given) 
<reinhard> i fixed that yesterday :)
<kilo> and what does it tell you then?
<reinhard> User 'foo' does not exist
<reinhard> or
<reinhard> Invalid password for user 'foo'.
<kilo> wow
<reinhard> and it does not in the form of a traceback
<reinhard> but like for other db connections, it shows the message in the login box
<kilo> unnatural, non-gnue-standard error messages???
<reinhard> hehe
*** dsmith has joined #gnuenterprise
<kilo> I suggest SachaS and holycow replay the whole scene, I listen closely and rewrite the whole wiki 8-))
<kilo> with updated messages...
<reinhard> lol
<reinhard> nevermind, i think with the new messages people will figure out what's wrong anyway
<kilo> reinhard: two country names are still non-ascii in the sample database
<kilo> Reunion and Ivory coast
*** dsmith has left #gnuenterprise
<reinhard> and in case someone uses an older snapshot, it's better to leave the warning there
<reinhard> kilo: yes, they actually should serve as a test if the database encoding is handeled correctly
<reinhard> at least that was my intention
<reinhard> are there problems with latin-2 ?
*** clad has joined #gnuenterprise
<clad> hi guys
<btami> reinhard: can you try the connection.gfd's Test connection button with appserver?
<reinhard> hi clad
<kilo> please put the info regarding latin-2 in the sample docs then
<reinhard> btami: yes, tried yesterday
<btami> i'v updated it yesterday
<reinhard> ah oh
<btami> but fails with appserver
<clad> Someone could help me about gnue designer, gnue reports??
* reinhard does svn up now
<clad> i don't know how to use it
<reinhard> kilo: what is the problem that occurs?
<kilo> clad: ask and will try to answer
<btami> reinhard, if you doesn't spec encoding, sample.gfd fails 
<clad> i would like to create a report using postgresql data
<clad> how can i do it?
<kilo> i took a sample report and manually edited it, but!
<kilo> there is a wizard in designer, from New chose wizard then simple report and there you go
<SachaS> reinhard: do you still have the bug, when designer connects to appserver it does not list any business classes, on your list?
<reinhard> SachaS: yes, we have not yet fixed that
<SachaS> ok.
<reinhard> btami: ah, thanks
<reinhard> btami: ok, so the problem is that the makefile doesn't specify an encoding, i see
<btami> yes
<clad> in fact i have a problem with gnue-designer, because its not running and it give me an error message : gizmo error
<kilo> clad: are you on suse 8.2? I tried it also and havent had any success
<kilo> something went wrong during the gnue setup
<clad> yes i'm working on suse8.2
<clad> yes
<kilo> wxpython or something else, dont remember exactly
<clad> yes
<kilo> i gave up
<clad> what can id do!!
<kilo> upgrade?
<kilo> you can ran reports or common without wx
<kilo> s/ran/run
<kilo> but it requires to hand-edit report code
<kilo> that is plain xml so shouldn't be a big problem however
<clad> i don't understand, i have suse 8.2, which solution i have?
<clad> allo
<clad> kilo>??
<kilo> are all prerequisits installed? python, wx, and so on
<clad> i think
<clad> when i launch ./gnue-designer
<clad> i have ImportError: No modul name Gizmo
<kilo> havent you tried to download RPM from wxpython download page?
<clad> yes i do
<btami> bbl
*** btami has quit IRC
<kilo> and did it install?
<kilo> what is the version of wxpython
<clad> i try to install rpm wxPythonGTK2.2 and i have also an error
<clad> another kind of error
<kilo> well, I did give up installing on suse 8.2 earlier
<kilo> you should consider upgrading to 9.0
<kilo> or debian
<kilo> seems like gizmo that Jason uses in designer is not in suse 8.2
*** lekma has left #gnuenterprise
<clad> ypu know where can i find an update of suse 8.2->9.0?
<kilo> well here severalcomputer magazines included it on their dvds
<clad> you don't have it?!:)
<kilo> i have it on cd, but havent installed it yet
<clad> could you send me it? or give me an access to download it on ftp server, it's verys important for me
<reinhard> clad: sorry to ask this stupid question
<reinhard> but can't you download from www.suse.com or like that?
<clad> i have to buy it you know!!
<clad> and i would like to try it before to buy it
<reinhard> you can't install suse from the net??
* reinhard knows why he likes debian :)
<clad> no
*** jamest has joined #gnuenterprise
<kilo> clad: got to go now, will be back in the evening
*** kilo has quit IRC
<clad> ok
<clad> thank u
* johannesV rebooting 
*** johannesV has quit IRC
*** johannesV has joined #gnuenterprise
<jamest> anyone know the state of mysql support
<jamest> got an email suggesting it was broken but I thought all mysql issues were cleared up in svn
<reinhard> johannesV: don't you have a mysql installed?
<jamest> also the only issues left in email are the dropdown sample reinhard set me (which I think he fixed right after he sent it :), and imports in gnue-forms
<jamest> which IIRC were going to be "fun" to fix
*** lekma has joined #gnuenterprise
<reinhard> jamest: what about the dropdown event passing weirdness
<reinhard> ?
<jamest> sigh, i forgot about that, with the mouse right?
<reinhard> yep
<johannesV> reinhard: yes
<jamest> i thought we were so close
<reinhard> johannesV: could you test if mysql works or not?
<johannesV> works with not
<reinhard> johannesV: parse error
<johannesV> reinhard: sample.gsd ?
<reinhard> johannesV: parse error == i didn't understand you :)
<reinhard> <johannesV> works with not
<reinhard> jamest: maybe you can paste here what you got in your mail
<reinhard> and johannesV can test if it is true
<reinhard> for current svn
<jamest> no details at all
<reinhard> oh
<reinhard> cool bug report
<reinhard> "doesn't work"
<reinhard> :)
<jamest> i can't even recall what issue was
<jamest> not even that much
<jamest> here
<jamest>  did the mysql work now or should i learn Postgres?
<jamest> there were mysql issues like a week or two ago when the ML sprung back to life
<jamest> but i purged all the mails on mysql which indicats to me that I considered the issue fixed
<jamest> was hoping we had a mysql user in here that could say "yep, things seem ok"
<johannesV> mysql:
<johannesV> DB000:   File "/home/johannes/prj/gnue/gnue-common/.cvsdevelbase/gnue/common/datasources/drivers/Base/ResultSet.py", line 307, in _cacheNextRecord
<johannesV> DB000:     rs = self._loadNextRecord()
<johannesV> DB000:   File "/home/johannes/prj/gnue/gnue-common/.cvsdevelbase/gnue/common/datasources/drivers/DBSIG2/ResultSet.py", line 72, in _loadNextRecord
<johannesV> DB000:     rsets = self._cursor.fetchmany()
<johannesV> DB000:   File "/usr/lib/python2.1/site-packages/MySQLdb/cursors.py", line 186, in fetchmany
<johannesV> DB000:     end = self._pos + size or self.arraysize
<johannesV> DB000: TypeError: unsupported operand types for +
<johannesV> using woody with python-mysqldb package
<jamest> thats special!
<johannesV> jamest: what's special ?
<jamest> that error :)
<johannesV> ok, found and fixed the error in cursor.py
<jamest> thank you
<johannesV> end = self._pos + (size or self.arraysize)
<johannesV> instead of
<johannesV> end = self._pos + size or self.arraysize
<jamest> some day i need to get a mysql 4 dummies howto
<jamest> as a typically get way too ticked off at mysql's security model setup and give up :)
*** dneighbo has joined #gnuenterprise
*** derek has quit IRC
<lekma> jamest: i still have this error whith checkboxes updating boolean in forms, would you help me track it down...
<reinhard> lekma: you have "useParameters" set in gnue.conf?
<lekma> useParameters = 1, yep
<johannesV> jamest: mysql is working with the latest commits
<jamest> johannesV: thank you
<reinhard> lekma: can you please run with debug level 1
<reinhard> and tell me the DBSIG2 Statement and DBSIG2 Parameters lines?
<lekma> svn up running... in a minute
<jamest> reinhard: you think this is a parameter issue
<reinhard> jamest: yes
<reinhard> probably
<reinhard> the sql generation stuff is completely new code when you useParameters
<jamest> ok, as iirc we had to cast booleans somewhere in common 
<reinhard> yes
<reinhard> actually i *did* that
<jamest> on the python side
<reinhard> at least i thought so
<reinhard> i cast to strings
<jamest> then do it twice, maybe it'll stick :)
<reinhard> '1' and '0'
<reinhard> lol
*** jcater has joined #gnuenterprise
<lekma> DB001: [Connection:263] DBSIG2 Statement: UPDATE t_lang SET modified_at=%(new_modified_at)s, modified_by=%(new_modified_by)s, lang_valid=%(new_lang_valid)s WHERE modified_by=%(old_modified_by)s AND created_at=%(old_created_at)s AND modified_at=%(old_modified_at)s AND created_by=%(old_created_by)s AND lang_id=%(old_lang_id)s AND lang_valid IS NULL
<lekma> DB001: [Connection:264] DBSIG2 Parameters: {'old_created_at': <datetime object at 0x403db0f0>, 'old_modified_by': 'malek', 'old_lang_id': 'DL', 'old_modified_at': <datetime object at 0x403db0e0>, 'new_modified_by': 'malek', 'new_modified_at': <datetime object at 0x403db0d0>, 'old_created_by': 'malek', 'new_lang_valid': 1}
<lekma> the boolean is lang_valid
<reinhard> lekma: thanks
<reinhard> what python version do you use?
<lekma> DB001: [GBaseApp:264] Python 2.3.3 (#1, Feb 13 2004, 15:28:56)
<lekma> DB001: [GBaseApp:264] [GCC 3.3.2 20031218 (Gentoo Linux 3.3.2-r5, propolice-3.3-7)]
<lekma> DB001: [GBaseApp:265] Run Options: [('--debug-level', '1')]
<lekma> DB001: [GBaseApp:266] Run Arguments: ['customizing/language.gfd']
<lekma> with useParameters=1, checkboxTrue = '1' and checkboxFalse = '0' in gnue.conf it's the same
<reinhard> lekma: thanks, i think i have what i need now
<reinhard> must boot my notebook where i have a python 2.3 installed
*** kilo has joined #gnuenterprise
<SachaS> bye (dinner)
*** SachaS has left #gnuenterprise
<jamest> reinhard: which section of gnue.conf does useParamters go in
<jamest> as i put in common and was about to report I don't see the boolean error
<jamest> but debug level 1 shows that sql is still not parameterized
<lekma> i put it in common
<reinhard> jamest: you have a typo in useParameters?
<reinhard> at least you had one here :)
<reinhard> it should be in common
<lekma> reinhard: btw i use psycopg as the db provider atm
<jamest> sigh, ignore me
<jamest> i was reading the postgresql error dump
<jamest> not the gdebug line
<jamest> it shows parameters there :)
<jamest> and the value is being passed in as an int, not a string
<reinhard> jamest: the problem is
<reinhard> checkboxes always send 0/1 (integers) to the dbdriver
<reinhard> they never send booleans
<reinhard> not even for python 2.3
<reinhard> (which IMHO isn't correct)
<reinhard> at least i wasn't aware of that
<reinhard> so i quoted int's only for python < 2.3
<reinhard> jamest, lekma: please svn up and try again
*** dcmwai has joined #gnuenterprise
<reinhard> and thanks for reporting :)
<lekma> reinhard: well it works for me, thank you very much for fixing it
<reinhard> lekma: thanks for reporting and testing
<reinhard> btw
<reinhard> i would really appreciate if useParameters was used
<reinhard> it would be good if we could make it default after the release
<reinhard> and completely remove the old way after the following release
<reinhard> (in case everything works ok)
<jamest> i've got it on at work now
<jamest> for me anyway
<reinhard> jamest: thanks
<jcater> fwiw
<jcater> I strongly disagree that a checkbox entry implies that boolean should be returned
<jcater> you could make the case that forms needs a boolean type for fields
<jcater> which might solve the problem
<jcater> but having an entry style determine/affect the returned datatype breaks our UI separation goals
<reinhard> erm
<reinhard> now every checkbox returns an integer
<reinhard> i don't see any basic difference if it implies an integer or if it implies a boolean
<jcater> I'd consider that broken then
<jcater> which we have always considered the checkbox implementation broken
<reinhard> ok :)
<reinhard> so the <entry> tag should have an attribute
<reinhard> trueValue= and falseValue=?
<jamest> i thought we set this up to always return a string '1' or '0'
<jamest> that way booleans worked even without the style checkbox 
<reinhard> jamest: unfortunately not
<reinhard> jamest: checkboxes always return ints
<jamest> the user just entered the T, f, 1, 0
<reinhard> 0 or 1
<reinhard> and common then has to typecast *all* ints into strings
<reinhard> because common doesn't know whether the int comes from a checkbox or from a real numeric entry
<reinhard> jcater: i'd agree that an option to have the checkbox return whatever we like would be best
<reinhard> but even then i'd think the default should be to return booleans
<reinhard> not ints
<reinhard> did i scare you all away?
<reinhard> hmmm
<reinhard> nice code snippet
<reinhard>           else:
<reinhard>             return 0
<reinhard>         return 1
<reinhard>       else:
<reinhard>         return 0
<reinhard>     else:
<reinhard>       return 0
<reinhard> :-)
<jamest> lol
<jamest> where is that?
<reinhard> in the pypgsql driver
<reinhard> as we were talking about duplicate code
<reinhard> there are some drivers that seem to throw exceptions on fetchmany() if there are no records left
<reinhard> instead of returning none
<reinhard> notable popy and pypgsql
<reinhard> these drivers have the complete _loadNextRecord duplicated in their code
<reinhard> would it be considered a good idea to have a
<reinhard> _broken_fetchmany = True
<reinhard> in the Connection class make the resultset ignore this exception
<reinhard> and therefore not having to subclass ResultSet at all?
<jcater> fine w/me
*** dcmwai has quit IRC
<jamest> reinhard: yip
*** sjc has joined #gnuenterprise
*** clad has quit IRC
<lekma> what is the exact syntax of form.setFeature()?
<lekma> form.setFeature('GUI:MENUBAR:SUPPRESS',True) doesn't seem to do much
<lekma> form.setFeature('GUI:MENUBAR:SUPPRESS',1) neither
<jamest> where you putting that?
<lekma> in on-activation form trigger
<jcater> I think that needs to be in on-startup
<jamest> those have to be in on-startup
<jamest> on-activation runs whenever the form becomes active
<jamest> which is after the ui is built
<lekma> ok
<jamest> the best place to see samples
<jamest> is gnue-forms/dialogs/*.gfd
<jamest> as they will all use that stuff 
<jamest> in fact it's a good example of using on-activation too
<jamest> :)
<lekma> i've seen those dialogs but there where no definitive explanation about where to put them
<lekma> btw does form.setStatusText() can't reside in on-activation?
<lekma> it throws me 
<lekma>   File "/home/malek/gnue-cvs/gnue/gnue-common/.cvsdevelbase/gnue/forms/uidrivers/_base/UIdriver.py", line 452, in updateStatusBar
<lekma>     if event.recordStatus != None:
<lekma> AttributeError: Event instance has no attribute 'recordStatus'
<jamest> odd, i can try to look at after lunch
<lekma> thx for your help guys...
<lekma> i must say it's quite difficult to grasp how everything works in GNUe
<lekma> all alone i would be lost
*** johannesV has quit IRC
*** mcDeke has joined #gnuenterprise
<mcDeke> any windows users here?
<mcDeke> having serious issues getting gnue forms to recognize mxDateTime
<mcDeke> not sure if syntax has changed or something is broken in the svn nightly
<mcDeke> running python 2.3 and mx 2.0.5 for python 2.3
<mcDeke> running gnue-forms.bat from the install instructions yields "This GNUe tool requires mxDateTime to be installed"
<mcDeke> however from a python shell
<mcDeke> i can do an import mx.DateTime
<mcDeke> and it works
<mcDeke> not sure if some path some where is dorked or if the call in the nightly snapshot of SVN is horked
<mcDeke> i am going to try to switch to the known releases, but am fearful that might make things even more muddy 
<jamest> no
<jamest> known release will not work on windows
<mcDeke> ?
<jamest> 0.5.3 is dead on windows
<mcDeke> hmm i have 0.5.3 working on a machine here
<mcDeke> might have "issues" but it works
<jamest> dropdowns?
<mcDeke> yeah
<jamest> odd
<mcDeke> at this piont the svn is dying horribly on me and im not sure why
<mcDeke> it can not find mx
<jamest> i was informed it was dead
<mcDeke> but mx is installed
<mcDeke> jamest: for true usability it IS dead
<mcDeke> but at this point i need to find out if mx is my problem or not
<mcDeke> if you can help me with path issues it would be great
<mcDeke> as im thinking its something silly 
<jamest> i can try to install on a win98 box in a bit
<mcDeke> or a version issue between mx
<mcDeke> and our code
<jamest> where did you get install docs?
<jamest> what version of python
<mcDeke> we are win200
<jamest> etc, etc
<mcDeke> python 2.3
<mcDeke> mx 2.0.5
<mcDeke> win2000
<mcDeke> hourly snapshot from late yesterday afternoon
<mcDeke> using the instructions from the WIKI
* mcDeke used these instructions with the other 0.5.3 stuff and everything installed
<mcDeke> it appears to have no failures, it just cant find mx
<mcDeke> what is crazy
<mcDeke> is installing forms checks for mx and FINDS it
<mcDeke> but running gnue-forms from c:\python23\scripts\ cant
<mcDeke> checking mxDateTime library
<mcDeke> is from the forms install
<mcDeke> checking wxPython library:  ok (2.4.2.4)
<mcDeke> as well
<jamest> i'll need to d/l everything to install
<jamest> so it's going to take a while
<mcDeke> okay i will check back in later
<mcDeke> have meeting to run to
*** mcDeke has left #gnuenterprise
<reinhard> mcDeke should try
<reinhard>  from mx.DateTime import DateTime, mktime
<reinhard> as that is exactly the line which is appearantly failing
<reinhard> maybe some versions of mx don't define mktime?
<jamest> (though I would prefer assembly or C, others have voted against them...)
<jamest> writing an ERP in assembly
* jamest shudders
<jcater> writing an ERP in c?
* jcater shudders at that thought too
<jcater> :)
<jamest> :)
*** dimas has quit IRC
<jamest> sigh
<jamest> latest mx datetime from windows installer goes in
<jamest> from mx.DateTime import DateTime,mktime
<kilo> I heard someone mention assembly or C 8-)))
<jamest> er, wait, that's what you said
<kilo> yes i know 8-))
<jamest> that was you ?
<nickr> its you!!!
* jamest thwaps kilo with a goat
* kilo assembles an assembly assembly
<kilo> what is the status of DCL?
*** SachaS has joined #gnuenterprise
<lekma> bye all, see you tomorrow (at least for me)
*** lekma has left #gnuenterprise
<jamest> kilo? it works
<kilo> Warning: pg_connect(): Unable to connect to PostgreSQL server: could not connect to server: Connection refused Is the server running on host "localhost" and accepting TCP/IP connections on port 5432? . in /usr/share/dcl/www/pgsql.php on line 74
<kilo> Template Error: filename: file ./templates//login.tpl does not exist.
<kilo> Halted.
<kilo> that's what i get
*** _florin_ has joined #gnuenterprise
<_florin_> hello
<SachaS> hi _florin_
<reinhard> hello _florin_
<reinhard> hello SachaS
<reinhard> SachaS: do you have any outstanding bugs?
<_florin_> hi SachaS, reinhard
<SachaS> hi reinhard: no outstanding bugs (as far as I know). but some feature requests :)
<SachaS> just back from dinner.
<jamest> SachaS: any chance you could upload that stuff into gnue-contrib?
<SachaS> jamest: which "stuff"? ;)
<SachaS> jamest: the documentation of it is incomplete :)
*** mcDeke has joined #gnuenterprise
<jamest> baby erp, it's still a great reference as is 
<SachaS> jamest: i saw the email: user are calling: we want code, we want code ;)
<jamest> exactly!
<jamest> don't make us come and take it!
<SachaS> :)
<jamest> er, where do you live anyway?
<jamest> :)
<kilo> in a tiny little country between the high mountains
<jamest> seriously, even some good file defs would be a plus
<jamest> as samples
<reinhard> jamest: i could manage that if need
<reinhard> (go and take it)
<reinhard> ;-)
<SachaS> hehe
<jamest>  /msg reinhard remember to use the softer bat, don't want to scare him off
<reinhard> lol
<mcDeke> jamest: were you able to look at windows at all
* mcDeke is grabbing the hourly and trying on a machine i have working currently with 0.5.3
<reinhard> <reinhard> mcDeke should try
<reinhard> <reinhard>  from mx.DateTime import DateTime, mktime
<reinhard> <reinhard> as that is exactly the line which is appearantly failing
<kilo> i am on win now
<mcDeke> reinhard: import mx.DateTime works
<mcDeke> i will try from mx.DateTime import DateTime, mktime
<reinhard> yes
<reinhard> maybe you have a version of mx.DateTime that doesn't define both these 2 symbols
<kilo> reinhard: second one works too for me
<jamest> mcDeke: i got mxdatetime installed
<jamest> that from line reinhard suggests works for me
<jamest> no farther yet
<mcDeke> from mx.DateTime import DateTime, mktime
<mcDeke> works on his machine
<mcDeke> in a python shell
<reinhard> next i would try is
<reinhard> in GFDisplayHandler (in gnue/forms)
<reinhard> remove the try: - except: block around the import line
<reinhard> so that you get the importError in form of a complete traceback
<reinhard> a thought i have been playing with is
<reinhard> if we catch exceptions and raise another one
<reinhard> like we catch some native exceptions in the dbdriver and raise a ConnectionError
<reinhard> to provide a property "detail" in these exception classes that contain the traceback of the cought exception
<reinhard> and for example in an error message in forms
<reinhard> have a "detail" button that displays that traceback
<reinhard> so we never lose the information where the actual exception happened and what kind it was
<mcDeke> welll batman
<mcDeke> good and bad news
<mcDeke> good news.. i think svn was hosed up
<mcDeke> bad news i dont know why
<mcDeke> in his copy its
<mcDeke> from mx.DateTime import DateTime, strptime
<mcDeke> NOT
<mcDeke> from mx.DateTime import DateTime, mktime
<mcDeke> and removing the error trap yields
<reinhard> mcDeke: that was about 2 days
<mcDeke> failure to import strptime
<reinhard> until we found out that strptime isn't available under windows
<reinhard> btami reported it
<mcDeke> reinhard: i believe he got the hourly.php snapshot from the website last night about 4:30pm Arizona Time
<reinhard> he must have taken a nightly snapshot 
<reinhard> err
<mcDeke> as long as its explainable we are all good
<reinhard> what?
* mcDeke just grabbed teh hourly and i will tell you what it has
<reinhard> then i think the hourly snapshot doesn't work
<reinhard> we fixed that several weeks ago
<mcDeke> from mx.DateTime import DateTime, strptime
<mcDeke> from http://www.gnuenterprise.org/downloads/hourly.php
<mcDeke> about 10 minutes ago
<mcDeke> good news is we have static ip addresses now
<mcDeke> so it should be very short time before hole is punched in firewall for us to get svn the normal way
<mcDeke> can someone update the hourly?
<mcDeke> so i can download adn test?
<reinhard> -rw-r--r--    1 jcater   jcater    6309582 Mar 20 13:05 gnue-svn-current.tar.gz
<mcDeke> check the .zip
<mcDeke> remember WINDOWS :)
<reinhard> -rw-r--r--    1 jcater   jcater    7958648 Mar 20 13:05 gnue-svn-current.zip
<reinhard> besides, current versions of winzip can unzip .tar.gz files
<mcDeke> well download and try for yourself
<mcDeke> i just downloaded 
<mcDeke> winzip is non free (tsk tsk)
<mcDeke> i think either
<mcDeke> 1. the hourlies are not pulling new copies of svn
<mcDeke> 2. i have some goofy cache happening
<mcDeke> 3. the change is still dorked in svn (i.e. its not fixed there yet)
<reinhard> mcDeke: 4. hourlies are not built any more
<reinhard> since Mar 20 appearantly
<reinhard> mcDeke: you have a *free* tool to unpack zip files under windows??
<mcDeke> yip that was my number 4 that this dumb thing did
<mcDeke> yes
<mcDeke> GPL
<mcDeke> http://www.info-zip.org/pub/infozip/
<mcDeke> actually looks like its BSD
<mcDeke> can someone run the process to build a nightly .zip?
* mcDeke can use ssh from here (yet) :(
<jamest> hmmmm
<jamest> i can find where the nightlies are created
<jamest> but not the hourlys
<mcDeke> i meant hourly but if a nightly that is full svn tree exists that woudl be okay too
<jamest> um
<jamest> we don't even list nightlys on the site anymore
<jamest> http://www.gnuenterprise.org/downloads/snapshots.php
<mcDeke> http://www.gnuenterprise.org/cgi-bin/twiki/view/Main/WindowsQuickStart#Install_GNUe_SVN_Snapshot_on_Win
<jamest> i imagine building these newer sdist releases every hour would be asking a bit much of poor old ash
<mcDeke> step 4
<jamest> ah, it's in the wiki, looks like something you can make notes on and fix
<jamest> :)
<jcater> I took hourlies off
<jcater> svn wasn't up to it
<jcater> (found it it was what was locking up svn)
<jcater> s/it//
<mcDeke> jamest: fixing wiki now
<kilo> sigh
<_florin_> does anybody have the time to give me a brief introduction on how to implement trigger support in a widget?
<mcDeke> just wanted to let you know that i actually pointed staff to gnuenterprise.org and told them to install
<mcDeke> to see what kind of problems our users normally bump into
<mcDeke> it actually all went smooth except the datetime piece 
<mcDeke> documentation updated
<mcDeke> also there is no mention of configuring gnue reports on the wiki how-to
<mcDeke> have had duane pencil those in and will add them later
<mcDeke> for now must really grab food :)
<mcDeke> sachas can you post to the net your gpd files for baby erp?
*** btami has joined #gnuenterprise
*** ogger has joined #gnuenterprise
<reinhard> jcater/jamest: as it seems, there is nothing to fix before the release
<reinhard> (except maybe the mouse-dropdown issue)
<SachaS> mcDeke: wait one minute.
<reinhard> not even SachaS knows of any outstanding bug
<jamest> wow, not even SachaS
<reinhard> so it would be the perfect time to release
<reinhard> ;-)
<btami> reinhard: the sap dbdriver is broken
<jamest> i'd like to get drowdowns fixed if at all possible
<reinhard> btami: seriously?
<jamest> imports too but that's a pipe dream
<reinhard> btami: i have no means of testing
<reinhard> err wait
<btami> i'v just tried the plugin.py
<reinhard> i think maxdb is gpl now, isn't it?
<jamest> i was going to see if we wanted to do a freeze and let translators do their thing
<reinhard> yes
<btami> sap=maxdb is gpl
<reinhard> if you look at the nightly build logs there are still some broken translatable strings
<jamest> so maybe a day or two of bug stomping w/ no new features
<reinhard> i just wrote johannes a mail he should fix them tomorrow
<jamest> and how about some, um, 
<jamest> damn, what do you call them
<jamest> begins with a D
<jamest> ducks
<jamest> no, no,
<jamest> docs
<reinhard> rofl
<jamest> yeah!  that's it!  how about some docs!
<reinhard> well i'd say
<btami> reinhard: just run python plugin.py
<reinhard> let's do the last fixes
<reinhard> jamest: let's do the last fixes within, say, 24 hrs
<reinhard> then do a prerelease tomorrow about this time
<jamest> i can try
<reinhard> then do translations and docs
<reinhard> and then release 2 days later or so
<reinhard> would be holy friday
<reinhard> or saturday
<reinhard> btami: you probably have to gcvs python.py
<reinhard> err
<reinhard> gcvs plugin.py
<jamest> ok
<jamest> so my outstandings are : dropdowns and import
<jamest> s/outstandings/outstanding issues
<jamest> i can try and take a look at dropdowns tonight
<reinhard> btami: i get
<reinhard> sapdb.sapdb: DependencyError: Module 'sapdbapi' is not installed.
<btami> ok
<reinhard> which is correct for me :)
<jamest> what else is out there?
<btami> no
<reinhard> jamest: i know of nothing
<btami> install sapdb's python modules and try again :)
<reinhard> jamest: except the weridness of a single form not working for PeterD but for everybody else
<jamest> ?
<reinhard> classrepository.gfd
<reinhard> i got me some mandrake cd's to try
<jamest> ah
<reinhard> but didn't get to it yet
<jamest> i've got 10 installed
<reinhard> really?
<reinhard> well
<reinhard> you've got appserver working there?
<jamest> yes
<reinhard> seriously?
<jamest> suse9 , fedora, sid, mandrake 10
<jamest> nope
<jamest> i don't have gnue working there
<reinhard> ah
<reinhard> ok
<reinhard> nevermind
<jamest> yes, seriously
<reinhard> i'll try one of these days
<jamest> i planned to do gnue install tonight
<jamest> mandrake and my system had issues i believe are now resolved
* jamest has 240GB of HD space for OS installs
<jamest> for testing, as long as it's freely available i'll add it to my list and do a quick run thru with it
<reinhard> jamest: cool
<reinhard> jamest: although he has no problems except with a single form
<reinhard> with a form that only works with appserver
<reinhard> btami: any chance you can save me from having to install and configure sapdb?
<btami> reinhard: ftp://ftp.sap.com/pub/sapdb/bin/linux/sapdb-python-linux-i386-7.4.03.31a.tgz
<reinhard> IOW
<btami> i have nosapdb installed too
<reinhard> could you give me the error/traceback/whatever?
<reinhard> ah oh
<reinhard> how do you know it's broken then?
<btami> just the gnue sapdb driver loading fails
<reinhard> can you give me the error message?
<btami> plugin.py man!
<btami> it's your pupy
<btami> :)
<btami> wait a min...
<reinhard>     connections = GConnections (r'D:\Python22\etc\connections.conf')
<reinhard> i wonder who wrote that...
<btami> not me :)
<reinhard> i figured that much
<SachaS> [Announce] babyerp is a new gnue-contrib project (licenced under GPL). project is uploaded. announcement email is on the way.
<btami> sapdb.sapdb: NameError: name 'DataObject_Object' is not defined
<SachaS> mcDeke: you find all babyerp files in gnue-contrib.
<btami> reinhard: i think it's half converted by jc :)
<reinhard> btami: just compare RecordSet.py to DataObject.py
<reinhard> and you see what's up with the sapdb driver...
<btami> hmm, yes
<reinhard> i would be curious how many of our dbdriver actually work
<btami> and how many was used ever...
<reinhard> thank god python is an interpreter language otherwise some of them would probably not even compile ;-)
<reinhard> btami: right
<reinhard> seriously i'd say we should "remove" the drivers we aren't able to maintain
<reinhard> "remove" == put into some branch in svn
<btami> btw. this plugin.py is wery useful to check if installed dbdrivers are OK or not
<jamest> reinhard: it's only my list
<jamest> i've got 45GB set aside for db installs
<jamest> :)
<reinhard> jamest: :)
<jamest> i imagine it'll basically be bare bones testing
<reinhard> btami: yes, i think there could be an option in GBaseApp
<reinhard> --list-dbdrivers
<btami> or --list-installed-dbdrivers
<kilo> must be there. i remember translating it today...
<SachaS> reinhard: I get an exception. in a commit() call.
<SachaS>   File "/home/sacha/gnuenterprise/gnue/gnue-common/.cvsdevelbase/gnue/common/datasources/drivers/appserver/appserver/Connection.py", line 100, in commit
<SachaS>     self._sm.commit (self._sess_id)
<SachaS>   File "/home/sacha/gnuenterprise/gnue/gnue-common/.cvsdevelbase/gnue/common/rpc/drivers/Base.py", line 232, in __call__
<SachaS>     return self._adapter._runMethod (self._methodname, *args, **params)
<SachaS>   File "/home/sacha/gnuenterprise/gnue/gnue-common/.cvsdevelbase/gnue/common/rpc/client.py", line 115, in __init__
<SachaS>     Error.__init__ (self, info [1])
<SachaS> IndexError: list index out of range
<jcater> --help-connections
<jcater> (anyone willing to fill in that TODO part is welcome to :)
<btami> jcater: you know the designer bug that editing only a trigger code in trigger editor,
<btami> then save doesn't save the new code
<btami> but if you modify another object, it's OK with new triggre code saving
<jamest> btami: i fixed that last week IIRC
<btami> have you committed it?
<jamest> i hope so
* jamest always gets burned by svn's reversal of log entry/password prompt
<jamest> sigh
<jamest> no
<jamest> i guess not
<jamest> i'll have to commit tonight when I get home
<btami> ok, thx
<jamest> it'll probably be close to 2 hours before I can commit 
<reinhard> SachaS: fixed the exception handling
<jamest> i have a hair cut after work
<reinhard> the server side exception should now be visible correctly
<reinhard> can you please try again?
<jamest> also have a fix for named trigger additions from the popup menu to commit it seems
<jamest> sigh
<jamest> i suck :)
<SachaS> reinhard: will try again.
<SachaS> reinhard: of course still a traceback, just a longer one ;)
<SachaS> reinhard: i think i see the problem
<SachaS> in a procedure I do:
<SachaS> create object 1
<SachaS> then create object 2
<SachaS> object 2 has a reference to object 1
<SachaS> when i call commit
<SachaS> it looks like it tries to create object 2 first, which fails as there is no object 1.
<reinhard> SachaS: so you get an error from the postgresql database?
<SachaS> probably yes. saying that there is a violation of foreign keys
<reinhard> yes that's from the db
<reinhard> hmm
<SachaS> seems like the order of creating objects ... is reverse? or random?
<SachaS> creating objects -> inserting into db tables.
<reinhard> random
<SachaS> actually committing objects.
<reinhard> hmmm
<SachaS> what? random ;)
<reinhard> i think it's not possible for appserver to determine the correct order of inserts
<SachaS> oops
<reinhard> a = new ('aclass')
<reinhard> b = new ('bclass')
<reinhard> a.foo = b
<reinhard> b.foo = a
<SachaS> bang!
<reinhard> now what should appserver do?
<reinhard> :)
<SachaS> shutdown?
<SachaS> ;)
<reinhard> rofl
<SachaS> my procedure might be bad practice ...
<reinhard> actually i'm surprised that postgres check referential integrity on insert and not on commit
<reinhard> SachaS: no
<reinhard> SachaS: IMHO even the above code should work
<SachaS> if .foo is null, it should be ok.
<SachaS> foo in both classes.
<reinhard> will have to think about this one for a bit
<reinhard> at least it made you find the exception handling bug :)
<SachaS> ;)
<reinhard> SachaS: just added your problem to BUGS
<reinhard> IOW
<reinhard> i don't think i'll be able to fix this quickly
*** reinhard has quit IRC
<SachaS> dah. OK.
*** reinhard has joined #gnuenterprise
<SachaS> welcome back.
<reinhard> <reinhard> SachaS: just added your problem to BUGS
<reinhard> <reinhard> IOW
<reinhard> <reinhard> i don't think i'll be able to fix this quickly
<reinhard> <reinhard> :(
<reinhard> you think you can work around it for the time being?
<SachaS> would it be possible, to create them in the sequence as they are created (new())?
<SachaS> i could with setting the property to null, right?
<SachaS> of course i can work around ;)
<reinhard> i don't think creating them in the sequence they are created would be right
<reinhard> it might be quite usual for a procedure
<reinhard> to create a master and several details
<reinhard> err skip that
<reinhard> you might be right :)
<reinhard> but i'd like it to not be dependent on the order you create them
<SachaS> as a first work around ;)
<SachaS> but yeah I agree
<SachaS> i see the problem.
*** jamest has quit IRC
<reinhard> actually
<reinhard> you even have the same problem with deletes
<reinhard> you also have to do the deletes in the correct order
<SachaS> true
<reinhard> off to bed for today
<reinhard> night all
*** reinhard has quit IRC
*** btami has quit IRC
*** kilo has quit IRC
*** jcater has quit IRC
*** ogger has quit IRC
*** jcater has joined #gnuenterprise
<chillywilly> hi ppl
<chillywilly> jason
<chillywilly> whaddup dawg?
<jcater> word
<chillywilly> up?
<mcDeke> hey all
<mcDeke> just an FYI
<mcDeke> win32 sucks arse
<chillywilly> haha
<jcater> that's w32 or windows, fewl
<mcDeke> bad and mean things should be done to those that make people endure it
* chillywilly points and mcDeke..<-- ha ha haaa haa haaaaaaaaaaaaaaa
<mcDeke> jcater do you know if you or jamest is using an auth plugin?
* mcDeke is getting ready to get head on into gnue app development of a grand scale
<mcDeke> and would you be opposed to me checking for operating system
<mcDeke> on login
<mcDeke> and if win os
<mcDeke> then attempting to grab domain login name
<mcDeke> and if it doesnt exist attempt to grab local machine login name
* mcDeke notes i would make this a gnue.conf option
<chillywilly> vehemently opposed
<mcDeke> turned off by default
<jcater> as long as an option, I don't care
* mcDeke is looking for documentation on how to obtain active directory credentials as a passthru
<chillywilly> ugh
*** nickr has quit IRC
*** mcDeke has left #gnuenterprise
*** sjc has quit IRC
*** nickr has joined #gnuenterprise
*** SachaS has quit IRC
*** jamest has joined #gnuenterprise
<jamest> anyone know what OS kilo runs?
<chillywilly> jo momma
<jcater> is that a BSD variant?
<chillywilly> yea mon
<jcater> yo mamma is like netbsd
<jcater> she'll work with any hardware you throw at her
<jcater> ba da bing
<jcater> yo momma is like Win98
<jcater> she goes down all the time
<jcater> wooo hooo
<jcater> I'll be here all week
* chillywilly holds his lighter in the air
<chillywilly> jcater!
<chillywilly> jcater!
*** icltlfatppl has quit IRC
<jcater> jamest: was he a suse person?
<jcater> or a gentoo
<jcater> we've had one of each lately, and I can't remember who's who
<chillywilly> gentoo == people who have too much time on their hands ;)
<jcater> ja
*** holycow has joined #gnuenterprise
<chillywilly> holycow!
<chillywilly> :-D
<holycow> oi :)
*** Vee has quit IRC
*** jamest has quit IRC
*** holycow has quit IRC
*** dcmwai has joined #gnuenterprise
<dcmwai> anyone here?
*** roche has joined #gnuenterprise
*** jcater has quit IRC