[00:00] Last message repeated 1 time(s).
<dtm> greetings ladies
<derek> you have a non dataware form you can post for me?
<derek> or you want me to create one
<derek> im extremely lazy today :)
<dtm> chillywilly: hitting it?  you'd better be nice and careful with that!  that's your brand new shiny Christmas crackpipe, Dan!
<Harald1> wait, I'll send you mine. does dcc work for you?
<dtm> You foolz all need to enable java browser support and roll on over to my site, http://ambcomm.com.  Dont click anything at all -- just sit there and wait to see what happens.
<dtm> Just indulge me and help me test it for a moment.
<dtm> RIGHT NOW
<chillywilly> don't have the java plugin
Action: dtm levels a switchblade at chillywilly
<chillywilly> and it takes too long to download
<dtm> AH CUT CHOO, FOO'!!
<dtm> hmm
<derek> Harald1: nope you can put it on a web site
<derek> or you can email it to derek@gnue.org
<dtm> chillywilly: Well, Jack, maybe we can work out a deal.
<derek> actually you can try dcc but i dont think it will work
<dtm> chillywilly: You're not gonna help me out, mah nigga?  :)  You gots mah back?
<dtm> chillywilly: oh well.
<derek> i wont download java :)
<dtm> i know it works.
<dtm> i know it right here
Action: dtm taps on his heart
<Harald1> derek: you should receive an email in a few moments
<dtm> derek: sucka!
<Harald1> don't mind the german field names
<derek> ok
<chillywilly> derek: are you happy that we will have a GOAT modules in GEAS ;)?
<chillywilly> GNUe Object Access Translator
<derek> sure
<chillywilly> (hey, it wasn't my idea)
<chillywilly> it's in the architecture guide reinhard did
<chillywilly> and I guess it is now tie for return of the GEDI
<chillywilly> s/tie/time
<chillywilly> see forms, common, etc. are too boring and not recursive acronyms, this is GNU you might as well be dead if you're not going to have fun with recursive acronyms ;)
<dtm> chillywilly: do you seriously have a component called GEDI?
<chillywilly> yes
<dtm> cool.
<chillywilly> it si planned anyway
<chillywilly> s/si/is
<jcater> we should call the reporting engine NIAGRA
<jcater> that's a cool recursive name
<dtm> yeah well just ocme up with acronyms -- we'll figure out  their meaning and what to implement behind them, later
<chillywilly> they are all figured out though
<jcater> Niagra Is A Great Recursive Acronym
<chillywilly> tp://www.gnuenterprise.org/~reinhard/whitepaper/whitepaper.html
<chillywilly> er
<chillywilly> http://www.gnuenterprise.org/~reinhard/whitepaper/whitepaper.html
<chillywilly> that's lame
<jcater> (it's not a modest acronym though :)
Action: chillywilly troutslaps jcater
<jcater> as opposed to gnu???????????
<jcater> btw
<chillywilly> Hurd s cool because it is double recursive
<chillywilly> s/s/is
<jcater> our acronym is an oxymoron
<dtm> hurd++
<jcater> it's a self-defeating acronym
<chillywilly> Herd of Unix Replacing Daemons
<dtm> jcater: self defeating in what way?
<jcater> GNUe's Not Unix Enterprise?
<jcater> but we ARE unix enterprise
<chillywilly> that's called a joke
<jcater> in GNU's case maybe
<jcater> but not ours
<jcater> :D
<chillywilly> GNU si very unix-like
<chillywilly> s/si/is
<derek> i thought we were double too
<derek> GNUe
<chillywilly> it's supposed to ne contradictory
<chillywilly> s/ne/be
<chillywilly> we are
<derek> GNU Nasty Ugly equines
<derek> crap jcater i meant to bug you about
<derek> on new record
<derek> trigger support
<derek> hell for that matter all 'datasource' triggers :)
<jcater> yeah
<jcater> there are 0
<derek> just on new record used to exist *iirc* and now it doesnt
<jcater> we need > 0
<derek> here is my plight (for your opinion)
<derek> all my tables have
<derek> created on, created by
<derek> modified on , modified by
<derek> the on's i can leave blank (iirc) and the getdate() function is tied to the field so that will auto popluate (though that is probably bad) as its not portable
<derek> but basically i need a way to do this for records that have changed
<derek> probably on an 'on-commit'
<derek> any ideas
<derek> looks to be pre-commit post-commit triggers at 'form' level
<derek> iirc
<derek> but i dont know how i would change the records that had been 'updated'
Action: derek thinks i scared him back into hiding
<jcater> for your contact manager, we used pre-commit irrc
<jcater> irrc = I righteously recall correctly
<jcater> which isn't a good solution
<jcater> but it is a solution
<derek> i dont think we did it in my contact manager
Action: jcater will look at datasource-level triggers tomorrow or monday'
<jcater> yeah
<jcater> your primary key sequence
<jcater> remember
<jcater> but it is an awkward way to do it :)
<jcater> but I'll do the datasource-level stuff next week if you need it
<jcater> shouldn't be hard now that jamest has the common-based trigger system
<jcater> I would offer to do it tonight
<jcater> but it's past midnight :)
<chillywilly> is that common based trigger system supposed to support multiple languages?
<jcater> yes
<chillywilly> ok..
<derek> yes i see that now
<derek> btw is taht still needed
<jcater> ?
<derek> or does new datasource handle that
<jcater> oh, yeah
<jcater> I think jamest did add an entry-level "autosequence" thingy
<jcater> haven't used it, but I think it works
<derek> i will have to hit him up on that
<jcater> chillywilly: the common trigger system defines namespaces
<derek> jcater otehr thing
<jcater> chillywilly: which is a small step from language-independence
<derek> i was getting datatype errors 
<derek> which i fixed by
<derek> str(blkName.WidgetName)
<jcater> chillwilly: of course, we haven't experimented with other languages
<derek> to get the value
<derek> shouldnt one be able to do blkName.WidgetName.Value
<derek> or something instead?
<jcater> we're looking into that
<derek> ok
<jcater> jamest did the other way for a reason
<jcater> a reason that jcater isn't sure of
<derek> just making sure i wasnt missing the magic incantation
<derek> as the 'notes' on triggers seemed to indicate discussion of
<chillywilly> what do 'namespaces' do exactly?
<jcater> there are still a few "gotchas", which is why we haven't released :)
<derek> getValue(blkName.WidgetName)
<derek> or
<chillywilly> well I know what they are for but I don't see the connection I guess
Action: chillywilly is dense sometimes
<derek> blkName.WidgetName.Value
<derek> or what ave you
<jcater> chillywilly: defines what methods/objects should go into the language instance's namespace
<jcater> chillywilly: and what they should be called
<derek> me thinks it might be cool to do
<derek> the .value
<derek> or support on fly type cast
<jcater> yeah
<derek> so you could do
<derek> bname.wname.value
<derek> or
<derek> bname.wname.AsString
<derek> bname.wname.AsInteger
<derek> bname.wname.AsFloat
<jcater> chillywilly: why do you ask?
<jcater> chillywilly: you have some needs?
<derek> but i suppose that isnt necessarily easier than 
<jcater> chillywilly: or just curious?
<chillywilly> because I thought that moght be a shared thing
<derek> str(bname.wname)
<chillywilly> but it is because it is on common right ;)?
<chillywilly> we have this thing we called GEMA
<jcater> derek: I have no problem with the As*  things
<chillywilly> GNUe Methods Adapter
<jcater> chillywilly: ok?
<chillywilly> which is FICTIONAL as of right now
<chillywilly> onyl exists on paper
<chillywilly> it's the multi-langauge methods thing
<jcater> well, we already have that
<chillywilly> supposed to be
<chillywilly> yes
<jcater> 80% or more of geas already exists
<jcater> it's called gnue-common
<jcater> :)
<chillywilly> I dunno if it is 80%, but it is up there
<jcater> we just don't have acronyms for everything
<chillywilly> yes I know
<jcater> :D
<chillywilly> GEDI == dbdrivers
<chillywilly> GEMA === triggers
<chillywilly> remote protocol adapter == gnurpc
<chillywilly> that's what I get from reinhard's document
<derek> Harald1: what was the problem with the form again?
<derek> it appears to function ok to me
<Harald1>  ok: start it, go to the multiline field, enter something, press return, enter something
<derek> if i press enter it goes to the next field
<Harald1> use shift+enter
<Harald1> in the second row, the latest character is after the cursor
<derek> not happening here
<derek> but i found antoher bug :)
<Harald1> ok seems the be windows specific, was newline under windows or linux two characters wide?
<Harald1> derek: what's the problem
<Harald1> ?
<derek> go to a new record in taht form
<derek> then just fill up the whole box with characters w/o hitting shift return
<derek> i.e. hold down a key until it stops (255) char i think
<derek> then try to delete one of them :)
<derek> it wont let you :)
<Harald1> derek: exactly
<Harald1> I'll submit "my" bug to dcl. btw I've already got a fix for it
<derek> i ONLY have this problem if i get to the end of the entry box (i.e. whole thing is full)
<derek> Harald1: in that case submit the bug
<derek> and email the diff :)
<derek> do we have an assignmetn from you?
Action: derek is bad at remembering and have only recently started trying to use DCL to manage this
<derek> btw: its kind of cool to see a german form (im guessing thats german)
<Harald1> assignment?
<Harald1> derek: yes indeed
<derek> we require all developers assign copyright to fsf
<derek> so we can properly protect GNUe 
<derek> it is a dual copyright deal so you would still retain full copyright authority as well fo ranything you contribute
<Harald1> derek: for me this happens earlier. about one and a half line before the end of the box
<Harald1> derek: ok, and how do i do it?
<derek> give me your email and i will send it to you
psu (psu@manorcon.demon.co.uk) joined #gnuenterprise.
<Harald1> derek: gruen-test@freenet.de
<Harald1> am I right: under unix \n is \013 and under windows \013\010 (or the other way around)?
Action: Harald1 is away: breakfast (back in about 10 min)
<psu> windows is both, unix is \013
<psu> and mac, just for the fun of it, is \010
<psu> standards are for wimps, obviously
<psu> New KC is up
<psu> http://kt.zork.net/GNUe/gnue20020330_22.html
<psu> announcements going to the usual places ;-)
<derek> Harald1: sent you paper work
Action: Harald1 is back
<Harald1> psu: thanks, that explains the bug
<chillywilly> hey, where's Sachas paper?
<derek> ?
<chillywilly> er, the one he was talking about
<chillywilly> BPML
<chillywilly> Business Process Modelling Langauge
<Harald1> derek: can I use another mail account for sending the assignment?
<chillywilly> nevermind
<chillywilly> http://www.bpmi.org/bpml.esp
<psu> http://www.openebxml.org/information/papers/sacha_ebxml_2001.pdf
<psu> sayeth the website
<derek> Harald1: yes
<chillywilly> cool
<chillywilly> Sacha wrote it?
Action: jcater is away: jcZzz
Nick change: jcater -> jcEasterBunny
<Harald1> derek: and changed file: should I leave it empty or should I put the file for which the bugfix is there?
<derek> psu what happened to our names having links?
<derek> in the kernel cousins?
<derek> either way
<psu> not sure
<psu> I know Zack is working on the code at the moment
<psu> trying to convert to using a "true" XSLT
<psu> instead of the perl scripts
<psu> may be a side effect of that
<psu> after all, it's so easy to break cvs ;-)
<derek> rofl
Action: derek scampers off to bed
Nick change: derek -> dnSleep
<Harald1> goodnight derek
<chillywilly> this is funky
<chillywilly> http://www.openebxml.org/realization/architecture/tarchitecture/tarchitecture.html
<chillywilly> is ebXML something that J2EE uses?
<psu> hmm
<chillywilly> www.ebxml.org times out here ;(
<chillywilly> psu: so sascha wrote that paper right?
<psu> IRC chat implies it - or at least part-authored
<psu> he was in apologising a few days back in case some of it sounded like a slam on GNUe
<psu> but since all he said was "er, not really any apps yet"
<psu> which is undeniable 
<psu> derek forgave him
<psu> probably group-authored if it's anything like most academic papers
<chillywilly> seems to be so
<chillywilly> er, only by sacha
<chillywilly> he did it for his doctorate or something
<chillywilly> why would it sound like a slam on GNUe?
<chillywilly> sounds like something we should read
<dtm> HOW ARE YOU GENTLEMEN
<chillywilly> it is basically business integration through document exshange stuufs ;)
<chillywilly> er, xchange stuffs
<dtm> so you foolz aren't even gonna look at my web site huh?
<chillywilly> that's it
<dtm> at all?!   :'-(
<chillywilly> nope, don't have the java plugin
<chillywilly> not going to download it either
<chillywilly> it is too big
<dtm> geez only one button on one page is java, chillywilly
<chillywilly> heheh
<psu> coz java is non-free
<chillywilly> well I was busy
<chillywilly> that too psu
<dtm> oh ok
<dtm> java is as free as anything.
<dtm> you can manually run the client for my site using blackdown if you want :)
<dtm> i even have a command with which to do that.
<dtm> you psychos  :)
<dtm> TEE HEE
<dtm> psu: how are you doing this fine evening, sir?
<psu> that sounds like trout-slappin' talk to me
<dtm> :-o
<dtm> I demand -- ah say -- ah deMAND satisfaction, suh!!
<psu> I'm having a happy easter morning
<psu> 08:52 here
<dtm> ;)
<dtm> oh 
<psu> but linux hasn't adjusted for daylight saving yet
<dtm> oh
<psu> so claims its only 07:52
<dtm> when's that?
<dtm> today/
<dtm> ?'
<psu> probably something bozo I've done here
<dtm> does that start today or something?
<psu> dtm - US/Canada daylight saving doesn't always co-incide with Europe
<psu> s/linux/GNU Linux
<psu> oops
<psu> ;-)
<psu> I always try to use FSF "house style" in Kernel Cousins, 
<psu> so refer to GNU/Linux, careful use of "non-free" etc
<psu> but probably miss a few here & there
<chillywilly> we are mentioned in appendix B.2
<chillywilly> of the paper
<chillywilly> well I ma definitely going to have to read this closer
<chillywilly> and check out all the links in the 'References' section
<chillywilly> looks to be some good research there
<chillywilly> :)
<psu> well, time for church
<chillywilly> have fun :)
<chillywilly> I should go to bed
<chillywilly> so I can get up for church
<psu> easter is always fun - kids get mini easter eggs
<chillywilly> we have a sunrise service tomorrow
<psu> ours was 3.5 hours ago ;-)
<chillywilly> where they finish acting out the thing after the roman guards have been guarding our tomb all night ;)
<psu> all the churches in town gather on the highest hill 
Action: psu is shamed to say he has never made it
<chillywilly> I don't think I will go to sunrise service
<chillywilly> it's too early
<chillywilly> scotty needs to sleep or he'll be a crabby little boy
<chillywilly> anyway
<chillywilly> g'night
chillywilly (danielb@d170.as12.nwbl0.wi.voyager.net) left irc: 
dres (dres@4.18.171.42) left irc: Read error: 104 (Connection reset by peer)
psu (psu@manorcon.demon.co.uk) left #gnuenterprise.
alexey (~alexey@195.151.214.34) joined #gnuenterprise.
Yurik (~yrashk@gw.telcos.net.ua) joined #gnuenterprise.
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) left irc: Read error: 54 (Connection reset by peer)
alexey (alexey@195.151.214.34) got netsplit.
dnSleep (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit.
jcEasterBunny (jason@HubA-mcr-24-165-193-24.midsouth.rr.com) got netsplit.
dnSleep (~derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise.
alexey (~alexey@195.151.214.34) returned to #gnuenterprise.
jcEasterBunny (~jason@HubA-mcr-24-165-193-24.midsouth.rr.com) returned to #gnuenterprise.
#gnuenterprise: mode change '+o dnSleep' by zahn.openprojects.net
alexey (alexey@195.151.214.34) left irc: "Client Exiting"
ajmitch (~me@p12-max6.dun.ihug.co.nz) joined #gnuenterprise.
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) joined #gnuenterprise.
drochaid (drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) left irc: Read error: 104 (Connection reset by peer)
drochaid_ (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) joined #gnuenterprise.
reinhard (~rm@N812P008.adsl.highway.telekom.at) joined #gnuenterprise.
<ajmitch> hi reinhard
<reinhard> hey ajmitch
<reinhard> long time no see
<ajmitch> how's things?
<ajmitch> yeah
<reinhard> happy easter to all!!
<ajmitch> heard you were doing GEAS in python so i decided to come back & hack some ;)
<reinhard> hehe
<reinhard> bad thing is that i'm supposed to lead the development and i'm buried in real work :(
<ajmitch> just means that i have to read thru the design proposals, etc, so i can comment :)
<ajmitch> yeah, cw said
<ajmitch> which is why i'm willing to help out as well :)
<ajmitch> just delegate some tasks to me & i'll try & complete them :)
dtm ([44qpgcLSX@ip94.promontory.sfo.interquest.net) left irc: Read error: 113 (No route to host)
dtm ([lwniZvk66@ip147.promontory.sfo.interquest.net) joined #gnuenterprise.
Nick change: drochaid_ -> drochaid
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) left irc: Read error: 104 (Connection reset by peer)
Action: ajmitch guesses drochaid is here
Action: drochaid gueses ajmitch is right
<drochaid> feasgar math
<ajmitch> yet another yet evening here :)
<ajmitch> wet
<ajmitch> is that 'tha i fliuch an-diugh' for it's wet today?
<ajmitch> haven't been learning it for quite awhile :)
<drochaid> it is indeed :)
<drochaid> tha i breagha an-diugh :D
<ajmitch> :P
<ajmitch> i think dunedin's weather can be nice & crap like scotland as well :)
<drochaid> heh
<drochaid> do you often get warm sunshine, driving rain and snow in the sane day as well then?
<ajmitch> we got that 2 days ago :)
<ajmitch> torrential rain, followed by thunder, lightening, hail, then sunshine
<ajmitch> all within about 30 minutes :)
<ajmitch> but during winter we'd get snow instead :)
<ajmitch> why do you think all those scottish settlers felt so at home? ;)
ajmitch (me@p12-max6.dun.ihug.co.nz) left irc: "http://www.freedevelopers.net"
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) joined #gnuenterprise.
alexey (~alexey@195.151.214.34) joined #gnuenterprise.
jamest (~jamest@fh-dialup-201034.flinthills.com) joined #gnuenterprise.
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) left irc: Read error: 54 (Connection reset by peer)
chillywilly (~danielb@d179.as21.nwbl0.wi.voyager.net) joined #gnuenterprise.
<jamest> i get to gnue some today!
Action: jamest does a dance
<chillywilly> ello
<jamest> ello
<chillywilly> sup dawg
<jamest> waiting for wife so we can go get a bit of breakfast
<jamest> then home and gnue
<chillywilly> ok
<chillywilly> I am about to got o church and go out to breakfast
<chillywilly> s/o/to
<chillywilly> why do you have to buy ISO standards?
<chillywilly> bah
<chillywilly> bye
chillywilly (danielb@d179.as21.nwbl0.wi.voyager.net) left irc: 
Action: Isomer declares himself to be standard
alexey (alexey@195.151.214.34) left irc: "Client Exiting"
Action: drochaid is the only *true* standard ... all others are FALSE!!!!!
ra3vat (ds@ics.elcom.ru) joined #gnuenterprise.
Nick change: jcEasterBunny -> jcater
Action: jcater declares himself to be substandard
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) joined #gnuenterprise.
<dneighbo> happy easter
<dnSleep> jcater: what is your schedule next week?
Nick change: dnSleep -> derek
<derek> er specifically the weekend
Action: derek is wondering if we can get you and jamest in here for 4 hours on the same day so we can have a 'bug day'
<derek> or 'developers day'
<derek> where we get lots of folks testing and submitting bugs and getting in the code and helping
Topic changed on #gnuenterprise by derek!~derek@cpe-24-221-112-50.az.sprintbbd.net: Shall we have a "developer" days?  (Free Donuts Included For All Folks Submitting Valid Patches)
<derek> hahaha a donut shop called
<derek> Grab your 'nuts and coff'
<derek> serving coffee and donuts
<derek> But we finally kept the name of Bust a Nut Donuts . We felt that it really summed up exactly the feeling you get when you are hungry and you want a donut. You are staring through the glass in amazement and you can never decide exactly which one is your favorite... then a friend leans over to you says just Bust a 'nut Man.
<derek> a donut shop with a mailing list :)
<derek> they do online ordering for friends
<derek> but i cant find out where tehy are at
<derek> roflmao
<derek> http://www.hotties.com/donut-fun-donut-defamation-league.htm
<derek> i cant stop laughing
<derek> D.A.R.E. (to keep kops off donuts)
<derek> Donut Abuse Resistance Education
<derek> must run
<derek> need donuts
derek (derek@cpe-24-221-112-50.az.sprintbbd.net) left irc: "um my gnue tree is bigger than your gnue tree"
jbailey (~jbailey@HSE-Toronto-ppp286260.sympatico.ca) joined #gnuenterprise.
<jamest> next weekend is a no go for me
jbailey (jbailey@HSE-Toronto-ppp286260.sympatico.ca) left irc: "Client Exiting"
Action: jcater is back (gone 08:53:10)
jbailey (~jbailey@HSE-Toronto-ppp319182.sympatico.ca) joined #gnuenterprise.
Nick change: jcater -> jcFamily
<jbailey> jcater's off to start a family?
<jcFamily> visiting my relatives... closer to ending a family than starting :)
<jbailey> *lol*!
<jamest> :)
sama (sama@frankye-194.micanet.it) joined #gnuenterprise.
<sama> hi all.
<jamest> hi
<sama> Is this the channel for GNU Enterprise?
<jamest> yip
<sama> great!
<sama> I'm peeking around the web site and trying to understand the general architecture of GNU Enterprise.
<jamest> ok
<jamest> how can we help?
<sama> is there any localization effort going on?
<jamest> yes
<jamest> back
<jamest> yes there is
<jamest> I've been swamped with real life stuff over last month or so, so I'm not sure of it's status
<jamest> the mail list archives may have some info on it
alexey (~alexey@195.151.214.34) joined #gnuenterprise.
<sama> jamest: i'll give a look, thanks.
<jamest> ok
<jamest> I'm going thru mail now 
<jamest> seems to be some talk in march
alexey (~alexey@195.151.214.34) left #gnuenterprise ("Client Exiting").
Nick change: sama -> sama|afk
<jamest> sigh
<jamest> Harald1: you here?
Nick change: jbailey -> angie
angie (~jbailey@HSE-Toronto-ppp319182.sympatico.ca) left #gnuenterprise ("Client Exiting").
<Harald1> jamest: now I am
<jamest> back
<jamest> you submitted patch for win32 support?
<Harald1> yes
<jamest> ok
<jamest> have you noticed any oddities when using entries with row > 1 
<Harald1> no, with the patch everything seems to work
<jamest> hmmm
<Harald1> why?
<jamest> I've got some forms at work that have issues on navigation on win32
<jamest> there was a bug in max_length support as well that I thought was win32 only but was crosss platform
<jamest> i've got a block set to rows=3 at work 
<jamest> when I'm in the 2nd or 3rd record and tab to move to next field it sometimes jumps to the 1st record
<jamest> s/record/row/g
<Harald1> derek has found another bug earlier
<Harald1> jamest: i'll try it
<jamest> derek found a bug?
<Harald1> yes:
<Harald1> <derek> go to a new record in taht form
<Harald1> <derek> then just fill up the whole box with characters w/o hitting shift return
<Harald1> <derek> i.e. hold down a key until it stops (255) char i think
<Harald1> <derek> then try to delete one of them :)
<Harald1> <derek> it wont let you :)
<jamest> that may be the bug I've fixed today
<jamest> haven't checked in yet
<Harald1> ok. the bug you described, has it anything to do with my patch, or was it there earlier?
<jamest> i have not recieved that patch yet
<jamest> but it's old bug that has plagued me for about a month
<jamest> i've just been swamped with non-gnue work so didn't have time to address till today
<jamest> to address it I have to update wife's pc with latest libs which takes a while
<Harald1> jamest: I've send it via email
<jamest> maybe I haven't got to it yet...i was about 300 mails behind in my inbox alone when I started reading it this morning :)
<jamest> hmmmm
<jamest> nope
<jamest> the last mail is the dcl mail tick #14 notice
<Harald1> it should have the topic: [Ticket # 14] Win: Multiline textfields cursor problems (the DIFF)
<jamest> did I get a copy or did you sent to a mailing list or someone else
<jamest> drat, i don't have it
<Harald1> ahh, my mistake jamest != james
<jamest> however my gnuenterprise.org account has been nuts lately
<jamest> Harald1:  you have any major win32 issues left outstanding that I need to attempt to address today?
<Harald1> jamest: to your bug: I've got the same problem. Additionally it seems that if you position the cursor in row 2 or 3 with the mouse, it first jumps to row one, and then to the right row
<jamest> which form are you testing with?
<Harald1> Harald1: no, the memory leaks of gfclient are normal
<Harald1> ?
<Harald1> and i mean jamest, not me
<Harald1> jamest: an own one I just constructed
<jamest> ah
<jamest> memory leaks?
<Harald1> jamest: yes when closing gfclient several wxWindows objects are not destroyed
<Harald1> but I'm not sure if this is a wxWindows or gnue problem
<Harald1> it says something like:
<Harald1> 20:32:45: There were memory leaks.
<Harald1> 20:32:45: ----- Memory dump -----
<Harald1> 20:32:45: wxBitmap at $13B9270, size 12
<Harald1> ....
<jamest> ah
<Harald1> altohether 80 objects
<jamest> i remember those
<jamest> been a while since I ran hybrid wxpython on win32
<jamest> been a while since I ran forms on windows at all :)
neilt (~neilt@66-61-51-107.wo4.cox.rr.com) joined #gnuenterprise.
#gnuenterprise: mode change '+o neilt' by ChanServ!ChanServ@services.
<jamest> neilt: hey
<neilt> hello all
<jamest> can you try docbook on ask
<jamest> ash
<neilt> sure
<jamest> or tell me how to test
Action: jamest is back (sorta)
<jamest> Harald1: did you resend that patch?
<neilt> ash seems real slow
<jamest> hmmm
<neilt> maybe just my imagination
<jamest> about 96-98% cpu is going to jade and friends
<neilt> havn't used it for a while
<jamest> it's old :(
<jamest> 350MHz
<jamest> Harald1: were you the mysql user?
<Harald1> jamest: yes and yes
<Harald1> jamest: jamest@gnue.org is right?
<jamest> yip
<jamest> hmmm
<Harald1> then I sent it 30 min ago
<jamest> gnue.org should redirect to gnuetnerprise.org
<jamest> gnuenterprise.org that is :)
<jamest> our servers are kinda spred out a bit
<jamest> could you send to jamest@gnuenterprise.org
<jamest> wait!
<jamest> i got it
<neilt> jamest: docbook seems to work fine
<jamest> cool, i'll close the dcl ticket
<neilt> what was the problem
<jamest> derek submitted a dcl ticket saying docbook docs could not be built on ash
crazymike (~crazymike@mke-65-31-133-230.wi.rr.com) joined #gnuenterprise.
<jamest> however this submission may be old
<neilt> did he say what the problem was
crazymike (crazymike@mke-65-31-133-230.wi.rr.com) left irc: Client Quit
<jamest> Opened By: dneighbo on 02/25/2002 
ra3vat (ds@ics.elcom.ru) left irc: "Client Exiting"
<jamest> Issue: a lot of the docs no longer build correctly on ash. it looks like the ps and pdf conversions are broken (fatal error). They all pass the parse test (first section before the space. Sample output showing error is at the end of this email. 
<jamest>  
<neilt> ah ok
<neilt> I had to rewrite the my docbook docs
<neilt> to work with the new template
<jamest> ?
<jamest> what had changed?
<neilt> could not find the gnu template
<neilt> so i changed templates
<jamest> ok
<neilt> hold on
<neilt> i'm loking at cvs 
<neilt> to jog my memory
<jamest> Harald1: on the mysql stuff....hows it going?
<jamest> sigh
<jamest> i applied that patch and pressing enter in the multiline entry returns me to the first entry on the form
<jamest> which doesn't make sense looking at the code
<jamest> it should work 
<jamest> Harald1: you were running under python 2.2?
<jamest> neilt: can I close that ticket?
<Harald1> jamest: mysql works, python 2.1.2
<Harald1> going to the first entry only happens with the patch?
<jamest> Harald1: i didn't have it running pre patch
<jamest> my win install needs work :)
<jamest> can I commit and have you test?
<Harald1> enter or shift+install
<Harald1> jamest: you could just send me your form for testing
<jamest> enter
<jamest> I'm using helloworld.gfd 
<jamest> with one field set to height="4"
<jamest> last field in the first block
<Harald1> you mean helloworld2.gfd?
<jamest> yes
<jamest> i added a height="15" to the form
<jamest> and a height="4" to the last field in the first block
<neilt> jamest: you will probably have to ask derek
<jamest> then I moved the second block down a bit
<jamest> neilt: thanks
<neilt> i had to change the files to make them work
<neilt> how come the bonsi does not show the latest revision of files?
<neilt> am i doing somethink wrong
<jamest> um
<neilt> its one revision behind and does not show my last checkin
<jamest> let me look
<jamest> where is bonsi running from
<jamest> gnue.org or gnuenterprise.org?
<neilt> gnue.org
<jamest> Harald1: to get mysql going did you have to change that fetchmany to something else
<Harald1> jamest: the same happens here, I'll remove the patch and test the old one
<jamest> gnue.org is the server in AZ under jade's control
<jamest> sigh
<jamest> i need a clone
<Harald1> jamest: yes, I remember resetting it once
<jamest> so should I adjust the mysql driver to fetchmany(0) or fetchall()
<jamest> which ended up working better for you?
<Harald1> If I remember correctly, both should work, I have fetchall() afair (but I can look it up)
<jamest> argh! postgresql support on win32 is toast
Action: jamest beats head into desk
<Harald1> jamest: the thing happens without the patch, too. It seems that this reaction is normal. When hitting enter in the last field of the second block, it jumps to the first field of the second block
<Harald1> jamest: ok I am having fetchall() there
<neilt> jamest: ok here is the story
<neilt> run this from the gnue cvs dir
<neilt> cvs diff -r 1.5 packages/sales/package-doc/Sales.sgml
<jamest> Harald1: i wonder which behaviour we want
<neilt> this will give you the needed changes to make docbook work again
<jamest> pwd
<jamest> so docbook is broken on ash then?
<neilt> no the old
<neilt> definition
<neilt> < <!DOCTYPE book PUBLIC "-//GNOME//DTD DocBook PNG Variant V1.1//EN" [
<neilt> does not work
<neilt> sorry
<neilt> actually that one does work
<neilt> we used to use OASIS DOCTYPE
<neilt> and it is broken
<neilt> or not available
<Harald1> Harald1: which behaviour? fetchmany(0) vs fetchall() or the return issue?
<jamest> return issue
<jamest> on unix enter key is newline in multirow text blocks
<jamest> on win it's next widget
<jamest> neilt: an apt-cache search oasis turns up 3 packages on docbook
<jamest> i just installed all 3
<Harald1> jamest: no, that's something you can change in gnue.conf. or do you mean something else?
<jamest> i can set that in gnue.conf?  I must be out of touch as I didn't know that :)
<jamest> i'll leave alone...seems everything is fine 
<neilt> ok, i'll test the regular docbook stuff
<Harald1> jamest: yes with EnterIsNewLine = 1, if we are talking about the same thing
<jamest> Harald1: you get that paperwork stuff from derek?
<Harald1> jamest: yes and I filled it out and sent it to derek and to fsf-records@gnu.org
<jamest> cool
<jamest> I'm updating the mysql driver to fetchall() any other mysql issues you've come accross
<Harald1> no, but I haven't really tested that. I'm implementing a bigger gfd in a few days which largely works with multiple tables. If I find any problems, I'll submit them
<jamest> cool
<jamest> I'll commit a new mysql driver now then that has the fetchany() in it
<jamest> if you could test it I'd really appreciate it
<neilt> jamest: actually i had it backwords
<neilt> we are missing the GNOME DTD
<Harald1> with fetchany? tell me when it's up
<jamest> Harald1: it's up
<jamest> neilt:  installed gnome-doc-tools
Action: neilt retrying
<Harald1> how do I tell the setup script that it should overwrite any files, even if they are changed?
<jamest> um
<jamest> it should do that automatically (setup.py)
<neilt> jamest: that worked
<neilt> jamest: ok i think you can mark it complete now
<jamest> happy happy
<Harald1> jamest: ok it seems to work. fetchany shouldn't change anything, right?
<jamest> nope
<jamest> the only change is on huge record setups
<jamest> /setups/sets
<jamest> s/setups/sets
<jamest> if you query all records in a fetchmany() driver then you get them in smaller chunks
<jamest> fetchany() requires all records load which is slower
<Harald1> jamest: so the new version is actually a slow down?
<jamest> only on huge record sets
<jamest> normally we pull 10 records (or cache #) at a time into memory
<jamest> this makes control return to user faster
<jamest> we had this issue with pygresql driver as well
<Harald1> ok, so it's an advantage and not only a workaround because fetchmany() sometimes doesn't work (crash) ?
<jamest> yes
<jamest> and no
<jamest> :)
<jamest> fetchany was the old way, fetchmany the new way but the pre-fetching doesn't always work right 
<jamest> so the new way is better responce to the user which is a big plus if you compare our responce time to something like pgaccess (slow) on huge recordsets
<Harald1> ok
<jamest> Harald1: is this clearer....
<jamest> Data not saved. Save changes or clear the form to proceed.
<jamest> or
<jamest> Data not saved. Save changes or clear the form before quitting.
Yurik_ (~yrashk@gw.telcos.net.ua) joined #gnuenterprise.
Yurik (yrashk@gw.telcos.net.ua) left irc: Read error: 113 (No route to host)
<Harald1> jamest: I think the first one is better. though there isn't much difference
<Harald1> ok, I'm going to bed. if you've got anything important that I should test, jamest, just send me an email
<jamest> ok
<jamest> take care
Harald1 (Peter1234@pD951731F.dip.t-dialin.net) left irc: 
jamest_ (~jamest@fh-dialup-201002.flinthills.com) joined #gnuenterprise.
jamest (jamest@fh-dialup-201034.flinthills.com) left irc: Read error: 113 (No route to host)
Nick change: Yurik_ -> Yurik
ToyMan (~stuq@c5300-3-ip103.albany.thebiz.net) joined #gnuenterprise.
neilt (neilt@66-61-51-107.wo4.cox.rr.com) left irc: "later all"
crazymike (~crazymike@mke-65-31-133-230.wi.rr.com) joined #gnuenterprise.
crazymike (crazymike@mke-65-31-133-230.wi.rr.com) left irc: Read error: 104 (Connection reset by peer)
ToyMan (stuq@c5300-3-ip103.albany.thebiz.net) got netsplit.
Yurik (yrashk@gw.telcos.net.ua) got netsplit.
jcFamily (jason@HubA-mcr-24-165-193-24.midsouth.rr.com) got netsplit.
Mr_You (rouzer@mankind.boredom.org) got netsplit.
drochaid (drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) got netsplit.
gnuebot (eggdrop@mail.libertydistribution.com) got netsplit.
pattieja (pattieja@www.pcxperience.com) got netsplit.
jamest_ (jamest@fh-dialup-201002.flinthills.com) got netsplit.
reinhard (rm@N812P008.adsl.highway.telekom.at) got netsplit.
Isomer (dahoose@210-86-56-169.jetstart.xtra.co.nz) got netsplit.
dneighbo (dneighbo@cpe-24-221-112-50.az.sprintbbd.net) got netsplit.
SirK (SirK@203.149.68.226) got netsplit.
pattieja (~pattieja@www.pcxperience.com) returned to #gnuenterprise.
dneighbo (~dneighbo@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise.
jamest_ (~jamest@fh-dialup-201002.flinthills.com) returned to #gnuenterprise.
jcFamily (~jason@HubA-mcr-24-165-193-24.midsouth.rr.com) returned to #gnuenterprise.
Isomer (dahoose@210-86-56-169.jetstart.xtra.co.nz) returned to #gnuenterprise.
sama|afk (sama@frankye-194.micanet.it) got netsplit.
SirK (~SirK@203.149.68.226) returned to #gnuenterprise.
gnuebot (eggdrop@mail.libertydistribution.com) got lost in the net-split.
Mr_You (rouzer@mankind.boredom.org) got lost in the net-split.
drochaid (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) got lost in the net-split.
reinhard (~rm@N812P008.adsl.highway.telekom.at) got lost in the net-split.
Yurik (~yrashk@gw.telcos.net.ua) got lost in the net-split.
ToyMan (~stuq@c5300-3-ip103.albany.thebiz.net) got lost in the net-split.
sama|afk (sama@frankye-194.micanet.it) got lost in the net-split.
bigbrother joined #gnuenterprise.
drochaid (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) joined #gnuenterprise.
<drochaid> weeee
sama|afk (sama@frankye-194.micanet.it) joined #gnuenterprise.
<jamest> argh!!!!!!!!!!!!!
<jamest> wxSomething isn't passing shift-tab events back to our event handlers on win32 platforms
Action: jamest bashes head on desk
Action: drochaid inserts a pillow between the desk and jamest
<dtm> hi guys.
drochaid_ (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) joined #gnuenterprise.
drochaid (drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) left irc: Read error: 104 (Connection reset by peer)
Nick change: drochaid_ -> drochaid
crazymike (~crazymike@mke-65-31-133-230.wi.rr.com) joined #gnuenterprise.
crazymike (crazymike@mke-65-31-133-230.wi.rr.com) left irc: Client Quit
sama|afk (sama@frankye-194.micanet.it) got netsplit.
sama|afk (sama@frankye-194.micanet.it) returned to #gnuenterprise.
ToyMan (~stuq@c5300-3-ip103.albany.thebiz.net) joined #gnuenterprise.
nickr (~panphage@e-172-IP28.empnet.net) joined #gnuenterprise.
crazymike (~crazymike@mke-65-31-133-230.wi.rr.com) joined #gnuenterprise.
crazymike (crazymike@mke-65-31-133-230.wi.rr.com) left irc: Read error: 104 (Connection reset by peer)
drochaid (drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) left irc: "Client Exiting"
dneighbo (~dneighbo@cpe-24-221-112-50.az.sprintbbd.net) joined #gnuenterprise.
ToyMan (stuq@c5300-3-ip103.albany.thebiz.net) left irc: "Client Exiting"
crazymike (~crazymike@mke-65-31-133-230.wi.rr.com) joined #gnuenterprise.
crazymike (crazymike@mke-65-31-133-230.wi.rr.com) left irc: Client Quit
Yurik (~yrashk@gw.telcos.net.ua) joined #gnuenterprise.
chillywilly (~danielb@d106.as15.nwbl0.wi.voyager.net) joined #gnuenterprise.
Nick change: jcFamily -> jcater
ajmitch (~me@p11-max5.dun.ihug.co.nz) joined #gnuenterprise.
alexey (~Alexey@195.151.214.33) joined #gnuenterprise.
<ajmitch> afternoon all
<chillywilly> good evening sir
jbailey (~jbailey@HSE-Toronto-ppp286585.sympatico.ca) joined #gnuenterprise.
<chillywilly> ello jeff
<ajmitch> jbailey!
<ajmitch> wassup?
<chillywilly> w00t
[22:12] Last message repeated 1 time(s).
<jbailey> ajmitch!  Long time!
<jbailey> Heya Daniel!
<ajmitch> looong time!
<ajmitch> jbailey: derek said something about DCL debs? you doing some?
<chillywilly> howdy jbailey
<jbailey> ajmitch: Yeah.
jcater (jason@HubA-mcr-24-165-193-24.midsouth.rr.com) left irc: Remote closed the connection
ra3vat (ds@ics.elcom.ru) joined #gnuenterprise.
<ajmitch> hey ra3vat
<ra3vat> hey ajmitch
<ra3vat> wb :)
<chillywilly> hi dmitry
<ra3vat> hello chillywilly
<chillywilly> happy easter
<ra3vat> yea, happy easter
<jamest> well
<jamest> kernel cousins 22 is out
<ajmitch> yeah?
<ajmitch> i think i read that one...
<jamest> mailman won't let me approve the fscking mail so I'll annouce in here for psu
<jamest> ?
<chillywilly> is that the last one form a couple days ago?
<chillywilly> or a brand new one?
<ajmitch> hi jamest, btw ;)
<ajmitch> same one as a couple of days ago
<chillywilly> yea, I am barely wuoted int hat one ;)
<jamest> it hit the approval queue today
<chillywilly> er, quoted
<ajmitch> i've been absent from quite a few ;)
<ajmitch> so who's the gnurpc expert?
<chillywilly> that's what happends when you disappear ;)
<ajmitch> hehe
Action: ajmitch cranks up designer to have a look
jbailey (jbailey@HSE-Toronto-ppp286585.sympatico.ca) left irc: "Client Exiting"
<chillywilly> I am pretty sure it is safe to say that jcater is the gnurpc masta
<ajmitch> ok
<jamest> jan got it working recently
<jamest> for his php client
<jamest> not sure what happened to jan though
<jamest> i need sleep
<jamest> l8r
jamest (jamest@fh-dialup-201002.flinthills.com) left irc: "[x]chat"
<chillywilly> that's who use it
<chillywilly> used
<chillywilly> I knew someone hacked it
<chillywilly> (other than jc)
<ajmitch> hmm k
<ajmitch> and how is reports? in planning? :)
<chillywilly> jcater has been working on a proposal
<ajmitch> ok
<chillywilly> on how the intermediate xml should be
<ajmitch> what doesn't jcater do?
<chillywilly> hehe ;)
<chillywilly> my sentiments exactly
<chillywilly> there's nothing that he hasn't gotten his hands on
<chillywilly> except geas, but I sure wouldn't count im out on that now that we are rewriting in python and integrating with common ;)
<chillywilly> s/im/him
Action: ajmitch needs to get into hacking 
<chillywilly> they really need to document common better though
<chillywilly> as I read the code and I immediately get lost on the details
<chillywilly> ;)
<chillywilly> s/on/in
<ajmitch> hehe
<ajmitch> which parts of common
<chillywilly> was looking at the triggers studd and GObject
<chillywilly> er, stuff
<chillywilly> our GObject that is ;)
<chillywilly> the triggers look very formsish though
<chillywilly> #   
<chillywilly> # A list of all valid triggers, and their "pretty" names
<chillywilly> #
<chillywilly> VALIDTRIGGERS = { 'PRE-FOCUSOUT':   'Pre-FocusOut',
<chillywilly>                   'POST-FOCUSOUT':  'Post-FocusOut',
<chillywilly>                   'PRE-FOCUSIN':    'Pre-FocusIn',
<chillywilly>                   'POST-FOCUSIN':   'Post-FocusIn',
<ajmitch> i wonder why?
<chillywilly>                   'PRE-COMMIT':     'Pre-Commit',
<chillywilly>                   'POST-COMMIT':    'Post-Commit',
<chillywilly>                   'POST-QUERY':     'Post-Query',
<chillywilly>                   'ON-SWITCH':      'On-Switch',
<chillywilly>                   'PRE-CHANGE':     'Pre-Change',
<chillywilly>                   'POST-CHANGE':    'Post-Change',
<chillywilly>                   'ON-NEWRECORD':   'On-NewRecord' }
<chillywilly> that
<chillywilly> well I thought the 'methods' or 'trigger' system was going to be more of a generic object thing
<chillywilly> that could be shared
<chillywilly> guess not
<chillywilly> I thought GObject might also fit the ODMG Object Model
<chillywilly> by adding that api to it
<chillywilly> the 'Object' api
<ajmitch> you love GObject ;)
<chillywilly> er?
<chillywilly> I love Gnome's concept of it
<chillywilly> i see no reason why we shouldn't have something similar
<ajmitch> a GNUeObject?
<chillywilly> yea
<chillywilly> shared object system
<ajmitch> you using DCL for stuff?
<chillywilly> but their concept of 'triggers' is not 100% compatible with our concept of 'methods' imho
<chillywilly> we are using it for bugs and stuff
<chillywilly> it's on ash
<ajmitch> yeah, but you personally?
<chillywilly> no
<ajmitch> ok
<chillywilly>  just don't take the time to play with everything like you fo ;)
<chillywilly> s/fi/do
<chillywilly> s/fo/do
<chillywilly> I am lazy you know
<chillywilly> ;)
<ajmitch> heh
<ajmitch> i love some of the names in the GEAS proposal
<chillywilly> heheh
<ajmitch> 'GNUe Object Access Translator (GOAT)'
<chillywilly> GOAT?
<chillywilly> :)
<chillywilly> GEDI is cool too
<ajmitch> yeah
<chillywilly> I don't think triggers can be shared with anything else
<chillywilly> it is too form specific
<chillywilly> *sigh*
<ajmitch> neilt's pic for GEAs makes stuff a bit simpler to understand
<ajmitch> so everything will basically be a python object? :)
<ajmitch> you want them to all derive from the mighty GNUeObject, so that they all have common methods, attributes?
<chillywilly> yea
<chillywilly> I think that would be cool
<ajmitch> it's a logical way to do things, in my limited experience :)
<chillywilly> seems to be a bit redundant wrt gtk+/gobject stuff
<ajmitch> sorry?
<chillywilly> gawd this stuff is so hairy]
<ajmitch> me no understand
<chillywilly> it's like we would be trying to do what is already done
<chillywilly> with the glib GObject
<chillywilly> and GTk+ 2
<ajmitch> i mean that for each .gcd defined object, they implicitly derive from the base GNUeObject
<ajmitch> not really
<ajmitch> that's for C, which doesn't have native OO support :)
<chillywilly> I suppose so
<chillywilly> lemme think then I guess I am smoking cracl
<ajmitch> good stuff?
Action: chillywilly is away: hitting the crack pipe
Action: chillywilly is back (gone 00:00:04)
<ajmitch> hehe
<chillywilly> heheh
<chillywilly> damn skippy
<chillywilly> so you are saying we don't need a shared object system?
<chillywilly> then why did they make one?
<ajmitch> dunno :)
<chillywilly> I think we do
<ajmitch> we do, but what would GObject give us?
<chillywilly> because odmg calls for one that implements certain metods for persistence, and  they use ot for an object that cna be described using xml
<ajmitch> sorry?
<chillywilly> s/ot/one
<chillywilly> look at the current GObject in GObjects.py once
<ajmitch> ok
<ajmitch> i am probably getting confused here
<chillywilly> hey use a common object that they named GObject
<ajmitch> GObject probably is a good thing, just a bad name ;)
<chillywilly> yes
<chillywilly> think of it as GNUeObject
<ajmitch> ok then, i can agree that GNUeObject is a goodthing(TM)
<chillywilly> I think they name a lot of things poorly in cvs, I mean look at all these cool names ;)
<ajmitch> hehe
<chillywilly> the db absatrction layer should be gedi ;)
<ajmitch> psu will probably get confused if he tries to summarise this ;)
<ajmitch> so we're in agreement then over GObject/GNUeObject?
<chillywilly> yes
<chillywilly> but I dunno if that counts for anything ;)
<ajmitch> hehe
<ajmitch> we'd need to agree on what common methods/attributes each child inherits
<chillywilly> I was thinking that the 'Object' inteface in odmg would be implemented by the current GNUeObject/GObject that lives in GObjects.py
<ajmitch> right
<ajmitch> but it could do more than just ODMG stuff, right?
<chillywilly> yes, it could support all common things that are base to our object system, imho
<ajmitch> make it easy for people writing business objects, give them a nice base to work with :)
<chillywilly> yea
<ajmitch> http://www.gnuenterprise.org/modules.php?op=modload&name=NS-My_eGallery&file=index&do=showpic&pid=31 <-- was just looking at this pic
<chillywilly> I am using that and the other documents to try and pull things together
<ajmitch> would we try & supply a few example methods with some sample objects?
<chillywilly> converting that into a UML pakcage diagram
<ajmitch> alright, can i have a look? ;)
<chillywilly> there's not a whole lot there right now
<ajmitch> the object repository will be interesting to build
<chillywilly> I see that as a ODL or ODL markup langauge repository
<ajmitch> i've played briefly with Oracle (evil thing), you can just pull your objects out of there
<chillywilly> then as the smalltalk binding does
<ajmitch> yeah, would be cool
<chillywilly> you can go several ways
<ajmitch> so basically a set of object definitions
<chillywilly> like that diagram I showed you
<chillywilly> (which I plan to exlpain)
<ajmitch> so there are object definitions, what holds the state of each object instance?
<chillywilly> ;)
<ajmitch> ok :)
<ajmitch> dang it's wet here
<chillywilly> they describe the interface, the exceptions...ecerything that IDL does except it has extra thing slike 'relationships'
<ajmitch> i don't feel like going outside anymore ;)
<chillywilly> heheh
<ajmitch> sure
<ajmitch> so as i asked, what holds the state of each object instance? the database?
<chillywilly> ODL also does 'state' of an object rather than just 'interface'
<ajmitch> alright
<chillywilly> umm
<chillywilly> tha database will have the 'state'; of source
<ajmitch> the state is basically just the attribute values anyway :)
<ajmitch> sure
<chillywilly> you could do lazy loading
<chillywilly> via a proxy
<ajmitch> oh?
<chillywilly> use a reference
<chillywilly> and only hit the db for the object state if it is used
<chillywilly> lieka  smart pointer in c++
<ajmitch> so objects & methods are still quite separate?
<chillywilly> chekc out virtual proxy pattern once
<chillywilly> design pattern
<ajmitch> yeah
<chillywilly> that's just implementation details of course
<chillywilly> we should have a methods adapter or 'server' even
<chillywilly> the way I see it
<ajmitch> sure
<chillywilly> you could get the meta objects
<chillywilly> process the odls
<chillywilly> go in any direction
<chillywilly> get the python stubs
<ajmitch> one that doesn't kill all of GEAS with a funkily-written method ;)
<chillywilly> (which make calls to the gema or methods server)
<chillywilly> like the 'schema' sompiler drawing
<chillywilly> er, compiler
<ajmitch> multithreaded python, or not?
<chillywilly> you can go any which way
<chillywilly> I would say that
<chillywilly> threads deal with transactions
<ajmitch> python doesn't scale wonderfully when doing MT, from what i hear
<chillywilly> the spec says
<ajmitch> since it has a global interpreter lock
<chillywilly> there's 3 ways to do it
<ajmitch> separate programs?
<chillywilly> on thread per transaction
<chillywilly> lemme look once
<ajmitch> sure
<chillywilly> I will document this soon ;)
<ajmitch> or psu will when it comes to KC time ;)
<chillywilly> one thread doing database operations, under one transaction
<ajmitch> ok
<ajmitch> what else?
<chillywilly> multiple threads each with their own separate transaction
<chillywilly> useful for writing servoce accessed by multiple clients on a network
<chillywilly> db maintains ACID properties
<ajmitch> you could split off things like the methods server, object repository into separate apps, so that they can run on separate machines?
<chillywilly> er, transaction properties
<ajmitch> i am just throwing in random suggestions, which may or may not be up to the required level of sanity & feasability required ;)
<chillywilly> programmers must not pass objects form on ethread to another running under a different transaction
<ajmitch> yeah
<ajmitch> cos you can screw things up, i guess ;)
<chillywilly> multiple threads sharing one or more transactions
<chillywilly> is difficult and requires the client to do concurrency control
<ajmitch> yup
<chillywilly> (locking, unlocking mutexes)
<chillywilly> so those are the 3 models
<chillywilly> I think I am leaning towards the second one
<ajmitch> depends on how much of this interface is hidden by the object server
<chillywilly> byw
<chillywilly> er, btw
<ajmitch> aye?
<chillywilly> the 'Object' interface also support locking and concurrency
<ajmitch> ok
<chillywilly> interface Object
<chillywilly>       {
<chillywilly>         enum    Lock_Type{read, write, upgrade};
<chillywilly>         // Obtains a specific lock on an object. If an attempt is made
<chillywilly>         // to get a lock on an already locked object, this operation
<chillywilly>         // will block until the specified lock can be acquired, some
<chillywilly>         // time-out threshold is exceeded, or a transaction deadlock is
<chillywilly>         // detected. If the time-out threshold is crossed a
<chillywilly>         // LockNotGranted exception is raised, If a transaction
<chillywilly>         // deadlock is detected, the transaction deadlock exception is raised.
<chillywilly>          void lock(in Lock_Type mode) raises(LockNotGranted);
<chillywilly>         // Returns TRUE if specified lock was obtained and FALSE if
<chillywilly>         // the lock is in conflict with an existing lock on the object
<chillywilly>         boolean try_lock(in Lock_type mode);
<chillywilly>         // Compares the identity of this object with another
<chillywilly>         boolean same_as(in Object anObject);
<chillywilly>         // Creates a new object that is equivalent to the receiver
<chillywilly>         // object, the new object is not the "same_as" the original object
<chillywilly>         Object  copy();
<chillywilly>         // Explicitly deletes an object from an ODMS by removing it
<chillywilly>         // from memory in addition to the ODMS
<chillywilly>         void    delete();
<chillywilly>       };
<chillywilly> another cool thing is the metadata interface
<chillywilly> and the Database interface support a schema() operations which wil allow you to traverse the metaobject graph of the 'object repository'
<ajmitch> alright
<chillywilly> every object will have a class and a metaobject describing the object
<ajmitch> how would you do object comparisons? is there an 'equals' method in python to override?
<chillywilly> so you get your intropection
<chillywilly> same_as
<chillywilly> does 'identity' comparison
<ajmitch> ah right, comments are above each method :)
<chillywilly> doe sthis object have the same OID and this one
<ajmitch> yeah, in java you have to implement the Comparable interface (ie, you have to supply a compareTo method)
<chillywilly> these are all basically 'interfaces' whihc need to be realized
<ajmitch> yep, i know
<chillywilly> yep
<chillywilly> so the spec is flexible that way
<chillywilly> / Accesses the root object that defines the schema of the ODMS. The
<chillywilly>       // schema of the ODMS is contained within a single 'Module' meta object.
<chillywilly>       // Meta objects contained within the schema may be located via navigation
<chillywilly>       // of the appropriate relationships or by using the resolve operation
<chillywilly>       // with a scoped name as the argument. A scoped name is defined by the
<chillywilly>       // syntax of the ODL and uses double colon (::) delimiters to specify a
<chillywilly>       // search path composed of meta object names that uniquely identify each
<chillywilly>       // meta object by its location within the schema. e.g., "Professor::name"
<chillywilly>       // resolves to the Attribute meta object that represents the name of class
<chillywilly>       // Professor.
<chillywilly>       ODLMetaObjects::Module schema() raises(DatabaseClosed,
<chillywilly>                                              TransactionNotInProgress);
<chillywilly> that's from 'Database' interface
<ajmitch> ok
<ajmitch> so you've got the Object methods implemented in GObjects.py?
<chillywilly> nope
<ajmitch> who wrote that?
<ajmitch> ah, jcater
<chillywilly> GObjects.py?
<chillywilly> of course ;)
<ajmitch> yeah, i just looked with cvs annotate
<chillywilly> the Database and Transaction interfaces are very cool
<ajmitch> ok
<chillywilly> they are the last 2
<chillywilly> in odmg.txt
<ajmitch> ok, i should read it i guess?
<chillywilly> most if them all have factory interfaces
<chillywilly> for creating objects
<ajmitch> cool
<chillywilly> I think it is a well designed standard
<chillywilly> but of course I am bisaed
<chillywilly> er, biased
<ajmitch> yeah :)
<chillywilly> (not really)
<chillywilly> I just didn't want to bother designing my own interfaces when there's something out there already
<chillywilly> ;)
<chillywilly> but it sure leaves the door open to implementation
<chillywilly> you still need to make it do something'
<ajmitch> yep
<chillywilly> anyway, I hope that helps
<ajmitch> yeah, it does thanks
<chillywilly> see I have a bunch of ides to organize ;)
<chillywilly> then we can get down to bidness ;)
<ajmitch> i'm ready to help out & hack GEAS :)
<ajmitch> i should code first, i guess, then get me some paperwork :)
<chillywilly> mayeb after i geta rough draft out someone can catch on and help me organize
<chillywilly> that someone being you ;)
<chillywilly> muwaahahhaha
<chillywilly> but I am slow
<chillywilly> and I really need to go to bed or mike will be unhappy if I am late
<chillywilly> :P
<chillywilly> good night ajmitchie
<ajmitch> night, talk tomorrow perhaps :)
<chillywilly> ok
<chillywilly> happy hacking
<ajmitch> happy sleeping ;)
chillywilly (danielb@d106.as15.nwbl0.wi.voyager.net) left irc: 
--- Mon Apr  1 2002