um have you all tested datasource on 'different' pages? i moved my master/detail from blocks to datasources and that worked well :) the thing that is odd is a have one detail on a different it has not been tested on all forms it seems to be functioning to a certain degree but for the life of me it wont let me exit the form pages should have nothing to do w/ datasources it says i didnt save/rollback when i did let me try one more thing as this form has this issue anyhow ok my bad argh.... we need better error handling :) or smarter users :) ConnectionError: error 'ERROR: pg_atoi: error in "big contact type": can't parse "big contact type" ' in 'INSERT INTO history (subject,location,contact_type,notes,person) VALUES ('some stuff','the location','big contact type','my notes','3')' ------------------------------------------------------------ NOTICE: current transaction is aborted, queries ignored until end of transaction block i missed that in my log window i had a 'data type' mismatch basically :) i had done the save got no error then played with other stuff and tried to exit it wouldnt let me when i looked at the error log it didnt say anything (as i had done a bunch of other stuff that got logged) so was thinking datasource issue (classic blame last issue syndrome) but real debugging shows just poor error handling and dumb user :) scary thing, a little clean up and we have semblence of a minor contact manager hi all hi and bye I'm off to get some sleep night night jamest (jamest@fh-dialup-202033.flinthills.com) left irc: [x]chat derek: what's up buddy not too much doing a dcl install as we speak well if my connection would stay fast i would be doing a dcl install anyhow gravy what kind of connection sir um wireless (satellite) 10Mbps down 19Kbps up notice the M and K diff in down and up respectively and the 10 down is probably more at about 3M most of the time Hmmm! oh yeah, sprint well that sounds alright i've got a buddy in Sunnyvale with one of those ajbusy (ajmitch@p61-max6.dun.ihug.co.nz) left irc: Ping timeout for ajbusy[p61-max6.dun.ihug.co.nz] he says the availability and throughput varies with the weather especially fog and rain clouds are ok, i guess and he's on the edge of coverage man if that does 10Mbps, doesn't it fry birds? ;) hmm i havent had issues with weather ajbusy (ajmitch@p62-max11.dun.ihug.co.nz) joined #gnuenterprise. though i cant remember last time we had fog or rain Action: derek is in arizona ra3vat (ds@195.239.64.162) joined #gnuenterprise. right on well 5 minutes == new dcl install gotta love that funny thing is it takes me longer to customize it than it does to install it dtm so what happened to you for the rest of LWCE? i didnt see you after first day i wasn't there i was on working here at home sorry bout thta i wanted to be maximally productive and i desperately have to get some stupid stuff done so i only broke from that for some cool FSF duties and saying hello plus i'm not sure how i would have dealt with Brad and his "behavior" i've had _quite_ enough of that with the ppc linux community :) and nobody called me! :) so i figgerd things were ok it sucks that i was so busy during the only time i get to see real buddies such as yourselves ;) :) no big deal, just felt bad only got to say briefly hello assumed you were gonna be there following day :/ i assumed so as well Mr_You (rouzer@mankind.boredom.org) left irc: ircII EPIC4-1.0.1 -- Are we there yet? uday (kotuday@host-64-110-96-159.interpacket.net) left irc: Sacha (sacha@chime-a-142.conceptual.net.au) joined #gnuenterprise. hi sacha well im running to bed to all a good night Nick change: derek -> dnSleep Action: dnSleep is away: sleeping ra3vat (ds@195.239.64.162) left irc: [x]chat reinhard (rm@N802P023.adsl.highway.telekom.at) joined #gnuenterprise. hi guys what an ambitious project! hi Sacha sorry still sleeping although i'm sitting in the office ;) are you developper for the gnuenterprise project? yes any questions i can answer for you ? don't know yet. let me think. actually i am studend and dealing with intelligent agents and e-business.... so erp is an integral part of e-business. well maybe even e-business is an integral part of erp :) now i am wondering: can i use an open source project to build an e-business company i hope so as this is my plan :) with all possible addons: CRM, SupplyChain Management, E-Procurement, ERP, Electronic Markets. some of them will be part of GNUe anyway like CRM, SupplyChain and maybe others can be added by you what we will have initially will depend on where we will get developers true. i see different open source projects deal with different aspects of e-business. integrations comes to my mind. yes we are in good contact with phpgroupware and bayonne any other project you were aware of? gnucash oh yes a developer of gnucash is in this channel regularly i forget his name :( ajbusy (ajmitch@p62-max11.dun.ihug.co.nz) left irc: Ping timeout for ajbusy[p62-max11.dun.ihug.co.nz] our plans are more like cooperation than integration we think that it's good to have several projects dealing with different needs like gnucash = home user gnue = corporate user atm it seems to be a very big effort to realise your ideas sure more like cooperation than integration. what do you mean with this? integration = merging the projects ok. cooperation = still different projects, but sharing ideas and maybe code, and building interfaces ok. very good. we had a lot of merging btw. obelisk gnu gl project sanity and 2 or 3 others where i don't remember the name i never heard of these projects. yeah that's why they merged probably like not so many people know where queensland is but everybody knows australia :) if you know what i mean actually i come from liechtenstein. nobody knows where liechtenstein is. what? (sacha@chime-a-142.conceptual.net.au) you have a .au internet address? i am postgrad student here in australia. overseas student. ah btw i know liechtenstein ich komme aus vorarlberg :) ja ich dachte es mir. cooperation sounds good to me. lately i started to read papers on ebXML, cXML, xCBL, OBI. Action: dtm looks up driving directions from California to Liechtenstein there are different commercial products which provide the infastructure to do e-commerce based on these standards. dtm Liechtenstein is a small country in Europe. Between Austria and Switzerland. It has only 30'000 ppls chillywilly (danielb@d75.as8.nwbl0.wi.voyager.net) joined #gnuenterprise. intelligent software agents are still research topics even there are a few commercial products (intelligent software agents do business) available. there are different software agent platforms available. fipa-os (www.fipa-os.org) is a known open source agent platform. ajbusy (ajmitch@p2-max4.dun.ihug.co.nz) joined #gnuenterprise. in my project i am thinking how agents could do business. so first i started to read in e-business in general. and came along your gnuenterprise project. ajbusy (ajmitch@p2-max4.dun.ihug.co.nz) left irc: Ping timeout for ajbusy[p2-max4.dun.ihug.co.nz] ajmitch_ (ajmitch@p44-max5.dun.ihug.co.nz) joined #gnuenterprise. Sacha: cool. welcome :) chillywilly: hi hi http://enterprise.kde.org anyone ever look at this? chillywilly: you want I should go bust em up, bawss? eh? you're a littl nutty tonight oh ok chillywilly: i don't think that project is something like we do it seems more like a collection of cases where "normal" kde apps (konqueror, koffice...) are used in business tell ya the truth I didn't even follow the link hehe I don't want jamest to start making us ude kdelibs :P why would you use kdelibs? cause jamest is now an official KDE weenie that site has nothing to do with GNUe, really it's just a site with info about using the KDE desktop in the enterprise #kde-enterprise Action: chillywilly joins heh it was empty haha so? we win bah how petty shutup what's wrong with you tonight ajmitch_ lost your sense of humor do I need to borrow you mine/? Action: chillywilly thinks ajmitch_ is either depressed or cranky yes i have lost my sense of humour, so don't bother me btw, when did you grow a tail? sorry? what new offensive comment are you making? well at least _I_ read the enterprise.kde.org link, unlike _some_ people. however, yes, we win. what is that wyatterp project? reinhard: i see you merged with OpenRP OpenRP? chillywilly. openrp.sourceforge.net that was back in 2000. sorry guys. Sacha: i remember but i don't know if anybody of OpenRP is still with us heheh reinhard. ok if i will follow the open source erp development i should allways have a look at gnue. is there an alternative os erp? not sure there was linux kontor which seems to have died meanwhile which would be sad as they were way ahead dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] weeeeee! #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ OK. I picked up FERMS. are they still alive? don't know from what i can tell Sacha (sacha@chime-a-142.conceptual.net.au) left irc: Killed (NickServ (This nick is reserved by another user)) sacha_ (sacha@chime-a-142.conceptual.net.au) joined #gnuenterprise. most of the other erp projects never got further than planning state afaik ya and the cool thing about GNUe is that we have no planning so we're still around ;) ok. i get a dc in a couple of minutes and will come back later. if you are here at U.S. normal time you can meet derek Action: chillywilly hides sacha_: i am glad you're makign such an effort who keeps an eye on other projects reinhard: so did you consider to reuse the code of kontor for example? what for? we got our own code chillywilly to use the good ideas from kontor. we pretty much have a our own direction plus no one has bothered to look :P maybe derek has sacha_ (sacha@chime-a-142.conceptual.net.au) left irc: Ping timeout for sacha_[chime-a-142.conceptual.net.au] Action: chillywilly waits and waits for the sgi kernel cvs to update sacha_ (sacha@chime-a-166.conceptual.net.au) joined #gnuenterprise. sacha_ (sacha@chime-a-166.conceptual.net.au) left irc: Client Exiting sacha_ (sacha@chime-a-166.conceptual.net.au) joined #gnuenterprise. sacha_ (sacha@chime-a-166.conceptual.net.au) left #gnuenterprise. sacha_ (sacha@chime-a-190.conceptual.net.au) joined #gnuenterprise. reinhard: is ferm also dead? i don't know just look at when the last activities were last mails on the list, last releases ok. can't access their website at www.ferms.org looks _very_ dead then :) GNUe is the only one who is still alive and kicking GNUe is dead! long live GNUe! Action: chillywilly slaps ajmitch_ with a trout I think I am addicted to compiling kernels brb chillywilly (danielb@d75.as8.nwbl0.wi.voyager.net) left irc: Read error to chillywilly[d75.as8.nwbl0.wi.voyager.net]: EOF from client hehe chillywilly (danielb@d142.as28.nwbl0.wi.voyager.net) joined #gnuenterprise. chillywilly (danielb@d142.as28.nwbl0.wi.voyager.net) left irc: good night ra3vat (ds@195.239.66.43) joined #gnuenterprise. hi ra3vat you found a good bug yesterday ra3vat (ds@195.239.66.43) left irc: Client Exiting ra3vat (ds@195.239.66.11) joined #gnuenterprise. reinhad can you develop on gnuenterprise while you work? are you a paid developper? well depends on how you see it i am self employed so i pay myself for developing gnue :) ok. Nick change: reinhard -> rm-eat ra3vat (ds@195.239.66.11) left irc: Ping timeout for ra3vat[195.239.66.11] sacha_ (sacha@chime-a-190.conceptual.net.au) left irc: Ping timeout for sacha_[chime-a-190.conceptual.net.au] sacha_ (sacha@chime-a-25.conceptual.net.au) joined #gnuenterprise. Nick change: rm-eat -> reinhard have to visit customer will be back later bye reinhard (rm@N802P023.adsl.highway.telekom.at) left irc: Those who say it can't be done should never stop those who are actually doing it. ToyMan (stuq@c5300-2-ip103.albany.thebiz.net) joined #gnuenterprise. ToyMan (stuq@c5300-2-ip103.albany.thebiz.net) left irc: [x]chat ra3vat (ds@195.239.66.40) joined #gnuenterprise. sacha_ (sacha@chime-a-25.conceptual.net.au) left irc: Ping timeout for sacha_[chime-a-25.conceptual.net.au] sacha_ (sacha@chime-a-25.conceptual.net.au) joined #gnuenterprise. dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. ra3vat (ds@195.239.66.40) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. ra3vat (ds@195.239.66.40) returned to #gnuenterprise. #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. ra3vat (ds@195.239.66.40) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. ra3vat (ds@195.239.66.40) returned to #gnuenterprise. #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ ra3vat (ds@195.239.66.40) left irc: Client Exiting neilt (neilt@dialup-166.90.69.174.Dial1.Chicago1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ reinhard (rm@N802P023.adsl.highway.telekom.at) joined #gnuenterprise. ToyMan (stuq@c5300-1-ip123.albany.thebiz.net) joined #gnuenterprise. neilt (neilt@dialup-166.90.69.174.Dial1.Chicago1.Level3.net) left irc: bbs and hour or so dres_ (dres@4.18.171.42) joined #gnuenterprise. dres__ (dres@4.18.171.42) left irc: Ping timeout for dres__[4.18.171.42] dres__ (dres@4.18.171.42) joined #gnuenterprise. dres_ (dres@4.18.171.42) left irc: Ping timeout for dres_[4.18.171.42] ra3vat (ds@195.239.66.37) joined #gnuenterprise. reinhard: you saw my yesterday's greeting yes it's a cool error message hard to fix? because when you look closer at it it points to a general design fault in geas :) the problem is whenever a field of an object is set with setField i still don't see it that close :( the data is written to the database immediately so if you have an object with two NOT NULL fields it is impossible to create a new instance because no matter which field you set first the database will bitch about the other one being NULL i was preparing a query ra3vat: yes it's happening for an internal object of geas but it would be happening for every other object that has 2 NOT NULL fields and the error was happen when I added second field it's not _that_ hard to fix but i want to talk to neilt first to decide which way to fix it but i think it will be fixed today via addField and then executeQuery(q) with one field it worked via addField and then executeQuery(q). with one field it worked jamest (jamest@hobbes.math.ksu.edu) joined #gnuenterprise. #gnuenterprise: mode change '+o jamest' by ChanServ!s@ChanServ reinhard: you here? dres_ (dres@4.18.171.42) joined #gnuenterprise. yes here dres__ (dres@4.18.171.42) left irc: Read error to dres__[4.18.171.42]: EOF from client hi guys neilt (neilt@dialup-166.90.69.174.Dial1.Chicago1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ dres_ (dres@4.18.171.42) left irc: Read error to dres_[4.18.171.42]: Connection reset by peer dres_ (dres@4.18.171.42) joined #gnuenterprise. hello all reinhard: thanks when I get the outline cleaned up, I'll send you a copy neilt: i hope my bio is ok and you are ok with that i wrote i coordinate geas as actually we both try to :) neilt: we have a design problem in geas the function dataobject.setField immediately flushes the data to the database that is not good if you have a calss with two NOT NULL fields and both have no init value em no default value then you can't set one of those fields without the database backend this is the error from ra3vat giving an error because the other is still NULL yes well the error from ra3vat is in a geas internal class but it could also happen for a normal class i see several ways of fixing it but my favorite is not to flush immediately but to have a explicit flush function OTOH this will certainly break all existing clients not an explicit flush that is in the code already at least in the IDL yeah void DataObject_flush (GEAS_object_reference * id, CORBA_Environment * ev) { /* not really needed: leave empty until further notice */ } that is it's implementation :) cool that should not cause any exceptions then, huuh lol we need a flush strategy this means that we flush if we can if we have a condition that should not allow a field to be sit, we then wait for more data and flush at the first posible moment i'm not sure if i like that implicit immediate flushing for example hold on no let me rephrase we need a interface method to set more fields at a time think we want to update 20 fields at once we now have to call setField 20 times which makes 20 SQL UPDATE's for the same record smells like bad performance yes it does i like the multiple fields idea there is another thing iirc andru wanted it and andrewm was against it but it was done then anyway if you call newObject (or what it is called) yes newObject then the object is only created in memory and only written to the database if you set a field afterwards so an object where all fields are still set to the default value is not persistent --- sigh too many logs to read Nick change: dnSleep -> derek Action: derek is back (gone 08:18:36) i think we should implement transactions another point to waiting until data is complete so the client can control it if we have a NOT NULL field and you forget to set it the object is not saved in the db at all silently which is not good imho agree so we need an explicit action if we are going to do an explicit action whether we call that flush () or commit () doesn't matter much in this point imho it should be done like a transaction without the rollback for the time being neilt: with the possibility to change multiple records of the same table within a single transaction? not right now ok something like connection.commit() simply flushes all objects that are dirty? yes hehe sounds like a job for a rainy sunday afternoon ;) or two of them :) or five :) another question yes what if the client disconnects without a commmit then all dirty objects are tossed out should be like an implicit rollback ok i think like all dirty objects are only held in memory and when we do commit() the sql statements are generated if the backend supports transactions we start a transaction cool do all INSERT's and UPDATE's and do a DB COMMIT if backend doesn't support transactions we don't do START TRANSACTION and COMMIT and pray instead :) :) we need start/end transaction for each type database if they support it then the code executes, if not it just returns so we can handle different db strategies yes well that's a todo like number 10 iirc we will talk about that again we need a quick fix for ra3vat imho yes how will that work in geas/src/geas_internal_classes.h comment out all lines that contain ODL_PROP_NOTNULL or maybe ok i just commited a quick fix NOT NULL is ignored for the time being so after database rebuild you should not get any more error messages you must drop the database and build up again i hope it works i must leave now :( will be back in 2-3 hours l8r reinhard (rm@N802P023.adsl.highway.telekom.at) left irc: Those who say it can't be done should never stop those who are actually doing it. reinhard (rm@N802P023.adsl.highway.telekom.at) joined #gnuenterprise. Nick change: reinhard -> rm-away Nick change: derek -> dnWork Action: dnWork is away: working bbl ra3vat (ds@195.239.66.37) left irc: Client Exiting neilt or rm-away found a neat tool last night http://www.dwheeler.com/flawfinder/ tried the RATS tool as well mentioned on that site since it works w/ python (it doesnt seem to work btw) but ran some of the geas code thru it and it had a very simple print-out "you have to be kidding" then it falted and you had to reboot :) s/falted/faulted/ :) no but it did return some possible security holes which I thought might be important seeing as geas is a server and all :) cool RATS however pukes on our python code :( so I cant scan it yet neilt (neilt@dialup-166.90.69.174.Dial1.Chicago1.Level3.net) left irc: Ping timeout for neilt[dialup-166.90.69.174.Dial1.Chicago1.Level3.net] dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. jamest (jamest@hobbes.math.ksu.edu) got netsplit. dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ jamest (jamest@hobbes.math.ksu.edu) returned to #gnuenterprise. #gnuenterprise: mode change '+o jamest' by forward.openprojects.net #gnuenterprise: mode change '-o jamest' by ChanServ!s@ChanServ sacha_ (sacha@chime-a-25.conceptual.net.au) left irc: Client Exiting sacha_ (sacha@chime-a-25.conceptual.net.au) joined #gnuenterprise. sacha_ (sacha@chime-a-25.conceptual.net.au) left irc: Client Exiting anil (anil@host-64-110-96-159.interpacket.net) joined #gnuenterprise. Mr_You (rouzer@mankind.boredom.org) joined #gnuenterprise. anil (anil@host-64-110-96-159.interpacket.net) left irc: Read error to anil[host-64-110-96-159.interpacket.net]: EOF from client chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) joined #gnuenterprise. jbailey (we-refuse-@fencepost.gnu.org) joined #gnuenterprise. jamest (jamest@hobbes.math.ksu.edu) got netsplit. dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. jamest (jamest@hobbes.math.ksu.edu) returned to #gnuenterprise. jbailey (we-refuse-@fencepost.gnu.org) left irc: leaving neilt (neilt@dialup-166.90.69.174.Dial1.Chicago1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ ello dudes neilt: how goes the article writting? writing even it goes well, not much progress today will send out outline tomorrow and have tohave article done by the 10th Nick change: rm-away -> reinhard dneighbo (dneighbo@mail.libertydistribution.com) joined #gnuenterprise. hey dneighbo hello I think I've got the golfing bug now I know what I'm asking my dad for xmas heh he and my brother play golf all the time brother is good brb reinhard (rm@N802P023.adsl.highway.telekom.at) left irc: Those who say it can't be done should never stop those who are actually doing it. wow hp bought compaq now that is a sign of the industry and its slump reinhard (rm@62.47.44.55) joined #gnuenterprise. yeah when you can only afford to pay $25 billion for a company it shows you're pretty weak financially :) well they will kill like.. 7k-17k jobs I think Mr_You well they buy each other then in 3 months with sales still dismal and that work force cut they consume dell and repeat the cycle at which time they are ripe for intel, sun, ibm or microsoft to buy of course sun and intel would lack the cash and ibm and microsoft would fear anti trust so aol-time warner would buy via default oh man as they have proven you can own the whole damn world as long as you dont own more than 'one' of any industry since they dont own a computer 'hardware' company this newsforge article just pointed something out.. they did do ANY PR for buy rofl they wouldnt have to worry about antitrust :) from geas-skeleton.c #ifdef STUPID_CRAP rofl reinhard i have some treshna code comments that go this is the proper way to do bracketing er // this is the proper way to do bracketing followed by this comment // no andru's way is only way to do bracketing all other ways are shite lol in geas there are comments like /* hi mom */ rofl hehe i like code that is funny yeah me too not being 'english' though the shite i read as shitty it's the only pace we can really act gooy and get away with it but even more i like code that works :) goofy even chillywilly that and irc oh yea.... :P dneighbo: the outline is basically how it will get sent to get a first impression the one in cvs that is you get any feedback form the editor yet? i will send it tomoroo morning ooooh chillywilly: have not sent it yet ;) gotcha damn I am tired Action: chillywilly wants AA fonts with his gnome http://philrsss.anu.edu.au/~josh/gdkxft/ jamest (jamest@hobbes.math.ksu.edu) got netsplit. dneighbo (dneighbo@mail.libertydistribution.com) got netsplit. reinhard (rm@62.47.44.55) got netsplit. dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] #gnuenterprise: mode change '-o neilt' by ChanServ!s@ChanServ #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. jamest (jamest@hobbes.math.ksu.edu) returned to #gnuenterprise. dneighbo (dneighbo@mail.libertydistribution.com) returned to #gnuenterprise. reinhard (rm@62.47.44.55) returned to #gnuenterprise. jamest: thanks for the links #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ jamest: thanks for the links to flawfinder however that program is not fulfilling it's purpose well most of it's output is false alarm it does not more than grep the source for certain keywords it considers harmful for example the "access" member of some structure (probably dirent) but we have a variable named access so it barks everywhere we use that variable brb chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) left irc: Read error to chillywilly[d42.as11.nwbl0.wi.voyager.net]: EOF from client neilt: quick question on your view about directory structure i would like to have all our own sources in the src/ directory ok from my understanding the lib/ directory is for supplemental "foreign" libs like md5 chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) joined #gnuenterprise. or if we would ship libuuid with geas (which would be a good idea anyway imho) but our own stuff belongs in src/ even the gcd parser and classdata.c makes no difference to me how we do the directories hmmm, can't tell much difference with these AA fonts as it seems like i partially rewrite the parser now anyway as long as the README has an explaination would you be ok with putting new code under src/classdef ? i dont understand why classdef is in its own dir except that it is used by the tools so you would move tools into src/ also eventually ok and if i was brutal i would say we need another subdir in src for all that files that implement the corba interface connection.c, dataobject.c etc. thats was the only reason i did not like src dir was because of all of those files neilt: i see us going through the idl files like we did for gcd syntax and remove all unused stuff stupid question one day but the new drivers for forms are pretty nice like tomorrow? and then redo the implementations any chance we can reuse them in geas? dneighbo: new geas dirvers ? er forms wait i said forms for geas? geas only supports mysql and postgres and the support is evil i.e. to add new functionality or add aditional drivers is painful dneighbo: you know anyone at IBM it is very simple in forms currently neilt : honestly i do we have full copies of DB2 currently dneighbo: why dont you get us copies of DB2 and are workign on ways to get things tested better on DB2 cool but beyond on that do we have source for DB2 or just x86 binaries a. maintianing two sets of drivers is painful b. geas drivers are really ugly currently im just thinking rather than rewrite geas drivers to be more clean why not write wrappers for forms drivers and just reuse them? as write one wrapper dneighbo: i don't think we can use python code in geas for db access not enough performance we are already too slow and voila all forms dbs are instituted yeah exactly what neilt said although i dont know why reinhard : hmmm i dont think thats an issue enterprises are using java and it is MUCH slower than python dneighbo: plus if things are too slow its fairly easy to convert things to C from python on a need basis geas needs much more functions in the db driver than forms dneighbo: java being slower depends, it can be compiled to native code reinhard: why? that does not exist for python technically 2 tier and 3 tier should be operating the same as much as possible for example i don't think that forms driver can create tables in the background reinhard: hmmm i think we support that if not we have to for designer basically the only thing the forms db drivers will not be able to do are things that are not provided via an API but geas drivers couldnt do that either if 2 tier and n tier work the same then we lose all object orientation hey neilt: i slightly mis stated that you guys ever look at swig we can build a C API to python code so we could share things that way the quick point is we have lots of working drivers in gnue-common it seems silly not to reuse our own code dudes swig would work great if python is too slow then those gnue-common drivers need to be in C dneighbo: yes, that effort should have been put into supporting geas :) not building unneeded drivers :) but i hate to see spinning wheels on rewriting working things for 'performance' at this point fwiw: im playing devils advocate here not saying its doable or not just seems logical db drivers in geas are not very painfull you basically copy from the manual using the c interface routines neilt: i REALLY hope you are kidding i dont understand the problem um problem i write a new db driver and i have to change like 5 files nope not any more and do a bunch of conditional ifdef statements not any more neilt changed that yeah one file well that helps things tremendously i would like to see us have ONE driver system period i dont care if its in scheme, java, python, C, monkeycode i vote fo monkeycode but its silly for us to not use shared code Action: chillywilly would vote for monkeycode, but that's ximian's expertise the coders for gnue-forms and geas can argue on who's code to use I vote for raw assembler but i think the drivers should be in gnue-common we should make our very own libGDA thingy (but better) and be usable to any gnue application or external application i.e. a libGDA done correctly :) have you preformance testing the python drivers yeah! neilt: i have not Mr_You: we use C because i ONLY said use gnue-common ones because a. i thought geas drivers were still a mess C combines the speed of assembler with the flexibility of assembler b. they support many more dbs reinhard: I was joking ;-) reinhard rofl oh hehe reinhard: and you forgot C code has an inheritly higher error rate than assembler lol C is like my bitch or somethin C the other white assembler. its so funny how biased billy grahm is dneighbo: there is a bad taste in my mouth wrt db drivers in python reinhard : thats easy fix quit eating them you have to fight the python binding for all the databases jamest are you around? to weigh in on this? didn't jamest have issues with popy for example oh this is somethin else or are you hacking the drivers with assembler dneighbo: what drivers dneighbo: the real difference between assembler and C is are begin hacked in assembler reinhard that was pygresql C is more portable :) dneighbo: ok python has assember drivers reinhard that is pretty true that sucks i think some of that might have gone away with new DBSIG stuff in python jamest...... Action: chillywilly pokes jamest neilt: read your outline and agree thanks huh python is for weenies someone told me that once give it a rest! you bg python weenie :P s/bg/big jamest: so now are you publicly admitting you like weenie? sheesh python weenies really know how to hold a grudge goats are one thing, but ...... hehe since jcater isnt here i will 'quote' him while he cant defend himself C is for old austrian hackers who cant seem to find thier razor.... lol huh? ouch that's not nice ;) Action: reinhard goes to look up razor reinhard im kidding of course reinhard: what you shave with reinhard : razor == tool you shave your beard with as for the python/C thing I'd be happy to benchmark if you like reinhard: what d they call it over there? s/d/do ah chillywilly : its cold there they probably all wear beards :) oh the performance of python vs C drivers but I won't be able to before the weekend Action: reinhard is amused how others imagine how he looks as for our new drivers you have reinhard : once you said you have long hair and beard like stallman and maddog? or am i confused? dneighbo: probably confused no beard datasource - tied to the native driver resultset - the results of a query recordset - a single record resultsets contain numerous recordsets reinhard: but you are a long haired freaky person? if the geas drivers are at all similar then sharing might be possigle possible chillywilly: yes somehow not very long though to the shoulders we have a geas db api so all of the drivers call the exact same functions and all the driver is is the glue to make it work and take care of differences but we are building ways to make performacne imprvements in the db driver rather than design to the lowest common denominator why not just make our own enterprise strength db abstraction layer with all sorts of bindings? so some functions use different ways on doing it depending on the db for example geas char = varchar in postgresql not is a char in mysql s/not/and ra3vat (ds@195.239.64.223) joined #gnuenterprise. jamest: if we had a geas driver for forms you would not even need to worry about db drivers they could all be in geas or use geas code thing is they are still in love with with 2-tier Action: chillywilly hides they will be until they try to use it in multiple db situations or multiple locations or with complicated structures or on a large scale basis or . . . . . . :) 2-tier is so yesterday :P but it is simple to understand you dont have to solve problems, just write code you shouldn't have to worry about how thngs are stored you can always create SQL to do anything ya cobol is so yesterday, why would anyone code in cobol why not they don't if 70% of your code base is in cobol, you would code in it and people do with mainframes this discussion starts to become unproductive it is still not very common a lot of the world works on mainframe code written 15 years ago reinhard: yep, just venting ya and then they wonder why we get Y2K issues neilt: anyway its very depressing not being shown at tradeshows and not being able to run example code they can't stop us from writing the worlds best multi db driver chillywilly: http://cobolworld.com/1_2press.html just after we finished the world's best classdefinition parser ;) :) i have to go in 30 minutes hold up neilt: my point was 2-tier may be so yesterday but it's still in wide use reinhard: did you have anythink thing neilt: no part of the problem is there is an us them mentality i work on parser and i learn a lot cool fwiw: i spoke of geas features more than forms features on the show floor reinhard: how long until something is in cvs i just dont demo geas as its not 'attractive' visually if drivers existed to use geas with forms neilt: 2 days then i would use geas in the demo reinhard: cool i only have 1-2 hours per day actually for coding dneighbo: this is not geas against forms dneighbo: my point exactly not at all problem is without they driver peopple ask what geas really is as in current state with forms its noting short of db abstraction dneighbo: we talk about geas, get people excited with geas give me a gui tool to build and manage gcds and i will demo it dneighbo: then write 2 tier drivers demo'ing emas w/ gcds is not impressive :) dneighbo: it is more than that it is OO interface to the data chillywilly you are missing the piont i KNOW that you KNOW that now tell the forms geas driver that :) yes, no eye candy frankly from the shows there is a large 2 tier market out there as well so really we need to support both IIRC old forms works with geas to the extent that it could n-tier should be our preferred jamest: yes but it had no way to find what objects where available when i say geas-driver if you knew the objects they you could reference them its a two edge problem some things need to be implemented in geas jamest: the object api has been there for months and the forms code needs to use those implementations since before andrewm left there is no reason they cannot talk to each other right now but to me this is all superflous to the immediate dicussion of dbdrivers even if we have workign geas driver forms is still going to support two tier which means it has to have data access w/o geas i think that its silly not to use same code if possible maybe its not possible i like the idea of own libGDA as part of gnue-common or outside of gnue-common language its written in etc is debatable return of the gedi chillywilly pretty much its sounds as though there is a divide here though :( one that probably wont be crossed dvide? chillywilly , im no communications expert but i see 2 tier vs n tier we just want you you use GEAS...we feel left out :( sql vs oop and C vs python well yeah as underlying divides OO is good i dont think we have divide I believe STRONLGY in OOA/OOD im thinking they need not be mutually exclusive i think we would just like to see forms support our other product STRONGLY even but maybe i am incorrect neilt : so would i :) that has really been ignored by everyone except for reinhard myself, and cw as im getting ready to write invoicing and g/l for a client right now :( i like what i see for the most part in gcd's but have no way to easily functionally use them dneighbo: exactly we need two thinks thnks well there's no decent methods....which is my fault things as you know i attempted to make entry screens etc for them methods and forms driver which are being worked on and could only do most basic forms because of inheritance not supported in geas driver of forms still but i think thats still a different issue than dbdriver inheritance should work fine maybe im wrong? jamest: i am not complaining about people, just venting my frustration at priorities chillywilly it does in GEAS, not in geas driver :) huh? dneighbo: why doesn't it work in the driver? i thought geas hid inheritance just curious you just use the object chillywilly because when jade wrote the driver nearly a year ago so gnuef wouldn't need to know anything about that reinhard and i just did away with inheritance anyway if it happen to inherit some things then you get that too object introspection did not work at least some forms of it so the geas driver was severly crippled ah yes that's another area basically it just acts like another sql db neilt: no inheritance? wrong words you talk to geas objects, but really can only do what a sql db could do no modules inside of modules where i had issues building forms no classes inside of classes neilt: aaah, that's fine though was the concept of say you have an org class and you descend from that we have a flat namespace now to make a vendor class yes that nested crap is unnecessary or somethign it is not gonna make a big diff i could make the org in forms but not the vendory Just like I can use Java to my heart's content and never have to deal with innner classes unless someone can explain the java dneighbo: anything that you do not like in GEAS right now we can change in a few days so give us a list based on current code excluding driver and methods 94 type person 95 { 96 char name_prefix<8>; # Mr. Ms Dr. 97 char name_first<25>; # given or first name 98 char name_middle<25>; # other name(s), not surname or first 99 char name_last<25>; # surname or family name 100 char name_suffix<8>; # Jr. Sr. III 101 char name_informal<25>; # nickname102 103 char work_title<25>; # president, vice president104 105 person::address [] address; # home, work address 106 person::comm [] comm; # email, phone, fax, web107 108 #text get_full_name(); # Mr. Frank N. Furter Sr. 109 #text get_polite_name(); # Mr. Furter110 }; since we're on the subject of the geas driver..... like this class i dont know how to make this work in forms masta what exactly is the issue again? i can make all work but how do i amk personn::address personn::comm something I've been looking at doing (yes, believe it or not I have been looking at this :) work in forms hmmm maybe someone should be the forms-geas liason and learn both it ditching orbit-python in favor of omniOrbpy jamest yeah or at least enough to knwo th eissues i vote for not DITCHING but SUPPORTING both as it should make a difference to geas which orb I use right ? even if that means orbit falls to untimely death it sounds better to support another orb than to kill one :) jamest: actually it should not jamest: not really that's why CORBA is a standard IIOP jamest: my understanding is that for client to server it HSOULD NOT make a diff was also wanting to look at an orb abstration layer to change from orbit to omniorb in server or support both any orb should be able to talk to any other orb in the server it would mean some work in geas ii understand correctly so that the reports engine can plug in ddifferent interfaces (omniOrb, xml-rpc, etc, etc) but client to server has been done alread jamest: we also want the same thing andrewm wrote a java client using jacorb i think to geas so there's another thing to share i LOVE that idea but am thinking it might be better off in different phase i.e. lets get corba workign I'm all for sharing however I'm much more productive in python than C (I love C btw) so I've been taking the approach write in python, profile, rewrite slow stuff in C dneighbo: yes, the rmote method thing is long term abstraction layer jamest: I have no beefs against python it looks very fun i personally prefer python but im all for what works best for the job if C is correct tool lets use C so lets summarize here we can share db abstraction layer and OO IPC abstraction layer right? if java is correct tool lets use java etc I just get the feeling that any python work I offer up is rejected on the basis it isn't C from the start nah w/o any profiling being done at all to justify rejection if it works and isn't too slow why change it? i definitely think there is a divide, and really believe it based on this last hour :) agreed there Action: chillywilly wants to bridge the gap im not sure as i said before if its a python vs c divide 2 tier vs n tier or which i think we need to close it though and get on same page at least we still all agree that masta is a PITA right? masta good jamest: you all are definitely NOT divided on that :) right cool no I am opne minded here I am not saying everything should be C....I believe in the right tool for the job too....I just want us all to play together now i dont think we have a divid, still you can start sending your complaints to free_da_coders@gnue.org everyone has had their head down working codeing etc we need a reality check neilt let me rephrase there's no divide...we just don't understand all the areas dneighbo: goats and ash can't handle that kind of mail load when i say divide i dont mean flaming personal hatred like gnome is having (divide) I don't get forms issues at all cause I haven't even gotten to play with it yet i simply mean not on same page divide and an understanding on GNUe prrorities well, I'm willing to dive into geas and make it play nice w/ python stuff (and vise versa) well I think we just recognized 2 common areas as you state mostly incurred from not looking up on current tasks :) i would like to install dcl on ash and do gnue like a real project er corporate project ;) does that mean paychecks? :) so we can see the 'projects' and 'todos' of all projects you are now da real masta instead of fishing in todo files Action: chillywilly bows to dneighbo I like that idea i realize with volunteers this just might not be feasible Action: neilt goes to check mail box for new paycheck neilt : um its small enough i just tape the quarter to an index card for payroll, checks are nt in our budget cool you lost mine, so send another one please ok I have one more idea for code sharing and that is a plugin system for various scripting langauges rofl jamest said he also wanted this chillywilly yip this is another 'shared' area possibly you can do triggers i various languages and we need methods in various langauges dneighbo: i still did not receive it, maybe just to be sure, you can send it again neilt : sorry must file in triplicate a lost quarter on index card form with USPS Action: neilt thinks its so easy to tripple ones salary in 6 to 8 weeks we should hear back from them 3 * 0 = 0 btw: in traditional corporate stewardship hehe oh crud i paid 6.50 to insure that quarter ;) hehe hehe dneighbo: could you send the tracking number please chillywilly: I started writting a design this weekend for trigger system roflmao jamest: ya? trying to take into account geas needs nothing huge yet ok and no code as I'm still working on the drivers in gnuef jamest: when it gets huge can we call it MS Trigger System i think MS == More Stuff dneighbo: as long as we change it every 6 months look in gnue-common/doc for the trigger spec needs tons of work and input ok neilt: how else do you expect me to keep sending quarters? exactly http://lxr.gnue.org/gnue/source/gnue-common/doc/TriggerSpecifications.txt oooooh sexy brb I am sick of this thing flashing ever snce I changed theme xchat is scting mental fwiw: on funny note, i remember insisting on lxr and someone said why lxr its pretty much cvsweb slowly i think people have grown to use it i didnt really realize how much until at the hotel room jcater and were researching a problem which ended in 'i cant solve this w/o cvsblame' anyone running debian 2.2r3? oh oh me me me in need to know the stock kernel installed uname -a 2.2.19 I think as I have 2.2 and it doesn't support the nic I've got poop i dont have here only at home potato? you have an account on that machine unless someone knows where to get iso's of woody iirc you can ssh and verify potato has 2.2.19-preX while I'm here what exactly do we require for geas/forms interface as I'm obviously missing something brb chillywilly_ (danielb@d42.as11.nwbl0.wi.voyager.net) joined #gnuenterprise. jamest: we need derek to understand the api and the driver to support the api aaaah chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) left irc: Killed (NickServ (GHOST command used by chillywilly_)) since i cant run the gui i dont know what we need Nick change: chillywilly_ -> chillywilly now for a touchier question Action: chillywilly foams at the mouth is there any advantage to looking at things the the enhydra API and try an accomidate their needs as well what is enhydra ? as the basis for later intergration with other systems wassat? jamest: based on what i have seen http://www.enhydra.org for forms i think it may be beneficial to support other application servers specifically jboss a java based app server kinda like jboss enhydra websphere kewl though i think we should focus on geas first none do what geas does and foremost agreed there what about that thing that webworks has they have an app server too but several people asked if forms would support other appservers than geas my quick two cents a. all our tools should work together or apart and play well with others (this would mean we should) b. most people using enhydra, jboss, etc already have a solution for data entry so will not be wildly interested in forms (imho) (this would mean not high priority) the only advantages i really see is if people had lots of jboss etc code but wanted to move away having this support woudl make that easier (not real likely to be this situation) or if we wanted immediate appserver out of box with user base at this point i think geas can do most of what the otehrs can anyhow im curious to see reinhard and neilt's opinion on this? Action: dneighbo guesses they all disappeard nope we dont even have geas working right now ok in the geas driver for forms you want to be able to call geas object methods right (this is the major shortcomming) so i dont see any need to even think about another appserver unless there is some design issue that should be considered jamest: right using the corba idl but i agree with a in principle and b also reinhard_ (rm@N802P023.adsl.highway.telekom.at) joined #gnuenterprise. derek (derek@cpe-24-221-112-50.az.sprintbbd.net) joined #gnuenterprise. reinhard (rm@62.47.44.55) got netsplit. dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. derek: you here dnWork (derek@cpe-24-221-112-50.az.sprintbbd.net) got lost in the net-split. reinhard (rm@62.47.44.55) got lost in the net-split. Nick change: ajmitch_ -> ajbusy ra3vat (ds@195.239.64.223) got netsplit. chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) got netsplit. jamest (jamest@hobbes.math.ksu.edu) got netsplit. dneighbo (dneighbo@mail.libertydistribution.com) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] #gnuenterprise: mode change '-o neilt' by ChanServ!s@ChanServ #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ jamest (jamest@hobbes.math.ksu.edu) returned to #gnuenterprise. dneighbo (dneighbo@mail.libertydistribution.com) returned to #gnuenterprise. ra3vat (ds@195.239.64.223) returned to #gnuenterprise. chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) returned to #gnuenterprise. weee! later all neilt (neilt@dialup-166.90.69.174.Dial1.Chicago1.Level3.net) left irc: bbs and hour or so sigh, once again #debian fails to even recognize that I asked for help :( chillywilly: you know where I can get boot floppies for woody? yep any debian mirror care to share the info or should I just install w2k http://http.us.debian.org :) etc. http://http.us.debian.org/dists/testing/main/disks-i386/current/ to be more specific assuming you want inhell (i386) that help dude? thanks! slurping copies now no prob wish me luck, this new dell isn't happy with GNU/Linux only thng that sucks with those disks is no pcmcia support no problem for me though ya I just need a 3c905 driver that sucks less (i think/hope) I figured that I would just roll my own kernel oh wait but you don't have any network connection :/ that's why I love kernel debs assuming you got a bunch of simlar machines but ya know whn you "assume" you make an "ass" out of "u" and "me" :P Action: chillywilly bables onwared... babbles onward i missed neilt i got called to meeting so we resolve anything here? sure ;) yeah we are absolutely 100 positive that masta is a PITA 100 = 100% Action: chillywilly likes da masta as chillywilly has masta -like qualities himself seriously, i think we resolved to work together ToyMan (stuq@c5300-1-ip123.albany.thebiz.net) left irc: [x]chat me too I really wanna be able to buils some apps soon ;) build rofl Mr_You (rouzer@mankind.boredom.org) left irc: ircII EPIC4-1.0.1 -- Are we there yet? jamest: whew ? i thought 100 positive mean 100 people were gonna jump me for being PITA much better to hear only 100% of 10 :) 10? I thought we were more like 6 -:- [Users(#gnuenterprise:11)] [@gnuebot ] [ dtm ] [ bigbrother] [ ajbusy ] [ dres_ ] [ reinhard_ ] [ derek ] [ chillywill] [ ra3vat ] [ dneighbo ] [ jamest ] oh ppl in the channel although i suppose -1 for gnuebot -1 for bigbrother -1 for dneighbo -1 derek so 7 guess you were closer to right :) guess so though mryou and toyman were here earlier nite all reinhard_ (rm@N802P023.adsl.highway.telekom.at) left irc: Those who say it can't be done should never stop those who are actually doing it. and i have been known to think me as a PITA I don't see it like that I see you as our leader ;) Action: dneighbo feels sorry for chilly why? it would be different if you were self appointed....uuuuh ok I admit it you're just a PITA ;P rofl lol actually, i asked him for the help Action: jamest thinks he musta been on some really wicked drugs that day hehe I am only joking I love da masta really I do you ever notice when they show good looking girls in the old star trek theyshow them all fuzzy looking? l8r chillywilly (danielb@d42.as11.nwbl0.wi.voyager.net) left irc: bbl jamest (jamest@hobbes.math.ksu.edu) left #gnuenterprise. jamest (jamest@fh-dialup-202033.flinthills.com) joined #gnuenterprise. Faster than a cray super computer! More powerful than a GNU/Linux Cluster! Able to bind data in a single line of code! Look! In the high back chair! Its a hacker! Its a coder! Its GNUe Man! Ah, but "Can he do all that in emacs?" is the real question. rofl should defintely change More powerful than a GNU/Linux Cluster! to More powerful than a Emacs! sigh, back to calc for a bit before class, then it'll be GNUe time! did you uncover more buglets in the new code? jbailey (we-refuse-@fencepost.gnu.org) joined #gnuenterprise. um yeah quite possibly must go home and test at home i have to get invoicing solution by end of month so will be using 2 tier gnue a lot i think how hard to make vpn for postgres? hi jbailey long time no see Heya dneighbo how is life? Am I reading right, like invoicing in gnue? Busy. I'm in the middle of looking for a job right now. neal very kindly spent the last 3 hours or so helping me with my resume. just remember crack-smoking-free-software-loving-hippie is spelled with a C not a K Action: jamest messed that one up on his last resume rofl Action: jamest is away: I'm busy jbailey yip gnue invoicing Is *that* why you can spend so much time in the #gnue channel. ;) dneighbo: Did you guys wind up doing multi currency, or just postpone that for now? its there in gnue-config iirc im doing custom invoicing as gnue official invoicing isnt ready so im using architecture to roll my own Nice as a proof of concept though. we have contact manager that is somewhat functional with minor crm capablities in cvs i will start to release such things on my personal page i think as 'non official' gnue applications I should get around to figuring out how to build gnue. My last attempts were not succesful. what did you try to build? if you want two tier should take 10 minutes or less to download and build (with high speed connection) if you have to get the dependencies maybe an hour but you run debian so all dependencies should be as easy as apt-get install dependencies The problem is that I didn't really know what I was trying to do, so I need to spend some time learning more about what I should try and build.. I get a DSL connection next week (and hopefully a computer RSN) so I'll look into it then. at the show floor we were installing dependencies on woody machines in about 30 minutes and having configured and running in 45 awesome. cool I should run. My web cafe time is almost up. geas also builds pretty easy after dependencies ok later bibi im off to run home sa well l8r dneighbo (dneighbo@mail.libertydistribution.com) left irc: [BX] I see your BitchX is as big as mine! jbailey (we-refuse-@fencepost.gnu.org) left irc: leaving Mr_You (rouzer@mankind.boredom.org) joined #gnuenterprise. nickr (nick@63.163.68.113) joined #gnuenterprise. ajbusy (ajmitch@p44-max5.dun.ihug.co.nz) left irc: Ping timeout for ajbusy[p44-max5.dun.ihug.co.nz] #gnuenterprise: mode change '+o derek' by ChanServ!s@ChanServ ajbusy (ajmitch@p44-max5.dun.ihug.co.nz) joined #gnuenterprise. ******************** greeting for reinhard ********************************* using fields of type "date" in query gives me this: __conversion_rule.sys_mod_user FROM currency__conversion_rule WHERE (upper(rtrim (currency__conversion_rule._to,' '))=upper(rtrim('63e9b2362bd94a2fbf7a248ac1b9e5 3c',' ')) AND upper(rtrim(currency__conversion_rule."end",' '))=upper(rtrim('200 1-09-05',' '))) ORDER BY currency__conversion_rule."end" DESC' PostgreSQL reported a fatal error PostgreSQL Error message: ERROR: Function 'rtrim(date, unknown)' does not exist You may need to add explicit typecasts (7) disconnecting... reconnecting... and typascing via ::text does not work in this case also gnue=# select rtrim(begin::text, ' ') from currency__conversion_rule; ERROR: Cannot cast type 'date' to 'text' ******************************************************************************* neilt (neilt@dialup-64.157.58.95.Dial1.Washington1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ hi neilt hello neilt (neilt@dialup-64.157.58.95.Dial1.Washington1.Level3.net) left irc: ra3vat (ds@195.239.64.223) left irc: [x]chat ra3vat (ds@195.239.64.223) joined #gnuenterprise. chillywilly (danielb@d132.as13.nwbl0.wi.voyager.net) joined #gnuenterprise. Nick change: ra3vat -> help help (ds@195.239.64.223) left irc: [x]chat ra3vat (ds@195.239.64.223) joined #gnuenterprise. Nick change: ajbusy -> ajmitch