off to lunch *** johannesV_ has joined #gnuenterprise *** johannesV has quit IRC *** menomc has joined #gnuenterprise *** mnemoc has quit IRC *** menomc is now known as mnemoc dimas: you here? yep I need your help ok can you tell me what "Russian" (the language) is in Russian? preferably in HTML encoding :) but I can convert to html myself do not understand the Russian translation of the word "Russian" http://www.fsfeurope.org has a list of languages at the top each name of the language written in *that* language Russian is missing Russky or something wyep but you do use cyrillic writing, don't you? yes wonder how to pass it to you ? write it in a mail and send it to reinhard at bytewise.at ok, let's see would be great thanks a lot sent thanks np, thank kilo :) ? was taught 8 yrs of russian, but damn dunno much :) kilo: i was behind you answering reinhard's question :) :) reinhard: should i translate the whole site now? :) lol dimas: :) well, if you are bored.... ;-) let me know if something needed to be done reinhard: svn co for gnue-luca doesn't seem to work yet for anonymous. michael301080: you need to swith depository switch can you check if the following command is correct: svn co http://svn.gnuenterprise.org/var/svn/gnue-luca/trunk ./gnue-luca svn switch --relocate svn://svn.gnuenterprise.org/var/svn/gnue/trunk http://svn.gnuenterprise.org/gnue/trunk michael301080: should be http://svn.gnuenterprise.org/gnue-luca/trunk aaahh, thx Checked out revision 23. ;-) ok another question: reinhard: gnue-contrib does not work though bbl how would you / gnue support a mobile user? e.g.: a laptop equipped salesperson, who wants to print an invoice or a sales order when he is with the client. based on my experiences with such kind of stuff prefered solution would be online connection via umts or something like that considering that invoice numbers need to be sequential and all. if it has to be an offline connection you *have* to use a separate block of invoice numbers for that which is usually not a problem when using a block of invoice numbers, they can hardly be sequential by number *and* date or is that not necessary? I can't speak for dutch law It's belgian law ;-) separate blocks are the way to go here too but for germany and austria, the invoice numbers must not all be sequentially in one sequence you can have for example invoices written at office starting with 10000 and then sequencially up invoices written at notebook A starting with 20000 and then sequentially up etc actually there is no law that invoice numbers have to be sequential at all, it's just a law that you have to prove your list of invoices is complete indeed (for which a sequential number is the most straightforward way to do) Are the invoice put in a seperate sales 'journal' or how do you call it? yes straightfw => I agree probably or, to talk more practically, put the copies in a separate folder yes, separate journals (we still have to archive copies of all invoices here) Because if they are in 1 journal they need to be sequential from now it is possible to archive them electronically... EDI... must search the EU directive... IIRC you need them signed electronically if you want to archive as a file therefore you need to make sure that you get to 19999 by the end of the fiscal year, for the numbers to 'match up' with a signature key that was issued by an authority and where the state has a copy of the private key michael301080: no, no need to close the gap between the blocks If they're in one journal, you can also use smaller blocks of numbers if you can explain why there are 2 blocks, you have no problem usually it will end like aaahh didn't know that office invoices: 10000 - 12837 notebook invoices: 20000 - 21847 no problem with that ah, OK. as long as there is *some* system behind it so there's no need to have a sales journal for every sales notebook. so they'll take your word for it, when you explain that you have 2 blocks of numbers ? you need to do the normal documentation for opening a block so if there are 2 blocks open, np kilo: what do you mean with 'normal documentation' dunno how they say that in english, bnut you have to put it down if you open a block. maybe reinhard know the correct term IANAA (I'm Not An Accountant) either. reinhard: when you mentioned umts: did you meant to use it for exmpl. to serve up the gnue-forms of a server and work remotely? yes, that's what I meant curses UI might come into play with that for example running on a nokia communicator or similar ideas I was thinking along the lines of a local gnue installation on the notebook and sending an sms to your server ... ... which can then reply with the next number in the invoice number sequence. But I think that'll be hard to maintain. *** btami has quit IRC michael301080: I would not want to implement that and I think it would be quite unreliable preallocating sequence numbers is a no-no imho yep right but that sms game could also run at invoice save time but as I said I would not want *that* user to know my phone number ;-) lol lol kilo: what do you mean with preallocating sequence numbers? bbl I was just trying to look for the solution that used the least bandwith *** klasstek has joined #gnuenterprise *** jamest has joined #gnuenterprise michael301080: will explain later, now must run bbl *** kilo has quit IRC *** johannesV_ has quit IRC *** chillywi1ly has joined #gnuenterprise *** klasstek has quit IRC *** chillywilly has quit IRC back *** iro has joined #gnuenterprise *** chillywi1ly is now known as chillywilly michael301080: if you are not scared by sms-worm, sms's could be maintained automagically :) and it should not be a person's number hi dimas: didn't notice you here how's it done automagically with an email->sms service provider? technicially you can attach cell phone to computer and get sms's in your application AT commands over serial i was using gsmlib once for that did it work to your satisfaction? yes had no sms-worms ? dimas: thank you again for your help on Russian, see http://www.fsfeurope.org - it's already online :-) it was used to accept small payment in controlled environment, there was not big pressure about special security there gsmsmsd -d /dev/ttyS1 -b 57600 no_stat -a "cat - >> sms.log" oh, micropayments reinhard: best way to setup remote users is therefore to give them a GPRS or UMTS connection and access the gnue-server ... and using a low bandwith UI. reinhard: thank you too for your care about "russian" :) but what if the users want a non-text UI, or you have lots of pictures in a catalog that you use to compose your invoice? wouldn't it get pretty bandwith intensive. michael301080: and i now use gnue with remote user on slow/nonreliable connection does it work OK? i was thought that each sides should have appserver to minimize traffic I would like for the users to be able to compose their invoice or sales order buy selecting products from a catalog forms first come as a flat frame until all data fetched I read discussion on that earlier I saw compiere disable it's controls until all data was fetched (but the controls were already in place) how do 2 appservers work together? is 1 master and the other slave? I'm interested in the answers but I got to go myself. thx already *** michael301080 has left #gnuenterprise if i ever know.... :( dimas: 2 appservers acting on same db server should work, but you then still have the appserver<->db traffic over the slow connection and the problem with forms being slow and needing too much time for startup is something that annoys me as well if you want/need completely offline systems, you need db on both sides, too and exchange data in a smart way where probably the automatic timestamps maintained by gnue-appserver can be of help reinhard: agreed about db and it depend on the system hardly maybe also sqlrelay is some key to a possible solution, but I haven't looked at it at all in one case i could move db on notebook too where is not in another for different reason mobile users of a centralized db system are always a problem a completely different approach would be to use curses UI in a terminal window yes, how i forgot about that :) or even running a gtk2 gui of forms via vnc but have no idea how much bandwidth that would need reinhard: do you have time now for a few questions? am i here? just about to prepare dinner for the kids bbl reinhard: looks like johannes added assert to gLeave but i'm getting assert errors now *** jcater has joined #gnuenterprise looks like a missing return on the noLeave when no results hi howdy chillywilly whaddup dudes? looks like the hurricane is missing the oil refineries which is good for gas prices jamest: should be fixed, thanks damn chillywilly, that's heartless. what about all those oil execs and their stock options without inflated gas prices they'll never aford that 4th vacation home fsck them ;) that's what I say jamest: your irony here is anti-american and anti-capitalistic shame on you he's a terrorist ;) to make profit out of people's poverty and bad luck is good and sane look at your biggest idol, your president * reinhard shuts up now damn terrorist traitors jamest, but think of those poor 4th vacation homes in Galveston they all have dimas: back here and ready for your questions jcater: please, check gnue-contrib there is no trunk dir in there, so the pattern does not work :) reinhard: ok oh well, just leave off the trunk/ part I guess we didn't use that pattern with gnue-contrib jcater: yes, i'm just to let you know.. okay thanks reinhard: small english translation issue idiot is spelled i d i o t not idol reinhard: does it make sence to have properties in .gcd that would be rw available in forms like others but not propagate into db but used on appserver level jamest: rofl dimas: what would they be for? how long would they be persistent? would that be something like an unbound field in a block? yes something like unbound field but that reach appserver with others and could be used in ON-Validate procedure for example if it is not obvious i'd better come back later with more clear example second - there should be some syntax to define some restriction on data on the appserver side like ro that forms would follow is it hard to define that syntax? ok, understood the undbound field stuff however I would in that case probably consider to just store it in the db, even if it's not needed about read-only restricion and similar there is the plan of a access rights management for appserver, which johannes has started to work on where this would be part of i'm ok with putting in db until then I would prefer handling this stuff in the form (well knowing that you can't use generated forms then) (this stuff == read only...) ok, thanks the rights management would end up with things possible like field a read only for user 1, read write for user 2, invisible at all for user3 however it depends on a good authentication system and johannes ended up with having to overhaul the rpc layer before being able to implement that there should be a setting to disable scrollbar in a multiline entry, and if i disable say horisontal - the text would be wrapped yes, I agree on that one is it possible to use say rows="5" for buttons, to put they in a table? never tried that but I *think* (not sure) I heard somebody saying it isn't and i'm missing rw/ro hide/show switches for most widgets yes, agree we need more manpower for forms or the people that were working on it to work on it again :( *** btami has joined #gnuenterprise reinhard: just committed the incremental search support for dropdown in win32 i know it's not perfect yet any feedback welcome will read the logs, but have to run now thx, bbl *** btami has quit IRC *** sjc has joined #gnuenterprise *** iro has quit IRC *** kilo has joined #gnuenterprise *** jamest has quit IRC *** jcater has quit IRC *** yure has quit IRC *** kilo has left #gnuenterprise *** reinhard has quit IRC *** jcater has joined #gnuenterprise hi all it is I *** sjc has quit IRC *** SachaS has joined #gnuenterprise *** ajmitch has quit IRC *** ajmitch__ has quit IRC *** SachaS has quit IRC *** iro has joined #gnuenterprise *** ajmitch_ has joined #gnuenterprise *** ajmitch_ is now known as ajmitch *** sjc has joined #gnuenterprise *** btami has joined #gnuenterprise *** btami has quit IRC *** reinhard has joined #gnuenterprise reinhard: hello was interrupted yesterday, thanks for answering my questions one more i have tried indexes if unique constraint enforced i get error: ConnectionError: ERROR: duplicate key violates unique constraint "ix_sys_indx" if i correct the data and try to commit again, next error: ConnectionError: ERROR: current transaction is aborted, commands ignored until end of transaction block ok this is a bug is there a solution or workaround for this? how urgent do you need a solution? a week?