*** jcater has quit IRC
*** johannesV has joined #gnuenterprise
*** sjc has joined #gnuenterprise
*** kilo has joined #gnuenterprise
*** reinhard has joined #gnuenterprise
*** btami has joined #gnuenterprise
*** kilo has quit IRC
*** kilo has joined #gnuenterprise
<reinhard> good morning all
<reinhard> does anybody use autocommitt with grid style forms?
<reinhard> i have the strange behaviour that autocommit breaks scrolling of visible records
<reinhard> or more precise
<wt> hi
<wt> how's it shakin' kilo & reinhard
<wt> I am thinking about a general ledger implementation
<reinhard> hi wt
<reinhard> i was just working on one yesterday evening
<reinhard> :)
<wt> do you have the packages committed?
<reinhard> wt: did you already look at gnue-luca
<reinhard> ?
*** kilo has quit IRC
<reinhard> separate svn tree
<reinhard> as to my scrolling problem
<wt> we need a list of svn trees in the wiki or something
<reinhard> output scrolls up even though there would be more space below the cursor
*** kilo has joined #gnuenterprise
<reinhard> reinhard: as to my scrolling problem
<reinhard> reinhard: output scrolls up even though there would be more space below the cursor
<reinhard> (repeated for kilo) ;-)
<reinhard> did anybody experience the same problem?
<wt> Do you define anything useful for the base packages?
<wt> well, I can't right now
<wt> is there anyway I could get ssh+svn readonly access?
<wt> so I can get it from here
<reinhard> wt: yes, I try to define useful stuff for the base packages
<reinhard> wt: sorry I have no idea about readonly svn
<reinhard> doesn't the same method work like for the main gnue repository?
<wt> reinhard, how do you access the svn?
<wt> reinhard, not from here
<wt> reinhard, at home, yes
<wt> i need ssh from here
<wt> reinhard, do you use ssh+svn?
<kilo> reinhard: /me is sitting here with a baby in my arms...
<kilo> so, what is the problem? 8-)
*** Amorphous has quit IRC
*** chillywi1ly has quit IRC
<wt> reinhard, how would I get access to the server, I will update the viewcvs, if I can
*** chillywilly has joined #gnuenterprise
<wt> kilo, maybe you can answer that question also
*** kilo has quit IRC
<wt> btw, I added some more links in the wiki
<wt> accounting topics
<wt> I learned a bunch about debits/credits in the last two hours also
*** Amorphous has joined #gnuenterprise
*** kilo has joined #gnuenterprise
<kilo> here again
*** holycow has joined #gnuenterprise
<kilo> reinhard: is scenario like this:
<kilo> autocommit, you put records in the grid, first is OK, second is OK, but after second the 2nd record gets pushed down to third line and you are again on 2nd line?
<holycow> hi guys
<holycow> he so generally speaking, say i'm developing some apps in gnue
<kilo> wt: great links in wiki, thx
<holycow> and want to expose portions of it to the web, or say deliver it all via web to make it simple
<holycow> obviously there are many ways to access the db, php, python, whatever
<kilo> btw why does our wiki look like it does??? it looked differently...
<holycow> how do you guys envision starting out designing in gnue, then delivering the app to web?
<kilo> wt: we haven't discussed your question about class inheritence and such things...
<holycow> http://onefinger.sourceforge.net/screenshots.html  <-- neato
<reinhard> back
<reinhard> wt: i use ssh but have developer account
<reinhard> i think only jamest and jcater can give you access to our server
<reinhard> kilo: exactly that
<reinhard> (scrolling problem)
<kilo> so than i can affirm that as a bug
<reinhard> can you say when it started?
*** chillywi1ly has joined #gnuenterprise
<kilo> hmm, welllll, i dunno.
<kilo> iirc it was there when i first tried to do such a thing, so it is not caused by a recent change
*** chillywilly has quit IRC
<reinhard> kilo: ok, thanks
<kilo> i first tried to use autocommit on hmmm after agx tried that. and iirc we had a discussion with you about M/D and you suggested that i used autocommit...
<kilo> so it could be found in the irc logs
<reinhard> ok
<reinhard> i remember
<kilo> great
<reinhard> your autocommit datasource is also a detail datasource, right?
<reinhard> i *think* the symptom is dependent on m/d
<kilo> johannesV: thx for your notes on Friday, i used intro.gfd to test and that does not include a dropdown... then i turned to our good old gauntlet.gfd
<kilo> reinhard: yes, yes
<kilo> reinhard: iirc is IS an m/d problem, as i can't remember anything like that in zip3tier project, where there is no m/d but autocommit is on
<reinhard> ok
<reinhard> thanks again
<holycow> sorry to bug you guys again :) just curious if someone can explain to me how technologically one would access a gnue app via the web.  i understand the gnue server can have drivers written for it to 'serve' the app via other channels... how might this work technologically for the web?
<reinhard> holycow: IIRC siesel has started on a PHP implementation of forms
<holycow> *ew*
<holycow> :)
<holycow> okay but how does it work, in terms of the big picture
<holycow> say you start with forms designed via designer
<holycow> the idea is to have a driver that automatically converts the forms to webforms?
<holycow> and drive the data back and forth how?
<reinhard> i have no clue what the idea was
<reinhard> like i said siesel started that
<holycow> ah okay
<reinhard> i didn't follow too closely
<holycow> i remember reading something about drivers for the app server
<holycow> drivers would interface with the different 'deployment channels' ... do i talk to siesel more about that?
<btami> johannesV: the report parameters form doesn't works for me with wx/win32 ui, just with gtk2
<btami> tamas@TAMAS-PC:~$ grcvs -u wx monthly.grd
<btami> DB000: Traceback (most recent call last):
<btami> DB000:    File "/home/tamas/svn/gnue/gnue-reports/scripts/gnue-reports", line 29, in ?
<btami> DB000:     GRRun().run()
<btami> DB000:    File "/home/tamas/svn/gnue/.cvsdevelbase/gnue/reports/client/GRRun.py", line 215, in run
<btami> DB000:     runui.run ()
<btami> DB000:    File "/home/tamas/svn/gnue/.cvsdevelbase/gnue/reports/client/GRRunUI.py", line 103, in run
<btami> DB000:     instance.addFormFromFilehandle (formBuffer)
<btami> DB000:    File "/home/tamas/svn/gnue/.cvsdevelbase/gnue/forms/GFInstance.py", line 267, in addFormFromFilehandle
<btami> DB000:     form = loadFile (fileHandle, self, initialize = 0)
<btami> DB000:    File "/home/tamas/svn/gnue/.cvsdevelbase/gnue/forms/GFParser.py", line 68, in loadFile
<btami> DB000:     initialize, attributes={"_instance": instance,
<btami> DB000:    File "/home/tamas/svn/gnue/.cvsdevelbase/gnue/common/definitions/GParser.py", line 108, in loadXMLObject
<btami> DB000:     parser.parse(stream)
<btami> DB000:    File "/usr/lib/python2.3/xml/sax/expatreader.py", line 107, in parse
<btami> DB000:     xmlreader.IncrementalParser.parse(self, source)
<btami> DB000:    File "/usr/lib/python2.3/xml/sax/xmlreader.py", line 123, in parse
<btami> DB000:     self.feed(buffer)
<btami> DB000:    File "/usr/lib/python2.3/xml/sax/expatreader.py", line 207, in feed
<btami> DB000:     self._parser.Parse(data, isFinal)
<btami> DB000:  UnicodeEncodeError: 'ascii' codec can't encode character u'\xe1' in position 1413: ordinal not in range(128)
<johannesV> and it works if you use -u gtk2 ?
<kilo> reinhard: do you remember how the Epydoc docs were generated? http://www.gnuenterprise.org/tools/forms/docs/api/public/ is out of date, May 8...
<btami> johannesV: yep
<johannesV> btami, that's strange since it is always the same xml-parse, no matter which frontend will be used
<holycow> is siesel ever around? i've not seen that nick here ... ever actually ?
<btami> johannesV: try to modify "output filter" string to contain non ascii in your .po
<johannesV> btami, ok, i'll try
<btami> gtk uses unicode internally but wx not, meybe this is the difference...
<johannesV> right, wx is b******t :)
<johannesV> :)
<btami> hmm
<reinhard> kilo: we have to beat jcater for that
<reinhard> s/beat/ask
<reinhard> :)
<reinhard> holycow: haven't seen him for a long time
<reinhard> maybe that's the reason phpforms is not finished yet ;-)
<kilo> reinhard: it depends on how much weight he has lost recently 8-))
<holycow> ehe, okay then, let me back track.  are there and design/specs/thoughts on the 'write plugin for gnue server publish anywhere' idea?
<holycow> i read that on the original site somewhere but i cant find it anywhere now :/
<btami> holycow: do you want web forms(forms in a browser), or just deploy forms ower the web, if the latter you van use .gfd URLs with gnue-forms
<btami> s/van/can
<holycow> i just wanted to understand some of the thoughts around what has been considered, and how it would work
<holycow> *nod* re urls to gnue forms :), i thought one of the core ideas behind gnue was to be able to write drivers for appserver and publish anywhere?
<btami> you can take a look at appserver's extensions dir
<reinhard> bbl
<btami> there is a webfrontend dir using javascript
<holycow> yes but what is the general idea on developing the interfaces... there is no documentation on that?
<btami> it's siesels work too :)
<holycow> i mean if all we are storing data in an rdbm, i could use anything on the web
<holycow> or whatever toolset on ipaq allows
<holycow> then gnue is replacable too no?
<holycow> ehe they mysterious siesel :) i gotta track this fella down :)
<btami> the only doc in these ideas is .py files only :(
<holycow> ah, k. :/
<holycow> sorry for all the questions btami ... just trying to focus in on the big picture here a bit
<holycow> if one cannot use appserver to deliver content to multiple vectors... then gnue is simply replicating work of other projects with some minor non overlapping stuff
<holycow> which is cool, there is a place for everything ... k. then tracking siesel down is the next item on the todo list :)
<holycow> does anyone know if he/she ha an email?
<btami> i think appserver can do it, using rpc, but i'm not the best person to ask about  :))))
<holycow> *nod* :)
<holycow> if we could show the write once publish many...
<holycow> at least in demo
<holycow> one could theoretically map out gnue development and driver 'sponsorship' to various pieces
<holycow> the one to many is what would begin to set it apart... especially if we can make basic stuff easy for secretaries in a safe sandboxed environment (i know, dreamon rob!) :)
<holycow> i'm sure lots has been said about this already :) thx for the info btami
<btami> bbl
*** btami has quit IRC
*** holycow has quit IRC
<wt> I am back
<wt> Love me!
<kilo> lol
*** sjc has quit IRC
*** wt has quit IRC
*** tim_s has quit IRC
*** johannesV_ has joined #gnuenterprise
*** kilo has quit IRC
*** jamest has joined #gnuenterprise
*** johannesV has quit IRC
<reinhard> jamest: I hope you don't mind me messing around in forms
<reinhard> trying to fix some annoying bugs
<jamest> i don't mind
<jamest> to be honest I've not even noticed it yet
<reinhard> jamest: just did 2 commits within the last 3 hours
<reinhard> two very minor things
<reinhard> but fixed 2 scrolling problems with grid style forms
<jamest> great
*** dcmwai has joined #gnuenterprise
<johannesV_> how can i get back a file which has been removed in svn but was there a month ago ?
<johannesV_> (sql-file in zipcode-sample)
<reinhard> svn update -r 5000
<reinhard> (or any other committ number
<reinhard> )
<johannesV_> ah, ok, thanks
<jamest> if you don't what to roll back to that number
<jamest> svn cat -r <rev#> file
<jamest> should, iirc, cat that specific rev number
<reinhard> jamest: ah, cool, didn't know that
<reinhard> jamest: i have a question on dropdowns
<jamest> ok
<reinhard> currently, dropdowns get their data from a datasource and as soon as the list of possible values is populated, the connection is "lost"
<reinhard> however, for several reasons it could be handy if the datasource (and especially a block bound to that datasource) would always follow the selection in the dropdown
<reinhard> so selecting an item from a dropdown would implicitly navigate the datasource to the selected record
<reinhard> what would you think about such a change?
<jamest> i can see a use for it
<jamest> i assume that this wouldn't impact older forms using dropdowns
<jamest> as typically nothing ties to the datasource the dropdown uses
<reinhard> jamest: yeah, that's the precondition
<jamest> i think it'd be handy
<reinhard> so i will go into forms break mode again ;-)
<johannesV_> :)
<jamest> as I do triggers to accomplish something similar here
<reinhard> jamest: yes we *tried* to do that with triggers
<reinhard> *but*
<reinhard> trigger only gets fired when you focus out of the dropdown
<jamest> i setup a static datasource/dropdown to filter a datasource query
<reinhard> and of course it's a lot of repeating code all over again
<jamest> and have same problem here
<jamest> fwiw, this is something I just hit in last month, not something I'd suggest leaving as triggers
<reinhard> ok
<jamest> so your solution might work for me as well :)
<reinhard> great
*** jamest has quit IRC
<reinhard> wow
<reinhard> two blocks in a form pointing to the same datasource
<reinhard> and you can do really funny stuff
<reinhard> :)
*** jamest has joined #gnuenterprise
*** sjc has joined #gnuenterprise
*** jcater has joined #gnuenterprise
*** kilo has joined #gnuenterprise
*** dcmwai has quit IRC
<reinhard> @all: please test my last commit
<reinhard> thanks
<reinhard> off to lug meeting
*** reinhard has quit IRC
*** jcater has quit IRC
*** cilkay has quit IRC
<johannesV_> reinhard: there are some issues with your patch
<johannesV_> records detail-forms are not following the record-change
<johannesV_> i'll explain it in detail tomorrow
*** johannesV_ has quit IRC
*** wendall911 has joined #gnuenterprise
*** wendall911 has left #gnuenterprise
*** holycow has joined #gnuenterprise
*** kilo has quit IRC
*** kilo_ has joined #gnuenterprise
*** sjc has quit IRC
*** cilkay has joined #gnuenterprise
<kilo_> jamest: what is needed to generate the epydoc api docs on the website? they are quite out of date (May 8 iirc)...
<jamest> jcater tied that into a nightly script I thought
<jamest> i think i can take a look at that script tonight after I get off work
<jamest> probably about 1 to 1.5 hrs
<kilo_> ok, thx.
*** kilo_ has quit IRC
*** jamest has quit IRC
*** jcater has joined #gnuenterprise
*** chillywi1ly has quit IRC
*** chillywilly has joined #gnuenterprise
*** jamest has joined #gnuenterprise
*** chillywilly has quit IRC
*** chillywilly has joined #gnuenterprise
* holycow waves to all
<holycow> who am i looking for again, with regard to writing drivers for gnue server to say publish to web or other platforms?
<holycow> seisel? is that the correct spelling
<holycow> ?
<holycow> http://www.rekallrevealed.org/toplevel/webaccess.shtml
<holycow> rekall now has basic publishing to web capacity as well
<holycow> neat
<chillywilly> bleh, KDE
*** jamest has quit IRC
<chillywilly> homo
*** jcater has quit IRC
*** wayneg has joined #gnuenterprise
*** holycow has quit IRC
*** holycow has joined #gnuenterprise
<holycow> hi guys