*** ccfiel has joined #gnuenterprise hellol.. hello..pll.. :D *** ccfiel has quit IRC *** jcater has quit IRC *** sjc has joined #gnuenterprise *** sjc has quit IRC *** sjc has joined #gnuenterprise *** reinhard has joined #gnuenterprise *** reinhar1 has joined #gnuenterprise *** reinhar1 has left #gnuenterprise *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** johannesV has joined #gnuenterprise *** sjc has quit IRC *** sjc_ has joined #gnuenterprise *** sjc has quit IRC *** sjc has joined #gnuenterprise *** sjc_ has joined #gnuenterprise see you later *** SachaS has quit IRC *** sjc has quit IRC *** kilo has joined #gnuenterprise good morning kilo: have got your mail will look at it in the afternoon ok, guess you dont have time now then to debate over it no, have to do a few things for my real job before noon ok. then i will send another one soon *** SachaS has joined #gnuenterprise hi SachaS hi kilo how are you? hmmm, well, not enough chocolate * SachaS wanders to the coffee machine *** btami has joined #gnuenterprise people close to Switzerland: Electronic business systems conference http://www.topsoft.ch/ 10. Fachmesse für Business-Software-Lösungen für Handel, Industrie und Dienstleistungen An der Herbst TOPsoft in Brugg-Windisch vom 5./6. Oktober 2004 werden 140 Aussteller über 300 Produkte präsentieren. Eintritt kostenlos. [Messeinformationen] 30 Fachreferate vermitteln Know-how, Trends und Erfahrungen. Eintritt kostenlos. [Programm] In unserer Produktdatenbank sind gegenwärtig mehr als 140 Produkte von 100 Anbietern von betriebswirtschaftlichen Gesamtlösungen referenziert. ERP CRM SCM Warenwirtschaft Add-ons gnue is not yet listed? ;) neither is compiere bbiab *** ajmitch has quit IRC *** ajmitch_ has quit IRC reinhard: shouldn't the fix to share/repository.ini be also applied to src/classrep/repository.ini or only one of them is used? *** ajmitch has joined #gnuenterprise lekma: src/classrep/repository.ini isn't used any more and should be removed (just did that) hi reinhard: when will appserver 0.2 be out? :) i figure about end of this week ok johannesV: what about utf8-ing database creation? There has been talk about it and we reached an agreement that it would be good yeah btami said he'd do it for interbase and johannesV for sqlite and postgres IIRC btami delegated it to me 8-))) and kilo to reinhard, and reinhard to johannesV :) won't work that way ;-) we should decide where to put it. hardwire it in the dbdriver or make schema scripter pass a parameter down to db drivers i have the right string ctrl+c-ed, now i want to ctrl-v it somewhere 8-)) I cannot think of any reason why somebody should want non-unicode database encoding for databases where you can select the client encoding independently of the db encoding so i'd say harwire it in the dbdriver ugh I've lost a cd :( lekma: I'm going to look at your issue with message() in OnInit triggers reinhard: thx is there anything new wrt that? nope i didn't try it again my current status is that calling message() in an OnInit trigger makes appserver forget to save this instance at a later commit and i must say i'm a bit upset right now is that correct? cause i lost one week worth of work yesterday lekma: you did? bad dropdb the status is correct as far as i remember ok which dropdb? su - postgres drpdb gnue exit :( too fast in typing oh and then... so i misinterpreted that you are upset at *us* or are you? no not at you at me ah ok well the real reason you should be upset at yourself IMHO is that you didn't do any backup yeah but dlt tape driver is broken since one month and i didn't have the time to look at it sigh you can still do a dumpdb to /home/backup yeah i know not trying to put you even more down than you are already i wish i would have been more rigorous on this one just sharing my experience mistakes *always* happen if only i could also make backups in real life ;-) :-) *** sjc has quit IRC *** sjc has joined #gnuenterprise lekma: did you mean the complete record is not saved or just the field you set in OnInit is not saved? yes i think i can't test it again right now, but that's what i seen, i was using generated forms on win2k *** wt has joined #gnuenterprise i meant the complete record is not saved i will test it works for me, and the message () is evaluated properly it even behaves properly when an exception is raised in OnInit it gives the exception and does not save it however you might have to notice that forms currently behaves a bit strange wrt OnInit creation of new records and editing them before first save is fully local so appserver doesn't get contacted until you save the newly-created record so the OnInit runs on save operation, not on insert operation (viewed from forms side) and to see if the fields are initialized properly, you have to reload the data we have to sort that out somehow but i need the help of the forms people for that (i think) bbl lunch hi wt hi, kilo I should be asleep of course, it won't happen you shouldnt sleep that early 8-))) just got another viagra and cialis email. how do they know that I need THAT much Viagra??? *** btami has quit IRC SachaS, efficient gnue based system tells them you need it wt haha kilo, are you going to do that encoding-stuff for interbase ? aye, just khmmm invsting it... at the moment we are here: (-204, 'isc_dsql_prepare: Dynamic SQL Error. SQL error code = -204. Data type unknown. Implementation limit exceeded. COLUMN GNUE_CODE. ') well, that means the string-length of gnue-code is to long think it's about 32k it is not specified in gnue.gsd. right this means "as long as possible" and the interbase-db-driver says: use 32000 chars for 'non-specified' strings but this should be an old problem then ? lunch time does anybody already started to work on automatic user/time stamping new/updated records? i would like to know what would be the prefered way to do this in gnue appserver would it be adding 4 properties (createdby, createdat, updatedby, updatedat) to all classes or store all stamping in another class referenced by objectid? (i don't like the later) lekma: i agree with you lekma: we will automatically add those 4 fields to every class like gnue_id is also automatically added reinhard: ok and johannesV will be working on it this and/or next week johannesV: and what does this meam? ProgrammingError: (-104, 'isc_dsql_prepare: invalid request BLR at offset 153. Implementation limit exceeded. block size exceeds implementation restriction. ') kilo, i'll check it (one by one ...) i've limited varchar length to 10000 so that one works * reinhard wonders that kilo asks johannesV about interbase actually it should be the other way around, shouldn't it? off again *** btami has joined #gnuenterprise reinhard, actually i (re)wrote most of that interbase stuff (new driver) at least that schema-creation/introspection stuff kilo, the former problem (gnue-code exceed): this occurs when changing character set to UNICODE_FSS (which means more bytes are used to store a single character, and thus reaching impl.limits faster) so a 'usual' varchar-column cannot hold 32000 characters any more (I'll look for a suitable length then, but it's not very *clean*) it is 10900 ah, ok so we have to determin if we like that as a max. restriction the other way would be using blob, but blob is somewhat very tricky ... ok have to leave for lunch now, bbl *** sjc has quit IRC *** btami has quit IRC *** dimas_ has joined #gnuenterprise *** dimas_ has quit IRC *** dimas_ has joined #gnuenterprise reinhard: what is that character named that signs double s in German? you mean ß sharp s ß scharf s? exactly scharfes s like scharfes gulasch that is used in strasse ;) yep oh, wonderful... it's used when the s is voiceless and after a long vowel 's i svoiceless'? is yes you can say an s with or without voice oh, it must be like out j and ly, both like y in yes but a little bit different... with sound = like a bee without sound = like a snake err s/sound/voice/ hmm, languages are beautiful... in Gaelic there is a rule that on each side of a consonant same category vowels must be used and it is a principal rule but there are 20 exceptions... 8-))) lol it is such a rule that was first introduced when the monks invented Gaelic writing in the early early middle ages gaelic is fun yes, it is kilo: i really don't like the thought of setup.py installing a dos batch file into /usr/bin under linux maybe you could try to make it conditional? and install the one or the other depending on sys.platform or os.system? blame btami. he was playing with Windows and did not want to commit it from there because it takes ages... so he asked me to do that. and as soon as I have done it, he left... 8-)))) ah, kilo did that unicode-thingy ok, it works, so i'll revert my changes in interbase-driver yes, sorry if i caused extra work np *** dimas_ has quit IRC *** kilo has quit IRC reinhard, gnue_company and gnue_year classes will be added to share/gnue.gsd, right ? *** jamest has joined #gnuenterprise yes gnue_company.gnue_code == string (8) ? yes name == string(35) ok, had this gnue_year.gnue_code = string (8) yes i hat changing gnue.gsd ... do we have a 'good' name for gcd's classes now (for company/year/None set) hmmm ok, question to all especially to those speaking English as a mother tongue in appserver we need some multi company ability to have data for different companies in a single database also we will have data that is dependent on the fiscal year we want to support that transparently in appserver so when you define a new class (2-tier people: read "table") you can say whether it is dependent on the company, dependent on the fiscal year or it is "global" to all companies (like, for example, a currency table could be) so the question in which of those 3 groups a class belongs is an attribute of the class we are looking for a name of this attribute :) --- roflmao reinhard, is a reference to gnue_company nullable ? (i'd say no) *** wendall911 has joined #gnuenterprise university of paderborn, germany, received a letter from Motion Pictures Assiociation of America (MPAA) telling them they offert the movie "Ghost" for illegal download and should immediately stop it and remove the download link however the link was actually for downloading GNU ghostview johannesV: no, not nullable johannesV: i# i'm tempted to say filter="company" *** johannesV has quit IRC *** johannesV has joined #gnuenterprise johannesV: i'm tempted to say filter="company" filter="year" as that is what it basically is also if possible please don't hardcode "company" and "year" anywhere except in the gsd we might add other filter possibilities at some point like sub-entities of companies (department, subsidiary...) think of CSS who had 10 subsidiaries all sharing the same database but every subsidary having its own customer data it's hardwired in gcd/GCParser.py there (as for gnue_id) all 'dynamic' refs are generated so to the outside it seems as if the refs was there in the gcd-file itself ok can't you change that to something like 'gnue_' + whatwasgivenasfilter well, i think so it should be no problem but what about combinations ? * reinhard is still not sure about the "filter" combinations? for company and year it's trivial since there's a m/d-relationship between them but if there are 'free' filters this might not be the case for all of them so is it possible to have filter="foo,bar" ? I'd say no ok, np at least not until we really see the need ok, then it works already [up to the gcd-parsing and schema-creation] i'm preparing for the dentist now, but i think i can continue with some work in the evening ok have fun i mean... ;-) well, it's not me who has to lay down at that chair *lol* *** johannesV has quit IRC *** dalen has joined #gnuenterprise a bug in Designer is that when you change a Page Caption the tab for that page in the layout editor doesn't change *** holycow has joined #gnuenterprise *** holycow has joined #gnuenterprise *** reinhard has quit IRC *** holy_cow has joined #gnuenterprise *** holycow has quit IRC *** jcater has joined #gnuenterprise *** insanekane has joined #gnuenterprise * SachaS using evolution1.5 from experimental :) morn SachaS :) any good? holy_cow sure :) dist-upgrade wants 555MB+ hmmm hehe :) sarge is going to be something like 11 cd's yes hay guys whats goin on in this biggest free software collection ever created sarge is supposed to be released in september, so any time now well base and standard are frozen but extra, etc. are freezing 'Real Soon Now' nickr: shipping with gnome 2.6? I don't know. nickr: probably 2.2 ;) 2.4 more likely. * wendall911 wonders about Debian at times they should have server and desktop distros that way you could actually see a working desktop with software written in the same decade Gnome 2.8 stable was just released that concept doesn't fit with the way debian does things. nickr: would be nice, hence the reason I only use Debian infrequently for desktops that sounds like a personal problem. heh *** insanekane has quit IRC *** kane has joined #gnuenterprise *** kane has quit IRC *** insanekane has joined #gnuenterprise no im sure ist shipping with gnome 2.6 why would 2.6 be in sarge now if it wasn't? *** wt has quit IRC *** wendall911 has quit IRC *** wendall911 has joined #gnuenterprise *** reinhard has joined #gnuenterprise hello reinhard hi insanekane all: I have commited a first step of a developer's guide please check it out and give me some feedback if it's understandable etc. thanks *** erkkie has joined #gnuenterprise reinhard, umm ... where is it ? reinhard, u mean it is in the svn repo ? insanekane: yes and in the web reinhard, when did u commit ? roday ? today ? (although not yet linked from anywhere) i committed about 1 hour ago aah ok ... (couldnt find it in the svn of yesterday) ... http://www.gnuenterprise.org/tools/appserver/docs/manual/ the devguide.* files please choose the format you like :) reinhard, the GCD editor is complete btw ... im working on now adding multiple modules in the same project .... reinhard, but that will take more time reinhard: in gcd i'm defining names for module, class, property in the db names would be mod_class for table names mod_prop for columns right? dimas, yes then when i define block in the form umm ... i dont know anything abt forms sorry i could use any prefix or none for _prop but if i reference it later in say simpleQuery i should use the name i used in block/field definition reinhard: is that expected behaviour? insanekane: thanks, anyway :) dimas: for field= you have to use the mod_prop syntax with correct module name insanekane: did you ever look at designer? i was told that it should be easily possible to have a gcd editor as a designer plugin reinhard: i have used incorrect module in mod_prop and it was allowed and ever worked dimas: really? let met test please seems so doh you're right that's absolutely unexpected behaviour it reconstracted proper mod_prop from _prop part and datasource it's how i could explain all this i was able to use even "prop" alone dimas: this is a bug and will be fixed before 0.2 thanks for reporting it reinhard: thanks too i'm sorry you will have to change your forms :( :) could you tell anything about timestamping feature? how it could look like? there will be the following properties available: (in each class, automatically) gnue_createuser gnue_createtime gnue_modifyuser gnue_modifydate (we are not sure about the names yet, but you get the idea) these fields are maintained automatically by appserver, much like the gnue_id and you can access them to find out date/time and user of instance creation and last change and how long it would take to get into svn? expect it during next week great reinhard, yes ive seen designer .... but i didnt mess with it ... i just made a very simple utility for reading .gcd files and editing them ... *** jamest has quit IRC *** jamest has joined #gnuenterprise dimas: fixed it - this really was a grave and dangerous bug thanks again what was your email address again? evolution 2.0 is nice :) evolution-2.0.0-1 from debian experimental reinhard: you're in good old jcater's mode someone raise problem the short period of silence * jcater hasn't been in that mode for a long time *** wendall911 has quit IRC then offering to cvs update with bug fixed :) glad to see appserver making progress though :) jcater: :) reinhard: ds at vbc dot elcom dot ru *** dpr has joined #gnuenterprise dimas: nevermind just decided to remove email addresses from THANKS file reinhard: i am having a problem with the messageBox function, i finally got it to work showing both 'yes' and 'no' buttons, but now no matter what the person clicks it returns "None" ? i'm using, print(messageBox('test', 'Question', 'test', False)) to test it out dpr: in a form trigger? reinhard: yeh, button trigger ok http://www.gnuenterprise.org/tools/forms/docs/faq.php Q: What about the curses client? Other clients? ajmitch: The curses client does not work properly but is under development. Work on an HTML client has begun. s/ajmitch/A is that correct? *** kilo has joined #gnuenterprise dimas: that's wrong curses client should be working basically need to update that stuff dpr: i can reproduce dpr: fixed, thanks for your bug report dimas, dpr: you see why we need no bug tracking system? ;-) ooh, people, this appserver bug was not a bug, it was a whole... emm... godzilla... kilo: you have lots of wrong forms? kilo: btw, you might want to update hu.po for appserver before the final release seems like 2 or 3 translations are missing dimas: also, if you want to update ru.po reinhard, u have put only api.* and guide.* with approx 2 chapters each yes ? dont think i have wrong forms insanekane: please forget api.* reinhard: when do you plan to release reinhard, okie ... devguide.* is the document i'm talking about reinhard, yeah i went through it .... and yes, it's only 2 or 3 pages reinhard, its kind of short ;) but want to get "early feedback" before i put more time into it reinhard, we should change that LOL :) kilo: will do -pre1 maybe even today and final over the weekend probably ok, will update po then regard appserver as "string freezed" reinhard, hopefully, when u (and I) get time to get into the tutorial, then the doc can be updated little by little everyday ... reinhard: pre1 with timestamping already? :) string freeze is on? ok then i will do it today dimas: timestamping will wait after the 0.2 release kilo: would you think we should also release common and forms again? there have been some bugfixes imho we are not needed to do so reinhard: and don't forget designer! we have that new feature, goat icon :) jamest, jcater, what would you think about releasing forms and common twice within 2 weeks? don't want release inflation I have no problem w/that but meanwhile i've broken down the release process to 2 minutes work per package (not including the update of the NEWS file) reinhard: thanks, just updated via svn and it's working 100% dpr: excellent reinhard: can we request new features? :) dpr: however, that bug was not *that* grave, so you'll have to report another few to make it into the THANKS file ;-)) dpr: you can try but cover your head ;-) reinhard: k, i'll keep working on it well, what i wanted was a function to automatically insert 3 spaces instead of a tab into the trigger editor 'cause u know how much python detests the tab dpr: ouch dpr: i'm not the person to talk about designer :)( i.e. when u hit tab, it inserts 3 spaces instead of a \t that's one of the few tools i hadn't put my dirty hands in yet i like the designer, it's improving more and more as each day goes by dpr: huh? ? dpr: apart from a new logo, there hasn't been a change for months now IIRC reinhard: really? from the release on the website? in designer yes 'cause i had the release from the website and it was unstable as hell kept crashing then i switched to the svn and it seemed a great deal better dpr: there were some changes to play it nice with wx 2.5 ah, that's prolly it then 'cause i'm using wc i mean, wx designer crashing is not a bug, it is a feature lol dpr: this was done 2004-08-03 may be designer is self improving? :) and before that, the last change was 2004-05-30 ah, yes, it has a hidden AI in it perhaps it's gentoo updaing packages which were causing problems with designer reinhard: any reason i get an attribute error when trying myDataSource.delete() it tells me there is no such attribute? no attribute delete? i presume it wants me to give a parameter for the data i want deleted? i assumed it would delete the entry i am currently working with? you sure delete() is the correct function name? according to the documentation i have? Developers-Guide.ps, pg. 61 hmm seems to be correct what exactly is the error you get and do you get a traceback? hold on, i think i am going to try blockName.deleteRecord() i think that's what i actually need ah * reinhard notes some bugs in his release announcement script is it available to switch button into enable/disable rendering? reinhard: next q, is there a way to disable the delete/append etc. buttons on the top of gnue-forms? i.e. i want to disable the delete button, but i still want to be able to delete records dpr: i don't know dpr: maybe somebody else knows but still this is fun prerelease out within < 5 minutes *** btami has joined #gnuenterprise including upload to web page, regenerating all documentation, rebuilding man pages and dtd's, announcement mail, tagging svn... btami: i would prefer gnue-setupdb.bat to be installed only under windows so it shouldn't be included into params ["scripts"] in a hardcoded way maybe something like if sys.platform == 'win32': setupdb = 'scripts/gnue_setupdb.bat' else: setupdb = 'scripts/gnue-setupdb' hi all reinhard: agreed hello to all from here also qman: ould i put --debug-file in subdirectory like in appserver://../class?--debug-file=subdir/debug.gfd just wanted to commit thet .bat not to forget until i go home btami: ok anyone could test that? reinhard: the same for other bash scripst in samples actually yes hello erkkie however i'm not too much after the sample files at least they don't get installed in /usr/bin :) ok i don't mind too much samples/ being installed completely i see it is more needed for gnue-packages a little bug, maybe: i get unicode errors, when i try to start appserver and when the backend database server (mysql in my case) is not running. appserver is trying to show me a error message, but fails with something related to unicode a have a error log also, should i paste that here, or send by e-mail or ...? dimas: i get a traceback with debug-file= something with subdirectory and a message that i should report it to the developers ;-) * reinhard loves our new error messages erkkie: what version? of appserver and common ok, than message received :) dimas: i will not have time to look at it today GNUe Application Server Version 0.1.90 GNUe Common Version 0.5.8 dimas: can you maybe remind me tomorrow to tell it to johannes? haven't tried cvs yet.... erkkie: ok, so please send the traceback here that's current "enough" :) reinhard: not urgent at all, will tell though dimas: thanks DB000: File "/home/erkki/gnue/root//lib/python/gnue/common/datasources/GConnections.py", line 464, in loginToConnection DB000: tmsg = u_("Unable to log in after 4 attempts.\n\nError: %s") \ DB000: File "/home/erkki/gnue/root//lib/python/gnue/common/apps/errors.py", line 222, in getException DB000: return (aValue.getGroup (), aValue.getName (), aValue.getMessage (), DB000: File "/home/erkki/gnue/root//lib/python/gnue/common/apps/errors.py", line 108, in getMessage DB000: return self._fmtUnicode (self.message) the traceback is longer, these are the last lines DB000: File "/home/erkki/gnue/root//lib/python/gnue/common/apps/errors.py", line 129, in _fmtUnicode DB000: return unicode (text, errors = 'replace') DB000: TypeError: coercing to Unicode: need string or buffer, instance found erkkie: sorry i was wrong gnue-common 0.5.8 is *not* current enough exactly that bug was fixed a few days ago :) ok, sorry:p erkkie: thanks anyway for reporting it confirmed me that we should release again :) reinhard: how to add/implement UNIQUE constrain to class property? don't worry... i'll stick around for a while.. i'm evaluating gnue stuff for a project and i'm sure i'll get some real problems also:D dimas: add an OnValidate trigger ok dimas: we have considered UNIQUE constraints in appserver but it will be a long time before they will get imlemented, if at all reinhard: if i am within a trigger for an entry box and i want the entry box to keep the focus if bad input is entered, why does doing a self.parent.setfocus(self) not work? dpr: there should be an abort() function or something like that to tell the entry to not lose the focus in PRE-FOCUSOUT but i'm not entirely sure if (and how) it works reinhard: could you explain as how calculated properties are used via ah just do hmmm actually i'd like me too :) off for today night all *** reinhard has quit IRC dimas: http://www.gnuenterprise.org/~kilo/images/packages-thoughts.png looking.. nice looking i do not undersatnd dimas' part though :) lol i was trying to explain that it would raise possible difficulties with the parallel storing of data of the same kind... please point me to that difficulties i was taught very strictly that the same kind of data should be stored in only one place, not scattered i'm with you there *** sjc has joined #gnuenterprise but also we have reinhard's opinion that he does not need/want schema to be too complex *** wendall911 has joined #gnuenterprise see yas later *** SachaS has quit IRC ah, back, I was fighting through my usual daily getViagra/million-$-from-nigeria/penis+17%/stop-spam emails... dimas: yes, i am eager to hear reinhard on that hope you won thx *** btami has quit IRC *** erkkie has quit IRC *** SachaS has joined #gnuenterprise *** insanekane has quit IRC *** SachaS has quit IRC *** SachaS has joined #gnuenterprise *** SachaZzzz has quit IRC *** kilo has quit IRC *** jamest has left #gnuenterprise *** jamest has joined #gnuenterprise *** SachaS has joined #gnuenterprise *** sjc has quit IRC *** sjc has joined #gnuenterprise *** sjc has quit IRC *** insanekane has joined #gnuenterprise *** MiXi^ has joined #gnuenterprise *** gsoti_away has joined #gnuenterprise *** jcater has quit IRC *** mixi has quit IRC *** holy_cow has quit IRC *** jcater has joined #gnuenterprise *** gsoti_away has left #gnuenterprise *** holycow has joined #gnuenterprise *** jcater has quit IRC *** jcater has joined #gnuenterprise yo ho ho and a bottle of rum wassup/ ? * chillywilly plays with a new mozilla sunbird release http://ftp.gnome.org/pub/GNOME/teams/art.gnome.org/backgrounds/mozilla1280.jpg cool new background ;) heheh chillywilly: figured you'd appreciate it ;) firefox and thunderbird are some pretty kick as apps azz agreed sunbird is getting there chillywilly: yeah, I'll have to build it sometime soon and check it out * chillywilly just downloads the pre-built releases ;P chillywilly: I have been using evolution for the last couple years bleh ;) only problem I have is you cannot export mail filters...or at least I don't see a way to do it chillywilly: I'll switch to Thunderbird sometime...it wasn't very stable last year.... I love how firefox scales images in the browser chillywilly: exporting stuff is a pain chillywilly: the new search is pretty cool chillywilly: 1.0_pre * chillywilly uses http://del.icio.us for bookmarks anyway yea I have 1.0 pre I like the search I even like nvu but it needs some work dreamweaver is still the best' chillywilly: bah, text editor um, nah freakin' hate html or dynamic, which is my preference http://design.metalexpress.net/mx/material/Nice%20Error%20Handling%20Dipshit%20:) hehe just kidding ben :) *** wendall911 has quit IRC *** ccfiel has joined #gnuenterprise hello..pll.. hello *** jcater has quit IRC chillywilly are you a user of gnue? :) *** Nicknam1 has joined #gnuenterprise *** Nicknam1 has quit IRC *** ccfiel has quit IRC *** holycow has quit IRC *** jamest has left #gnuenterprise