*** reinhard has joined #gnuenterprise
*** Amorphous has joined #gnuenterprise
*** Morphous has quit IRC
*** dimas has joined #gnuenterprise
*** btami has joined #gnuenterprise
*** dimas has quit IRC
*** SachaS has joined #gnuenterprise
<reinhard> anybody here have experience with VNC?
<reinhard> hey Amorphous
<reinhard> are you siesel?
<Amorphous> no i'm afaik not
<reinhard> ok
<reinhard> we had another jan from germany here in the past :)
<Amorphous> yea this name is fairly common
<SachaS> good morning
<SachaS> have to run for lunch
<reinhard> <SachaS> good morning
<reinhard> <SachaS> have to run for lunch
<reinhard> *lol*
<SachaS> reinhard: will have to talk to you later.
<reinhard> you should get up earlier
<reinhard> ;)
<SachaS> :)
<SachaS> when i went to bed it was about 5am.
<SachaS> so day starts up side down
<SachaS> i have done something  last night, which might be either ok or very bad practice
<SachaS> well talk to you later.
*** SachaS has left #gnuenterprise
*** chillywilly has quit IRC
*** kilo_ has joined #gnuenterprise
*** dimas has joined #gnuenterprise
*** chillywilly has joined #gnuenterprise
*** lekma has joined #gnuenterprise
*** lekma has quit IRC
*** dneighbo_ has joined #gnuenterprise
*** dneighbo has quit IRC
*** lekma has joined #gnuenterprise
*** btami has quit IRC
*** holycow has quit IRC
*** kilo_ has quit IRC
*** chillywilly has quit IRC
<lekma> hi everybody
*** dsmith has joined #gnuenterprise
<lekma> is there any 'commit' mailing list archive?
<lekma> ok found it
<lekma> reinhard
<reinhard> hi lekma
<lekma> in svn 5582 you remove the import of dyn_import in gnue-common/datasources/Gconnection.py
<reinhard> yes
<lekma> i needed it to make my own authentication wrapper
<lekma> do you pla to replace this by something else?
<reinhard> where do you need it?
<reinhard> you have changes to GConnections.py?
<lekma> in connections.conf i have a custom_auth parameter
<lekma> that dynamically import
<lekma> a python script
<reinhard> so you do have changes in GConnections.py?
<lekma> nope
<reinhard> you can place the import wherever you want
<lekma> yes i know
<reinhard> oh
<reinhard> sorry
<reinhard> just saw it
<reinhard> sorry, sorry
<lekma> np
<reinhard> i replaced a call to dyn_import and wasn't aware it's still called at another point
<reinhard> so i thought i can remove the import safely
<reinhard> it's fixed in svn 
<reinhard> thanks for reporting
<lekma> thx
<lekma> i need some advice
<lekma> i have a project sold with gnue technology
<lekma> and we must go live in october
<lekma> it's a kind of babyerp
<lekma> should i go for the appserver
<lekma> or should i stay with only gnue-forms
<lekma> ??
<reinhard> sigh
<reinhard> if i only knew for sure :)
<reinhard> as appserver stands now
<reinhard> i think it will in no respect have a downside against 2-tier
<reinhard> that is
<reinhard> i think you will be able to do everything with appserver that you can do without appserver
<reinhard> and it might well be that you can use some features that you wouldn't have without appserver
<reinhard> so my view is
<reinhard> you can't loose much
<reinhard> you can win
<reinhard> and after all
<reinhard> also the gnue project would win :)
<reinhard> if you'd do it with appserver
<reinhard> you could be assured that we'd fix the bugs you find with highest priority
*** SachaS has joined #gnuenterprise
<reinhard> hi SachaS
<lekma> that is a big incentive
<SachaS> hi reinhard
<SachaS> this lunch took forever :)
<lekma> and i think you're right
<SachaS> good i brought the laptop along ;)
<SachaS> reinhard: do you think there is a need for instance business class procedures?
<SachaS> i might get sidetracked to a completely wrong path.
<SachaS> but i created a inventoryHelper business class.
<SachaS> of course, there needs to be just one of such an object.
<SachaS> with instance procedures, I would not even have to create an instance (object). 
*** nenhum has joined #gnuenterprise
<reinhard> hi nenhum
<reinhard> SachaS: i think you are being sidetracked
<SachaS> :|
<reinhard> sorry but a bit under time pressure at the moment
<SachaS> ok. will talk to you later again.
*** nenhum has left #gnuenterprise
*** lekma has quit IRC
<SachaS> reinhard: i think i am just exploring different ways of doing things.
<SachaS> its pretty new to me.
<SachaS> calculated fields: seem like they have a procedure assigned to them and each time they are accessed, the procedure is called .
*** jcater has joined #gnuenterprise
*** wendall911 has joined #gnuenterprise
<reinhard> hi jcater, wendall911
<reinhard> SachaS: that's quite much what calculated fields are, yes
<wendall911> good morning :)
<wendall911> at least it's morning here ;)
<jcater> morning
*** lekma has joined #gnuenterprise
*** jamest has joined #gnuenterprise
<reinhard> good morning jamest
<jamest> afternoon reinhard
<jamest> (i think :)
<reinhard> jamest: correct
<SachaS> reinhard: i have removed that inventroy helper class ... back on track.
<reinhard> SachaS: excellent ;-)
<jamest> please no commits for a bit
<reinhard> hey
*** wendall911 has quit IRC
<reinhard> who told you that you could stop coding?
<jamest> i am fixing our fscking svn setup
<jamest> does that count?
<reinhard> so you found out what causes our problems?
* jamest publicly bitches about using the debian subversion packages
<jamest> ash got upgraded subversion package again
<jamest> and the wrappers were not all replaced
<jamest> i'm going to chattr +i the files
*** wendall911 has joined #gnuenterprise
<jamest> so the next apt-get upgrade will fail
<jamest> :)
<reinhard> :)
<reinhard> as long as it fixes our svn issues, i'm ok with what you want
<reinhard> :)
<SachaS> good news: my inventory sales order processing works (whatever that means) :)
<SachaS> eg. calculating new masterdata_inventory(should be item).masterdata(should be inventory)_quantityOnHand
<SachaS> have to run
<jamest> sigh
<jamest> that won't apply to a symlink
<reinhard> bbl
*** chillywilly has joined #gnuenterprise
<jamest> hi chillywilly
*** lekma has quit IRC
*** jamest has left #gnuenterprise
*** jamest has joined #gnuenterprise
*** ogger has joined #gnuenterprise
<jamest> hi ogger
*** dimas has quit IRC
*** _florin_ has joined #gnuenterprise
<_florin_> howdy
<chillywilly> is it possible to test if a list of values is a subset of another list... or that several values are in a list?
*** kingpin2k has joined #gnuenterprise
*** kilo has joined #gnuenterprise
<jcater> chillywilly: not that I know of
<jcater> short of a for loop
<jcater> though I bet Numerical Python adds support for it
<jcater> (but that'd be a separate package)
<jamest> i think that's in python 2.4
<jamest> think is the key word
*** siesel has joined #gnuenterprise
<SachaS> hi siesel
<siesel> hi SachaS
<SachaS> siesel you are not so much here anymore. are you busy?
<siesel> yes, got a job :(   
<SachaS> congratulations.
<siesel> ;)
<siesel> thx
<siesel> how are you? You're back from downunder right?
<SachaS> back in liechtenstein, yes.
<SachaS> have a little bit of time to work with gnue :)
<siesel> nice
<SachaS> i am building a sample app, so, that I can, one day, try to have gnue talk to gnue.
<jamest> hi siesel
<siesel> hi jamest
<jamest> so you get a little money in the pocket and now you're too good to hang out in here huh?
<jamest> :)
<siesel> yeah, I got corrupted ;)
<jamest> probably for the best
<jamest> we're a bad influence
<siesel> But there is still some free time in my life... gnue --> you can't get rid of me so easy :)
<jamest> great!
<jamest> now get to work
<jamest> :)
<siesel> lol
<SachaS> jamest: if I have a block with rows = 5. for one entry box (part of the block) i want to set the "post-query" trigger. atm there is only one trigger for one field, even if the block has the rows attribute. is there a way i can access the entry field of each row.
<jamest> not sure if this is what you are asking
<jamest> but this iterates thru a block manipulating all the fields
<jamest> IIRC however the focus will still change thru the records
<jamest>       <button name="setTrackNum" Char:height="1" Char:width="5" Char:x="88"
<jamest>               Char:y="11" label="Set">
<jamest>         <trigger type="ON-ACTION" name="setTrackNumTrigger"><![CDATA[
<jamest>             for record in blkPO_ITEMS:
<jamest>                 blkPO_ITEMS.fldTrackNo = quickSet.trackingNumber
<jamest> ]]></trigger>
<jamest>       </button>
<jamest> sigh, kopete bitched about that paste
<jamest> please let me know if you got it
<SachaS> i got it
<SachaS> thanks.
<jamest> is it what you needed?
<SachaS> yes thats what i needed.
<SachaS> things like this are hard to document
<SachaS> would more like a gnue cookbook with different examples....
<jamest> we have a few
<jamest> in the forms dev guide
<jamest> and if you're up to it....... :)
<jamest> http://www.gnuenterprise.org/cgi-bin/twiki/view/Main/TriggerRecipes
<jamest> lol
<jamest> someone took one of my mails and posted it to the wiki 
<jamest> man, i hope i got everything right in that mail
<jamest> :)
<kilo> aye, twas me
<jamest> ah :)
<kilo> i found that email perfect
<jamest> cool
<kilo> the next one yu wrote was also perfect but i havent processed that yet
<jamest> well, i hope they are still completely accurate
<jamest> should be
<jamest> but i always hate writing stuff like that
<jamest> i fear that I'll submit and get
<jamest> <jcater> dude, i completely rewrote that in the last 15 minutes, it now uses quantum vacum fluxuations with recursive inheritance 
<kilo> then you simply delete the wiki and thwap jcater...
<jamest> good idea
*** siesel has quit IRC
*** kingpin2k has quit IRC
*** _florin_ has quit IRC
*** dsmith has quit IRC
*** Vee2d2 has quit IRC
*** ogger has quit IRC
*** wendall911 has quit IRC
*** dneighbo_ has quit IRC
*** Amorphous has quit IRC
*** Vee has quit IRC
*** icltlfatppl has quit IRC
*** siesel has joined #gnuenterprise
*** kingpin2k has joined #gnuenterprise
*** _florin_ has joined #gnuenterprise
*** wendall911 has joined #gnuenterprise
*** dsmith has joined #gnuenterprise
*** dneighbo_ has joined #gnuenterprise
*** Amorphous has joined #gnuenterprise
*** Vee2d2 has joined #gnuenterprise
*** Vee has joined #gnuenterprise
*** icltlfatppl has joined #gnuenterprise
*** ogger has joined #gnuenterprise
<jcater> fwiw, I'm not sure I'd emphasis the
<jcater> block.zip = "foo" syntax
<jcater> as it's so inconsistent
<jcater> i.e., works sometimes and other times doesn;t
<jcater> (and isn't pythonic, so might confuse ppl)
<jcater> might be better in an advanced section
<kilo> go ahead and edit you own material 8-))
*** siesel has quit IRC
<SachaS> wb reinhard.
<reinhard> SachaS: thanks
<SachaS> whats going on with appserver? any new features implemented?
<SachaS> :)
<reinhard> well
<reinhard> will be just finished with real life work in about 10 minutes for today
<SachaS> dah?
<SachaS> :)
<SachaS> you have been busy then.
<reinhard> well
<reinhard> this evening i did the kind of work i like most
<reinhard> writing the monthly invoices ;-)
<SachaS> hehe :) good sign you were busy till 22.45!
* SachaS would like to be in a position to write monthly invoices :)
* SachaS is rather in a position where he receives monthly invoices
<reinhard> lol
<jcater> reinhard: I think I'd prefer
<jcater> writing out the deposit slips
<jcater> but that's just me
* _florin_ ignores the monthly invoices he receives... (bad policy?)
<SachaS> reinhard: have you heard of the buspro product? 
<reinhard> no
<_florin_> is there any possibility of doing heavy computation in gnue-forms while the UI still remains responsive (some kind of timer or similar?)
<jamest> _florin_: not currently
<_florin_> any workarounds possible (i.e. some kind of threads/forks?)
<jamest> i would think you could import python threads
<jamest> but I'm not sure how you'd interact with the UI
<jamest> you're moving into unknown teritory :)
<_florin_> again? :-)
<jamest> sigh
<jamest> i'm parting out an old pentium box that is burnt up
<jamest> the mfg glued the cables into the slots
<_florin_> wave a dead rat over it... might help :-)
<kilo> it will be good for a firewall 8-)
<jamest> might work a little better as trash
<jcater> uggh
<jcater> I'd rather think about handling dead rats
<jcater> than about forms and thread support
<jamest> but jason
<_florin_> writing multi-threaded applications has a lot to do with waving all kind of dead animals over your code...
<jamest> first we embed forms in navigator
<jamest> then we let them flip between executing threads each containing a forms instance
<jamest> wouldn't it be lovely
* jcater beats jamest with a dead rat
<kingpin2k> better stick that dead rat in a locked state first, otherwise we may grab the same rat
* jcater walks off to get some Tylenol
* reinhard goes to get a cat
<_florin_> how difficult would it be to make the event-passing system thread-safe (i.e. give me the possibility to send events from one thread to another)?
<kingpin2k> semaphores are your friend
<_florin_> (rotflmaybitc)
<_florin_> s/y/o
<reinhard> bitc?
<_florin_> biting in the carpet
<reinhard> ah
<reinhard> :)
<jamest> _florin_: my first guess would be it would require a rewrite of, um, everything
<_florin_> hm... bad.
<kingpin2k> does event passing really require thread safety?
<jamest> _florin_: what do you need threaded anyway?
<_florin_> it's not like i really _need_ it... but it would be an elegant way to do time-consuming computations to fill in values various lists
<_florin_> think of that win32-directory-property-dialog, where the size appears some time _after_ you open it...
<_florin_> as i understand it, searching for lots of items in big tables is not an issue (< 1 sec).
<_florin_> saving data in internal structures... well, that takes some more time, but still not bad.
<_florin_> but as soon as you have to do something nasty for each entry (say... one more DB request per entry)
<_florin_> it gets ugly.
<kingpin2k> so maybe just a couple of thread-enabled widgets?
<kingpin2k> and have them implement some type of progress checking procedures?
<_florin_> yes. but the computing thread needs a way to tell the others "i'm finished"... or the others need a way to poll from time to time "are you finished yet?" without user interaction.
<SachaS> _florin_ everthough of having the computation done by the appserver?
<SachaS> it could be a big fat machine.
<_florin_> yes but... :-)
<_florin_> 'big fat machine' is not my favorite answer to speed issues :-)
<_florin_> it's not the time that's the problem, but the fact that the user cannot interact with the software while it's computing (i.e. say "abort -- it takes too long")
<kingpin2k> if the computing thread has to loop and call AreYouDoneYet() 1000 times, the user could still interact, no?
<kingpin2k> not the computing thread...the main thread
<_florin_> yes.
<_florin_> that would be not the most elegant method, but probably the easiest to implement for now
<_florin_> as it doesn't require the main thread to be actually thread-safe :-)
<kingpin2k> and leave it up to the designer to do the checking
<kingpin2k> right
<_florin_> (i naively assume that i can define/use DataSources/DataBlocks in the computing thread without disturbing the rest of the application)
<reinhard> night all
*** reinhard has quit IRC
<_florin_> night reinhard
<kingpin2k> you could get into race conditions if you had widgets doing comparable things
<_florin_> i'm afraid i don't understand...
<kingpin2k> where widget B commits before widget A even though widget A should have been first
<_florin_> design issue. if you can settle on "main thread is the only UI-thread" and "main thread is the only write-access thread" it's a lot simpler.
<jamest> nothing is gnue (that I know of) has been tested in a threaded environment
<kingpin2k> hmmm...that increases complexity a bit
<_florin_> jamest: worst case: open a new connection to the DB (just take 'user' and 'pass' from the form) -- that should work, as it's technically like running a whole new application...
<_florin_> right?
<_florin_> kingpin2k: what increases complexity?
<kingpin2k> having main thread be the only write access thread means that main thread now has to be completely thread-aware, right?
<_florin_> no. it simply means that the compute-thread "promisses" not to save any changes :-)
<kingpin2k> ahhh
<kingpin2k> basically only threaded reads...like the windows control you mentioned
<_florin_> but still need a way to poll from main thread (i.e. a timer)
<_florin_> yes.
<_florin_> threaded writes would be kinda suicidal
<kingpin2k> are timers implemented in any of the python ui stuff?
<kingpin2k> gtk, wx, etc?
<_florin_> the are in Qt. i don't know about the others...
<kingpin2k> hmmm...
<jcater> yes
<kingpin2k> cool
<jcater> afaik, they all have one
<jcater> wxTimer, f.e.
<SachaS> can i call a trigger from within a trigger?
<kingpin2k> gotta run...bbl
<SachaS> oh... my "post-query" trigger is not called when the datasource is prequery.
<_florin_> i read something about "idle events" in wxGtk.
<_florin_> i know Qt had similar too
<_florin_> could something like this be added to gnue without big changes/breaks?
*** dsmith has quit IRC
<jamest> widget specific stuff?
<jamest> i wouldn't want to see qt only stuff in forms
<_florin_> me neither :-)
<jamest> i don't have a problem with stuff becoming thread safe at some point
<_florin_> but since these idle-events are a quite useful alternative to threads when it comes to heavy processing, i think that most GUIs have it. except for curses, of course...
<jamest> but it seems we could use a bit more of the basic stuff people need implemented first
<_florin_> agree.
*** kilo has quit IRC
*** jcater has quit IRC
<SachaS> i experience navigator problems.
<SachaS> DB000:   File "/home/sacha/gnuenterprise/gnue/gnue-common/.cvsdevelbase/gnue/forms/GFInstance.py", line 296, in activate
<SachaS> DB000:     self.manager.handleStartupError(
<SachaS> DB000: AttributeError: Instance instance has no attribute 'handleStartupError'
<SachaS> does it have to do with, different forms have different properties, such as on-startup, readonly on form level, readonly on block level? ...
<jamest> hmmm
<jamest> is it a random combination of forms that trigger this
<jamest> or 1 form in particular
<SachaS> so far random combination...have not nailed it down.
<jamest> is there more to the traceback?
<jamest> if so msg it to me
<jamest> please
<SachaS> now it just went through all forms, without a hicksup.
<SachaS> 11 forms
<SachaS> could it might be when you double click too fast?
<SachaS> will try.
<SachaS> yes.
<SachaS> dont double click too fast ;)
<SachaS> full traceback (not that long)
<SachaS> if you get errors, you are working too fast.
<SachaS> :|
<jamest> :)
<jamest> forms embedded in navigator are a quick hack
*** jcater has joined #gnuenterprise
<jamest> i'm suprised they work as well for you as they have
<SachaS> its pretty neat.
<SachaS> early night
<SachaS> good night
*** SachaS has left #gnuenterprise
<jamest> whoops
<jamest> an email from RMS
<jamest> Please don't use ``win'' as an abbreviation for Microsoft Windows in
<jamest> GNU software or documentation. 
<jamest> they either want us to spell out windows or go w32
<jamest> vs win32
<jamest> because
<jamest> In hacker terminology, calling
<jamest> something a "win" is a form of praise.  If you wish to praise
<jamest> Microsoft Windows when speaking on your own, by all means do so, but
<jamest> not in GNU software. 
<nickr> O_o
<nickr> that is asinine
<jamest> so
<jamest> we have 4 locatoins w/ that file name
<nickr> please use only the official updated newspeak dictionary
<jamest> as a project do we abdicate
<nickr> thoughtcriminal
<jcater> sigh
* jcater beats head against desk
<jcater> well, geee
<jcater> in that case
<jcater> let's not call stuff free
<jcater> as it cheapens it
<jcater> (to those who don't know better)
<jcater> same thought process, imho
<jamest> well, this has been in my mailbox since, um, well
<jamest> Sep 17th
<jamest> =)
<jcater> lol
<jcater> in thinking about this
<jcater> FSF is last group I'd think of worrying about name connotations
<jcater> after all, who calls stuff GNU w/a straight face
<jcater> or what connotations does Hurd come with?
<jamest> well, i'm purging the mail
<jcater> (I can only think of a few words that rhyme with that)
<jamest> i've notified the group
<jamest> and seems like the only feedback in here has been negative 
<jamest> so, before I'm lynched
* jamest runs
* jcater grabs the rope
<jcater> seriously,
<jcater> this reminds me of congress decreeing you can't call French Fries by that name
<jcater> must use Patriot Fries in gov't cafeterias
<jamest> please tell me that didn't happen
<jamest> i put nothing pass our politicians
<jcater> not a "decree" per say
<jcater> but, yes
<jcater> they are Patriot Fries now
<jcater> have been for a little while now
<jamest> sigh
<jamest> when I was in high school 
<jamest> way back in the 80s
<jamest> the deficit was, well, basically what it is today
<jamest> rather than deal w/ tough issues the guberment was making a huge deal out of flag burning
<jcater> I remember that
<jamest> anyway, this seems like more of the same
<jcater> sigh
<kingpin2k_away> what about the financial connotations of "float"...we should definitely not use that anymore
<jcater> no
<jcater> that is on the chopping block
<jcater> we have a fixedpoint implementation in gnue common
<jcater> just need to get stuff converted over
<kingpin2k_away> hehe
<kingpin2k_away> it's funny that this isn't even about copyright...it's about other uses of the word formed by an abbreviation
<kingpin2k_away> usually, the word win would elicit copyright claims
<jamest> i've always figured the win abreviation was intentional on MSs part
*** jamest has quit IRC
*** kingpin2k_away has quit IRC
*** chillywilly has quit IRC
*** chillywilly has joined #gnuenterprise
*** _florin_ has quit IRC
*** dcmwai has joined #gnuenterprise