for gsd, what do you mean by non-vendor? .gsd appears to add actual values to the database is that correct? one gsd file creates .sql files for oracle, postgres, mysql, sybase, firebird, sapdb, etc um it supports row insertion but its purpose was defining schemas .gcd also appears to do schema type things but as part of that, you can define default rows to be inserted in control tables, etc yes, different targets .gcd defines appserver classes appserver in turn creates a .gsd (possibly implicitly) which is used to create the database backend so, if using appserver, you should only use .gsd to insert rows? but .gsd can be used independently I am not a good person to ask about appserver stuff, as a lot of development and decisions on it are recent but I'd say, that is likely correct if you plan to use appserver, you'd want to create .gcd, not .gsd can you not "insert rows" or create instances of objects with .gcd files? you would define objects, yes I think that includes inserting rows, but am not sure appserver creates the needed tables based on .gcd files okay I think in the process, it internally creates a .gsd that's all I was saying there are some "attributes" of the objects that are defined with python methods yes these are presumably just class methods although gcd's create tables and store stuff in tables do they store their results in the database? you really can't think in terms of those tables when dealing w/appserver and gcd's or does it generate them when needed? when needed, iirc no caching of the generated values, I assume dunno but would guess not feel free to add any of this to our wiki :) *** bluesbaron_ has joined #gnuenterprise you have a wiki? maybe you should just make the whole site a wiki like freedesktop site is generated automatically from our svn repository so if someone modifies gnue-forms/README, for example then the forms/ descriptions, requirements, etc are updated there is a Community Wiki link on the left menu bar *** bluesbaron_ has left #gnuenterprise added the filetypes just need to tidy it up my god, AppServer is not in the wiki? wtf? ? AppServer is not a wiki word now it is a wiki word rock wiki is really new, relatively speaking we've been trying to push it lately ah well...I added AppServer as a wiki word with a link to the page for it I am thinking it might be good to just make the appserver page point to the wiki instead of the static page on the site our poor webserver needs some tlc and upgrading what does it run on? 300Mhz, iirc basically, hand-me-down machine that was considered hand-me-down 5 yrs ago it seems pretty speedy for a 300 I could give another 300...:-P what do you think of the new pages? exactly what I was hoping someone would do define schema in the context of databases, it would be table, index, and sequence definitions why do you interchagably use class and object in the appserver descriptions? I'll have to defer that one to reinhard but I'd guess class is definition, object is an instance of one but it probably is used inconsistently then creating objects is incorrect verbage for the AppServer howtos I don't know which they want to be the proper term icky icky icky blkANIMAL_Basic.call('ANIMAL_Move', {}) you have to manually mangle the names? and pass parms with dictionaries? icky poo I have no idea what that's from is that appserver stuff? *** jcater has quit IRC yep I know it's not your expertise and you left wt you seem to have some knowledge about these things why are you checking out gnue, just curious? *** btami has joined #gnuenterprise holycow, me? I have been intrigued with it for a while I'd like to see it get the use it deserves cool i've been looking for something like this too, a form based easy dev tool, but with the right framework in mind i love the idea of the appserver and being able to publish anywhere with the right driver as an option *** reinhard has joined #gnuenterprise *** johannesV has joined #gnuenterprise morning all wt: appserver peop on the board, time to ask your remaining question :) ah btami, is there a script that will load the base package .g* files in the right order? for instance typedef before item btami, you there? yes, kilo has something like that... and why is the gnue-packages/base/typedef not called */*/core? ask kilo :) (when he comes) and can you create instances of classes in .gcd files? ie adding database records that define instances of a class ask reinhard/johannesV :))) btami, I thought you were an appserver person no, i'm not an appserver guy yet... wt: to add instances you can (must) use a gsd file look at sample.gsd in appserver/sample directory there we add several instances to both address_person and address_country reinhard, where do the gnue_ids come from in the insert-en-currency.gsd? I am thinking about adding the country codes also in french since there is a list or would that be an application of gld? also, I think vat should be in an extended version of the item module not everywhere has VAT reinhard, you there? *** bluesbaron_ has joined #gnuenterprise Reinhard, I have a patch, I think reinhard, you have an email that I can send this to? wt: sorry for bad responsiveness it's okay i'm at work I am sending it to the devel I am just excited, I guess and as a profession, i'm not programmer but telephoner ;-) wt: the gnue_id's are random numbers (we want to have a good algorithm to create uuid's at some point) and to add country codes with french names you certainly would use gsd, not gld and, fwiw, we would appreciate very much french translations (the fr.po files are quite out of date for most tools) reinhard, i've already made an algorithm for uuid's is there a way to make them be generated when creating instances of classes? but i don't know why we haven't took it over such that you don't have to manually gen them? wt: actually you could enter the data in a form then gnue_id's get generated hmm you don't want that you want to import from a file, right? yep he could export that data afterwards using gnue-gsdgen and then ship that gsd johannesV: how would he import from a file that has no gnue_id? the message should be at the list soon can one import a gsd against an appserver connection? if wt is using appserver he could use the generated form appserver://connection/form/class reinhard, not against appserver itself but appserver's connection johannesV: my understanding is that wt already has a file with the data in it ah, ok, well than this won't work and simply wants to get a gnue_id for each record wt: you can do the trick we do here when importing data and how can I tell who is accessing a db in postgres? you can generate a non-numeric gnue_id this could be accomplished with some minor changes in gnue-schema, i think; but it won't work for relations wt: so for example for country "FR" you would simply generate a gnue_id of of what? IMPORT_COUNTRIES_20041028_FR for example ah so it is a string yes okay all gnue_id's generated by appserver are pure numeric well...I don't think I am ready for that so you could be sure your imported gnue_id's will *never* collide with a generated one is there no way to make appserver gen the id's automatically so I wouldn't even have to worry about it? the problem is when you import more than a single table say you also import addresses that reference the countries you want to import an address with country = france so you would have to know the gnue_id for france in the gsd, you give the table name oh I see you because you have to fill that into the address_country field it would be nice to be able to import lists that don't reference foreign keys reinhard, has my message posted yet? wt: yes that's what johannesV said it should be possible with not too much effort sounds better than current, at least although i'm not sure if it is really good to have 2 different ways of importing stuff where one way doens't work all the time reinhard, i haven't looked into gnue-schema for that problem, but i can imagin it's not too complicated I honestly don't understand why a schema file would have the functionality of inserting rows wt: your message has not arrived right, i don't like that idea too that so wierd because gnue-schema has no means of a gnue-id wt: gsd started as "schema only" it seems like you should have a data import tool of some kind instead the purpose of a gsd is to transport schema and data in a db-independant format at some point the possibility of data was added just think gsd means "Gnue Schema and Data" :) where it can contain both parts independantly how do I make postgresql realize that no procs are accessing it? or maybe just list the ones that are although I don't think any are ddl and dml should be different like in real databases :-P wt: you have a running appserver? no killed it i have no idea actually how to see what processes still use postgres wait a bloody minute appserver is still running oops I forgot that gacvs is the svn name reinhard, if an object has no __nonzero__ but a __len__ the latter would be used this makes an additional __nonzero__ in language.ObjectList.py obsolete why are there cross depends between the packages? johannesV: excellent seems to me that the extended item with vat support belongs in the financial package the extended item I made that is my message is in please give is a look see shit it broke the lines wt: I think kilo would want to look at that but from a quick view it looks logical and then move the item-vat.gcd into the financial package hell, I don't even think the price for an item belongs in the base package damn syntax is slightly wrong I forgot to update the patch file before posting wt: i think it's quite clear what you want actually *** kilo has joined #gnuenterprise *** johannesV has quit IRC reinhard, I just attached the right patch so it can just be applied kilo: wt just sent a patch for gnue-packages to gnue-dev good morning ah yeah good morning :) konnichiha sorry, had a terrible night... kilo, I submitted a patch and shook up the wiki to make it a good day for you oh *** johannesV has joined #gnuenterprise www.laphroaig.com tha would make a good day for me 8-)) kilo, a list of all known gnue file types is listed with a short desc of what they do in the wiki kilo, is that better than Jack Daniel's? kilo, I didn't update the .dia, though definitely. for me at least. I am a dedicated fan of Scottish single malt whisky, so it is #1 for me. I even have my own parcel next to the reservoir of Laphroaig... wt, thx. will look at it asap, and will update the dia too kilo, can you svn-up and try your INV_Head.gfd again ? johannesV: ah, yes, i noticed you committed m/d things I must sleep now later everyone oyasumi night wt good night johannesV: splendid johannesV: hmmm nite *** holycow has quit IRC kilo, so it is working ? wrt to trigger in line 148, it does work but it gives an error when i enable the trigger in line 78 Traceback (most recent call last): File "", line 8, in __main___NAMED_getItemPrice File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/logic/NamespaceCore.py", line 315, in __call__ return self._objectFunction(*args) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/forms/GFObjects/GFBlock.py", line 702, in callFunction parameters) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/datasources/GDataSource.py", line 182, in callFuncOfCurrentRecordsetEntry rset=self._currentResultSet.getRecord(n) File "/home/gabor/SVN/gnue/.cvsdevelbase/gnue/common/datasources/drivers/Base/ResultSet.py", line 121, in getRecord return self._cachedRecords[record] IndexError: list index out of range hmm, sounds bad looks bad reinhard: didn't you have a dream that OnValidate should work just as i told you? 8-) *** bluesbaron_ has quit IRC *** kilo has quit IRC *** btami has quit IRC *** bluesbaron_ has joined #gnuenterprise *** bluesbaron_ has left #gnuenterprise *** jamest has joined #gnuenterprise *** johannesV_ has joined #gnuenterprise *** wt has quit IRC *** johannesV has quit IRC *** johannesV_ has quit IRC *** jamest has quit IRC *** reinhard has quit IRC *** holycow has joined #gnuenterprise *** bluesbaron_ has joined #gnuenterprise *** sjc has joined #gnuenterprise *** jamest has joined #gnuenterprise *** bluesbaron_ has left #gnuenterprise *** wendall911 has joined #gnuenterprise *** bluesbaron_ has joined #gnuenterprise *** bluesbaron_ has left #gnuenterprise *** ajmitch__ has joined #gnuenterprise *** kilo has joined #gnuenterprise *** ajmitch_ has quit IRC *** jamest has quit IRC *** jamest has joined #gnuenterprise *** kilo has quit IRC *** jamest has quit IRC *** jamest has joined #gnuenterprise *** wendall911 has left #gnuenterprise *** sjc has quit IRC *** wt has joined #gnuenterprise who is into the curses client? the curse forms client? *** holycow has quit IRC *** havoc has quit IRC *** havoc has joined #gnuenterprise *** havoc has quit IRC *** havoc has joined #gnuenterprise havoc, hello? hi what do you do with GNUe? I lurk ;) with GNUe? hang on, brb *** havoc has quit IRC *** havoc has joined #gnuenterprise ok better what was that all about ? just messing w/ irssi, it was pissing me off all better now :) I'm friends w/ chillywi1ly and I occasionally harrass the gnue guys ;) ah do you cause havoc? :-P more or less *** bluesbaron_ has joined #gnuenterprise *** havoc has quit IRC *** holycow has joined #gnuenterprise *** jcater has quit IRC *** havoc has joined #gnuenterprise *** havoc has joined #gnuenterprise *** jcater has joined #gnuenterprise *** jamest has quit IRC hola dudes hey so i'm evaluating a shitty windows form based app basically windows exe that servers up html over the network and saves in an access db gonna talk to developers soon to find out if they wanna port app to linux if they say no im thinking of hosing them and hacking out the app in gnue :) ehe good luck ill need it :) i have no clue about gnue forms dev stuff except i've seen people whip up stuff in a jiff