*** rocky__ has quit IRC *** sjc has joined #gnuenterprise *** SachaZzzz has quit IRC *** holycow has quit IRC *** qman has joined #gnuenterprise *** reinhard has joined #gnuenterprise *** qman has quit IRC *** btami has joined #gnuenterprise *** holycow has joined #gnuenterprise *** johannesV has joined #gnuenterprise *** jcater has joined #gnuenterprise *** bigbrother has joined #gnuenterprise *** bigbrother has joined #gnuenterprise *** nickr has joined #gnuenterprise *** kilo has joined #gnuenterprise good morning *** btami has quit IRC *** btami has joined #gnuenterprise good morning kilo hi johannesV ah, btw, can you tell me something about this new auth system? it's not a new system; it's *no* system it just eliminats authentication completely, since it's not very usefull atm hmm, i like it 8-))) but one can still keep as it was before, using 'authentication = True' in gnue.conf's appserver section well to be a 'real' authentication system it would need much more than only checking some classnames but we plan to add *better* system in a near future until then we do it this way; btw. this means gnue_useraccess is *optional* now btw i used to use "gsscvs useraccess.gsd" to add hacker etx to gnue_useraccess but it doesnt work now what has changed? s/etx/etc i don't know useraccess.gsd, but there shouldn't be something different on db-backend side the clue is, appserver does *not* check gnue_useraccess anymore hmm, ok, then it can be forgot... *** lekma has joined #gnuenterprise johannesV: when gsscvs creates a db (--createdb), does it use the encoding setting from connections.conf? i'm not sure which db backend ? interbase hmm, the actual creation is done by kinterbasedb module our db-driver is using that module's create_db () function which takes no 'encoding' parameter so i guess it's using a 'default' encoding maybe btami knows more about that ... well he is looking at it right now brb *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** Amorphous has joined #gnuenterprise *** Amorphous has left #gnuenterprise johannesV: am i right that postgresql sets its default encoding at install time? kilo: yes that's right but i think one could superseed this on creation of a new database does it? because interbase does kinterbasdb lets us define the encoding of the db to-be-created in create_database *** wt has joined #gnuenterprise so it could be used to do this but consulting with btami we think that it should be an extra parameter to gsscvs when --createdb is used not read from connections.conf or introduce clientencoding(=encoding) and dbencoding in connections.conf so that one can create a db say in utf8 and define different accesses to it with different say chinese settings anyhow, some solution for non us is needed when creating new db *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has left #gnuenterprise *** reinhard has joined #gnuenterprise btami: are you using win32 wiy=th gnue s/wiy=th/with ? sometimes most of time i'm using uhu-linux i would like to setup a svn version on win2k hawe you read wiki? the setup-cvs-win.py script ask for utilities that i can't seem to find not yet hmm but i'm goiing to check it now setup-cvs-win.py was an attempt by arturas to use links on ntfs but it needs some non free utils from ms *** reinhard has left #gnuenterprise *** reinhard has joined #gnuenterprise so i just need to run setup-cvs.py? i'm suggesting to use python setup.py install for all gnue-* source packages setup-cvs.py is for linux only ok thx bbl *** btami has quit IRC *** sjc has quit IRC *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has left #gnuenterprise *** reinhard has joined #gnuenterprise *** lekma has quit IRC *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** reinhard has left #gnuenterprise *** reinhard has joined #gnuenterprise *** SachaS has joined #gnuenterprise hi everyon hi everyone *** holycow has quit IRC *** cilkay has quit IRC *** Vee has quit IRC *** mdean has quit IRC *** chillywilly has quit IRC *** havoc has quit IRC *** andrewm has quit IRC *** holycow has joined #gnuenterprise *** cilkay has joined #gnuenterprise *** Vee has joined #gnuenterprise *** mdean has joined #gnuenterprise *** chillywilly has joined #gnuenterprise *** andrewm has joined #gnuenterprise *** havoc has joined #gnuenterprise *** reinhard has left #gnuenterprise hi SachaS *** reinhard has joined #gnuenterprise hi kilo how are you? having lunch, feeling high... *** jfontan has left #gnuenterprise cool: knock is a server/client set that implements the idea known as port-knocking. Port-knocking is a method of accessing a backdoor to your firewall through a special sequence of port hits. This can be useful for opening up temporary holes in a restrictive firewall for SSH access or similar. reinhard: do you plan to implement automatic created/changed by/when fields in appserver and if yes, when? kilo, yes we do so it will be added right after the next release :) when is the clustered version available? j/k which will happen at end of week (but reinhard decides on that) oh, no, then i have to translate po files again... kilo, maybe you talk to reinhard about timing then ... :) johannesV: when will the appserver version for Z80 proc be ready? 8-)) what is "Z80" ? i am sure you know you mean the cpu ? a nice modern system depends when you can get python ported yes, yes :) good old days when i knew hex codes of MOVA and JMP... johannesV: I will release as soon as you think our code is ready for release reinhard, which code do you mean ? exception stuff seems ok for me the few items we talked about monday is all of that fixed? no, rearanging samples dir is missing *** lekma has joined #gnuenterprise *** kilo has quit IRC *** gsoti_away has joined #gnuenterprise *** hiolo has joined #gnuenterprise kilo, dont forget the gold nugget currency in the base currency package ;) *** holy_cow has joined #gnuenterprise *** wt has quit IRC reinhard: i hate to bug you again but i am fairly new to python and i don't get the find() to work in an external module imported in an OnValidate * SachaS was playing for 1/2 hour with live compiere demo system can you help, please *** agx has joined #gnuenterprise lekma: do you get to work to access properties in a OnValidate? thats where i had issues reinhard: here i am agx: welcome again eg you can access properties of that external modules but not of the underlying module lekma: you have to pass a function pointer as parameter to the function lekma: have not played with it but I had issues with accessing properties def myfunction (param1, param2, find): .... reinhard: ok, my problem is that any existing package is difficult to use in Italy since we have really tons of stupid laws about our business that change very frequently :-( I've seen for example that a package like compiere.org should be heavily modified in order to be used by italians. *** reed has joined #gnuenterprise SachaS: yeah i can (or i don't understand the question) i think most countries have stupid laws :) hi reed :) hei reinhard :) reinhard: just lurking, no worries :) reed: S.M.? agx: if you are willing to invest a little time into gnue reinhard: not gnue but ERP packages developed with GNUe agx: you would have the chance to influence the development of the tools err of the packages of course like i have started an accounting package a week ago or two reinhard: well but i see a really big problem, and i'm investigating it more, what will work for italy cannot work for other country :-( if you follow that development you can tell what we would have to change to make it work for italy hmmm i mean, phonebook can be a common module but the payment of employers is really different, in Italy it's a mess :-( payroll is a bad bad thing it's different in *every* country I guess btami and kilo are working on it reinhard: change? everything i suppose :-) it's not a matter of adding attributes/fields is really a different approache... something like "non linear" that's the problem and looking for people that can explain how payroll works in different countries and please don't underestimate the silliness of laws in countries outside italy ;-) reinhard: and in the OnValidte i pass find() as the param find? like myfunction(self.property1, self.property2, find())? no lekma: without () just ... self.propert2, find) reinhard: OK :-) with find() you call the function reed: Stefano? with find you pass a function pointer agx: yep, again lurking :) hmm reed: we'll talk about this in Milan next week, i promise, i would like to anticipate something to Reinhard started a db recovery some 15 minutes ago, still running reinhard: basically, i've find a company that can sponsorship some money (enough to pay some developer for 2 year) but i've very strict requirements to satisfy ah reinhard: a) no Java b) Libre Software licensing and development c) 3 tier d) multi channel (accessible via native client, web interface and text/IVR) reinhard: it's seems to me that GNUe satify this, it also have windows binarie for who want to continue to use stange OS :-) reinhard: ok thx what is IVR? reinhard: i'll try more GNUe during those weeks, if everything is OK and the project can be also be supported officially by GNUe i have to start finding other italian companies that want to join this adventure. I think i need to coordinate almost 50 developer to have something that can be go into production into 2 year :-) reinhard: IVR ==> asterisk or bayone; you call a phone number and a voice ask you a choice and to press # what do you mean with "supported officially by GNUe" ? reinhard: to be listed under "project" (if i rembember the current web site structure) ah *** btami has joined #gnuenterprise this will surely be no problem because we list all contributors *** johannesV_ has joined #gnuenterprise agx: btami would be the one to talk to about payroll reinhard: wonderfull :-) so i have to start making some serious test reinhard: basically i want to try to develop a phonebook my self to see how it works (a menu, a form with a grid, some reports and so on) reinhard: there is a such example available? yes, in gnue appserver's sample directory hi all hi btami agx: i've also started to write a documentation walking the reader through an example, but hi reinhard sigh sigh life didn't agree with that plan so far ;-) jcater: you here? jcater: i have problems with svn jcater: started svnadm recover /var/svn/gnue and it runs *way* too long now reinhard: can you send it to me? doesn't matter if it's incomplete; i didn't know python yet but i've almost used amost any develoeer tool around :-) reinhard: the only problem now it that i'm using Debian packages, should i use CVS? Let's start with a simple address management application. For now, we will assume that gnue-appserver has been installed properly and is already running, that gnue-forms and gnue-reports are also installed, and that your connections.conf file is set up correctly to allow connections to the running application server with the [appserver] connection. --- that's all i have so far :(( agx: yes you should use development version which would be svn in our case :) reinhard: i get errors when want to set up a clean gnue.fbd with setup-db on XP c:\python23\scripts>python gnue-schema --connection=gnue sample.gsd 'sample.gsd' gsd fßjl bet÷ltÚse... SÚma frissÝtÚse ... Adatok frissÝtÚse ... 'address_country' tßbla frissÝtÚse ... (-551, 'isc_dsql_prepare: no permission for read/select access to TABLE ADDRESS_COUNTRY. ') maybe the samples dir needs som TLC this seems to be a permission problem in the db? yes, but it worked before without errors hmmm i can't remember changing anything that would have such an effect how long ago is "before"? btw, there is no base-*.gld too reinhard: ok, send it please, in the meanwhile i'll start svn-ing down the source :-P i was testing it at the latest appserver release reinhard: i get: RemoteError: 'dict' object has no attribute '_children' when i test with find in external module agx: svn is doing a database recovery at the moment you will have to wait reed: it's all for now, i've go to take a coffè, if you have news or answers drops me a quick phone call! reinhard: http download of snapshots? agx: not sure if they are online please look at www.gnue.org lekma: will try it myself and see if i can reproduce reinhard: any UML of the ERP packages? didn't find them ok agx: as we have no ERP packages yet, we also don't have UML of them ;-) it's a simple find that search another class than the the calling class there is some .dia there... reinhard: you develop a big ERP project without UML? :-P reinhard: no snapshots available :-( ok i'll wait reinhard: here is the complete "bugreport" Microsoft Windows XP [verziószám: 5.1.2600] (C) Copyright 1985-2001 Microsoft Corp. c:\python23\scripts>setup-db.cmd c:\python23\scripts>python gnue-schema --connection=gnue --createdb base.gsd 'base.gsd' gsd fßjl bet÷ltÚse... ************************************************************ Attempting to log into "Security Database" (as SYSDBA): Password: ************************************************************ SÚma frissÝtÚse ... Adatok frissÝtÚse ... 'gnue_module' tßbla frissÝtÚse ... beszÿrva 1 sor, frissÝtve 0 sor 'gnue_class' tßbla frissÝtÚse ... beszÿr grr, i will send you in a mail instead btami: i got error messages for encoding problems yes please send mail *** johannesV has quit IRC reinhard: i'll tank you very much you're really kind! I'll go back to you via email when i'll have any news. Thanks you very much again. C U, good evening. *** agx has left #gnuenterprise i go too, bye *** reed has left #gnuenterprise reinhard: sent jcater: you here? * btami reboots to try setu-db on linux too *** btami has quit IRC *** jamest has joined #gnuenterprise *** btami has joined #gnuenterprise jamest: i was just trying to do a db recovery however it hung (> 1 hour) so i just killed and restarted it and it seems to hang again reinhard: with setup-cvs-db.sh i get tamas@beus:~/data$ ./setup-cvs-db.sh 'base.gsd' gsd fájl betöltése... ************************************************************ Attempting to log into "Security Database" (as SYSDBA): Password: ************************************************************ Updating schema ... Updating data ... updating table 'gnue_module' ... Inserted 1 row(s), updated 0 row(s) updating table 'gnue_class' ... Inserted 7 row(s), updated 0 row(s) updating table 'gnue_property' ... Inserted 49 row(s), updated 0 row(s) updating table 'gnue_procedure' ... Inserted 1 row(s), updated 0 row(s) updating table 'gnue_parameter' ... Inserted 4 row(s), updated 0 row(s) /usr/bin/python2: can't open file '/home/tamas/svn/gnue/gnue-appserver/scripts/gnue-gcd2sql' ./setup-cvs-db.sh: line 5: gldcvs: command not found 'sample.gsd' gsd fájl betöltése... Updating schema ... Updating data ... updating table 'gnue_useraccess' ... (-204, 'isc_dsql_prepare: Dynamic SQL Error. SQL error code = -204. Table unknown. GNUE_USERACCESS. At line 1, column 81.. ') tamas@beus:~/data$ i think this is exactly what johannesV_ still has to commit (and can't because svn is broken) we have just restructured the samples today bbl ok bbl *** btami has quit IRC reinhard: fixed *** Vee2d2 has quit IRC *** Vee2d2 has joined #gnuenterprise *** wendall911 has joined #gnuenterprise *** holycow has quit IRC *** wendall911 has quit IRC btami: gnue-gcd2sql doesn't exist anymore; it has been removed by reinhard and is now known as gnue-readgcd bbl :) everyone busy sigh * jamest kicks svn' hi people dinner time *** SachaS has quit IRC *** holycow has joined #gnuenterprise *** btami has joined #gnuenterprise btami: creating new db for appserver should work again btami: there is a script in appserver scripts/gnue-setupdb or scripts/gsetupdbcvs to set up an "empty" database (without address sample, but with all gnue_* tables) and the scripts in samples dir add the sample data *** filc has joined #gnuenterprise lekma: find() in imported procedure works nicely for me reinhard: searching another class? yes do you get a traceback also? (i suppose you use svn) yeah i get a taceback i'm using win2k now so i can't copy the traceback but the 2 last are common\rpc\drivers\Base.py, line 232 and common\rpc\drivers\xmlrpc\pw_xmlrpc\ClientAdapter.py, line 99 *** filc has quit IRC def test(self, testId, find): result = find ('gnue_useraccess', {'gnue_username': testId}, ['gnue_username'], ['gnue_username', 'gnue_password']) print result is what i'm trying to test is there sthing wrong reinhard: it's ok now if i'm using username=gnue and password=gnue for [gnue] in connections.conf but have the same permission error if i want to use username=sysdba password=masterkey as i had used before without error but yes, i know using sysdba instead of adding users not a good practice :) i will test it on XP later too lekma: i can reproduce your error reinhard: ok which means we are close to victory ;-) :-) btami: can you check who is the owner of the database tables? btami, just to check, what happens if you provide username/password on command-line ? now for testing sake i have put just this in OnValidate result = find('gnue_useraccess', {'gnue_username': self.testId}) print result and i have the exact same error than whenn i try to run it from imported modules *** bluesbaron_ has joined #gnuenterprise *** bluesbaron_ has left #gnuenterprise *** SachaS has joined #gnuenterprise hi everyone mornin SachaS hi holycow announcement: I got a job today :) hey right on :) doing what? so if i show up here less you know why i try not to but u never know i have to rebbot brb *** lekma has quit IRC *** lekma has joined #gnuenterprise *** holycow has quit IRC SachaS: congratulations thanks reinhard it was about time :) SachaS, great ! is it that job in germany / usa ? reinhard: about merging gnue-import/gnue-schema ... thanks johannesV_ ye it is s/ye/yes gnue-schema already is able to do updates on records or does gnue-import provide functions which are missing in gnue-schema SachaS: good, what kind of job ? lekma: it will be some ebXML software development closed source :( SachaS: but nof too far from gnue eb in ebXML stands for e-business right? right anyone ever restore a winblows ntfs file system from knoppix? and did it work ;) johannesV_: i don't really remember, so i would like you to have a look at it lekma: bug is fixed svn up and it should work thanks for reporting reinhard: i'll do it right now reinhard, i've checked it gnue-import does nothing more than gnue-schema can do already ok, i'm off for today johannesV_: great please delete it then (tomorrow) :) good night ok, will remove it tomorrow good night *** johannesV_ has quit IRC *** sjc has joined #gnuenterprise reinhard: it works thank you lekma: excellent, thank you for helping to debug it reinhard: i didn't help much, but i'm glad that it works i/me makes happy dance *** holycow has joined #gnuenterprise * SachaS makes happy dance bah everyone ignores me poo on you reinhard: when i'm using username=sysdba for [gnue] all gnu_* tables granted to gnue but chillywilly: i think nobody ever tried to restore an ntfs fs from knoppix :) address_* files granted to sysdba btami: aha btami: that looks like a bug in gnue-schema then i get Loading gsd file 'sample.gsd' ... Updating schema ... Updating data ... updating table 'address_country' ... (-551, 'isc_dsql_prepare: no permission for read/select access to TABLE ADDRESS_COUNTRY. ') gnue-schema seems to not honor username and password from connections.conf while gnue-readgcd does yes, seems to so the gnue_* tables (from gnue.gsd) are owned by gnue yes * lekma is away: feeding the cat... address_* tables (from sample.gcd) are owned by sysdba yes but then the insert of sample data from sample.gsd fails because that again runs as gnue hmmm quite clear :) :) ********************* johannes, please read the above lines in the log ********************** :) good night all *** btami has quit IRC *** kilo has joined #gnuenterprise hi kilo hi reinhard wow * jamest is about to svn up gnue * reinhard hopes not on a production machine ;-) well i was going to ask how stable common, forms, designer are as it's been a long time let me put it that way: i would be interested in *your* answer to that question tomorrow :) lol there have been many changes and i wouldn't risk a mission critical system but i will be here for another hour or so and help you with any problems you find * kilo has to chose between drinking a dram o' whisky or programming in Visual Basic... lol i'm off to a good start kilo: one or the other? jamest@hal:~/svn/gnue/gnue-common$ ./setup-cvs.py Traceback (most recent call last): File "./setup-cvs.py", line 36, in ? CVSDIR = os.path.abspath(os.path.join(os.path.dirname(__file__),'..')) and I always thought one was a prerequisite for the other jcater: it is. money for programming means more whisky 8-)) jamest: is there more of an exception? NameError: name '__file__' is not defined that's all of it i thought python always has __file__ defined it *has* i'm running 2.1 here and jamest@hal:~/svn/gnue/gnue-common$ python Python 2.1.3 (#1, Sep 7 2002, 15:29:56) [GCC 2.95.4 20011002 (Debian prerelease)] on linux2 Type "copyright", "credits" or "license" for more information. >>> __file__ Traceback (most recent call last): File "", line 1, in ? NameError: name '__file__' is not defined that's the same for me but if i write a x.py with this line: print __file__ it works shouldn't import sys sys.__file__ work? i thought any module had __file__ sys.__file__ doesn't work interactive puython session gives error at >>>__file__ even on python 2.3.3 it is not interpreted at an interactive session i think 6116 jcater CVSDIR = os.path.abspath(os.path.join(os.path.dirname(__file__ ),'..')) 6116 jcater BASEDIR = os.path.join(CVSDIR, '.cvsdevelbase') so these lines really seem to have changed lately we are at 6289 now ------------------------------------------------------------------------ r6116 | jcater | 2004-08-01 22:06:24 +0200 (Sun, 01 Aug 2004) | 1 line make it so setup-cvs.py doesn't have to be run from a specific directory (i.e., determine base dir based on __file__) jamest: can you please test if python 2.1 really doesn't have __file__ defined when running non-interactively? sure that'd be listed in __builtins__ right? hmm hmm Vee: you know how to patch SupyBot? I need it to intercept any pastes of svn logs #!/usr/bin/python print __file__ and s/jcater/derek/ is my x.py python2.2 x.py ./x.py python x.py all result in NameError: name '__file__' is not defined doh i don't have a 2.3 box damn derek always breaking stuff man, i know i've been absent for a long time but i'm back now and trying so you don't have to call me such bad names didn't mean to call you "derek" ;-) however as a quick fix you could set CVSDIR to a constant string works on 2.3 as a quick fix, you could cd .. gnue-common/setup-cvs.py :( actually we should just do jcater: no go ? cd .. gnue-common/setup-cvs.py NameError: name '__file__' is not defined you know who's fault this really is? debians! dereks why do we still have to support 2.1? debian! * jcater grumbles nasties under his breath because sarge is frozen so it's changing too rapidly for me to switch this box too not sure if there is a way to find out the directory at all in 2.1 :( there's nothing wrong with debian! nothing right, either sarge is frozen? news ot me there has to be a way as we did tricks like that in gnue-designer *** wendall911 has joined #gnuenterprise to load wizards * jcater thought it was __file__, but maybe not ajmitch_: you joke right? there was a freeze with a target of mid sept release today IIRC jamest: there is a freeze of the base system not of everything ajmitch_ even uploaded new gnue packages a few days ago and they will make it into sarge most probably lol this is funny our online svn browser is down i like the new error dialogs jamest: can you please svn up and try again? jamest: we've changed quite a bit of the error handling so that would be something to test "intensively" jamest: and you didn't see the error dialogs for gtk2 :) :) it worked cool you mention it not needed for 2.3 but you're still using the import just leaving in place till we drop < 2.3 support? yes i guess so ok sigh i think i'm seeing timestamp issues again did gnue stdize on passing datetime objects to the drivers ? or text don't think so I think we agreed to but never implemented it man i am a derek everywhere I go i break things isn't there a way to flip the db drivers from %(mode) to old style? it's puking on woody pypgsql with parse errors useParameters = 0 in gnue.conf thanks brb *** reinhard has quit IRC *** jamest has quit IRC *** jamest has joined #gnuenterprise *** reinhard has joined #gnuenterprise jamest: anything else? jamest: i'm about to go to bed dunno go to bed i'll read the source if i have more issues thanks for the help did the useParameters help? couldn't get it to take tried in [common] and [DEFAULT] building a glimpseindex now to read the code should work in [common] odd unless the debug message still shows as params i'm sure I'm hitting a timestamp issue and I'm sure i'm having issues with older pypgsql drive you using a primarykey on that datasource? doubtful sigh, nope ok, off for today night all *** reinhard has quit IRC jcater you might know if Oracle runs on 64bit linux ... yes Opterons w/SuSE's AMD64 port cool *** jamest has quit IRC thanks night nigth all *** lekma has quit IRC *** kilo has quit IRC *** agx has joined #gnuenterprise Hallo, anybody knows how to disable the default menu and bars of a form? *** agx has left #gnuenterprise *** jcater has quit IRC @last --with vee Vee: Vee: you know how to patch SupyBot? patch? haha.. oh I see the context now.. actually, there is a way to do that I think with the filtering. *** jcater has joined #gnuenterprise *** holycow has quit IRC *** gsoti_away has left #gnuenterprise *** MiXi has joined #gnuenterprise *** MiXi^ has quit IRC *** sjc has quit IRC *** holycow has joined #gnuenterprise *** wendall911_away has quit IRC whaddup my niggas? this channel used to be so hip are you saying that it isn't hip now?