:) dneighbo@latitude:~/cvs/gnue/navigator/samples$ ~/bin/gncvs -u web sample.gpd DB000: Traceback (most recent call last): DB000: File "/home/dneighbo/cvs/gnue/.cvsdevelbase/gnue/navigator/GNClient.py", line 147, in ? DB000: GNClient().run() DB000: File "/home/dneighbo/cvs/gnue/.cvsdevelbase/gnue/navigator/GNClient.py", line 144, in run DB000: interface.Instance(nav).run(self.connections) DB000: TypeError: run() takes exactly 1 argument (2 given) dneighbo@latitude:~/cvs/gnue/navigator/samples$ is error when trying web version yummy i patched it hi anyone know of any small/medium business accounting software implemented using GNUe? how feasible is it to do something like that at the moment? nola ? its not gnue but well ok Action: jcater must run jcater (~jason@cpe-024-165-193-024.midsouth.rr.com) left irc: "nite" bye jcater its accounting that looks promising to do some gnue stuff too Action: ajmitch hates it when he scares ppl away alright, was just talking to someone about doing something using GNUe Action: derek needs to run to bed too so i can get up at 04:00 to watch the england v argentina game it can be done using existing tools? Action: ajmitch listens to the silence SachaS (~Sacha@dialup-196-242.wasp.net.au) left irc: Read error: 110 (Connection timed out) psu (psu@manorcon.demon.co.uk) joined #gnuenterprise. hi aj hey psu how are you? NOLA is, iirc, definantely not GPL fine it was discussed on Kernel Cousins recently it's web-based and might have some non-free depends (java springs to mind) ah ok sad it was just suggested to me as a project, since most businesses in NZ are small (< 5 employees, iirc) no, I'm wrong - it *is* GPL http://kt.zork.net/GNUe/gnue20020427_26.html#7 It's a LAMP application (Linux, Apache, MySQL, php) screenshots look moderately sophisticated (but then so do ours) jcater downloaded for a look-see, you might want to ask him when he wakes jcater just ducked out 20min ago :) well, there's an online demo looking now based on 5 minutes, look and feel isn't terrible, isn't wonderful and being GPL gets you lots of gold stars in my book ;-) yeah would still like something GNUeish tho web-based is ok but not ideal for things like printing reports, invoices, etc Action: psu needs to have a look at the AppServer ROADMAP we should be able to tell now what release we will have an AppServer yup that will support releasable applications great are you up really early or something? or late? ;) not that early - almost 7 a.m. early for me as far as I can tell, I am the only person in the country going to work today heh, why's that? for me, 04:45 is early ajmitch: an association football match, of some interest to the masses ;-) England v Argentina is a 10:30 am kick off our time ah i see... ack HTML requires align=center, doesn't support align=centre Action: ajmitch should try & learn forms a bit better then ;) heh I thought it recognised both US English and English English for most stuff Action: ajmitch dislikes americanisms :) grey/gray etc. you mean US English & Proper English? ;) Action: psu goes off to change all his pages NZ is still a lot closer to UK style than US Action: ajmitch doesn't like NOLA's UI that much forms needs to be spiced up a bit, gray forms are so boring ;) it's LAMPish. what can I say? heh DCL could be called LAMPish, but i think it has a better UI so does phpGW NOLA is GPL Action: derek decided to stay up and watch the first game and record eng v arg (no way i can wake up at 4) nola is far from perfect BUT i think its far enough that its a decent start and it would be worth building gnue screens for and such i have a need for nola and will probably get gnue screens for it ASAP as well as make it support fund accounting fund accounting == cash accounting == non-accruals? ? i.e. invoices go in the period they are paid, not period they are raised? sounds good derek um no my question was different i understand cash vs accural fund accounting is generally for non profits Action: psu would record Eng v Arg, but it's 100% likely I would get told the result before watching it i.e. a fund is an accounting entity with self balancing set of accounts usually used to observe some restriction or limitation of resources multi-organisation psu: that would be a private sector use probably pretty much all govt's in US are required by law to do fund based accounting my personal need for it is 'church management software' i.e. church accounting where there are funds like building fund general fund missions fund childrens fund etc i like to think of 'fund accounting' as bank accounts on paper i.e. its like a 'strictly enforced budget' the 'fund balance' is like the 'account balance' at the bank We do much the same at work at moment - each directorate has own balance sheet but the "Cash" figure is not real cash only instead of having a ton of bank accounts you have paper (or virtual) ones right just their share of the central balance sheet generally every fund has which is part cash, part Creditors, part Debtors etc assests, liabilities, fund balance I keep wavering as to whether this is a good idea in our sitation or not as it's not required Action: derek doesnt like this talk as it reminds me of work :) and the control aspects can be done w/ good row security cant we argue about football instead :) derek: just put it down on your timesheet Action: derek shudders as mention of timesheet "01:00 - meeting to discuss alternative arrangements for fund accounting" reminds of work still needing to be done to institute cost based accounting now talk about fun cost based fund accounting Eng v Arg is not fun unless we win ;-) we have a saying we put the fun in dysFUNctional oh crap missing start of sweden v nigeria bbl Action: derek is away: football hooligan holiday all praise Chipaca if he hadn't replied to jcater's email thus carrying that thread over to next week this week's KC would be > 100k heheh that would be impressive Action: ajmitch waits breathlessly for this week's KC to go online bbl hot damn you see that goal by nigeria that just rocked psu (psu@manorcon.demon.co.uk) left #gnuenterprise ("the daily commute - all 6 minutes of it"). btami (~tamas@ip102-205.ktv.tiszanet.hu) joined #gnuenterprise. morning woooo hooo larsen nets one for sweden this has been best game (as far as action goes) that i have seen yet in korea how can one use gnu-rpc from php? xml-rpc let me rephrase that right now you cant use gnu-rpc really from php but you could talk to geas from php via xml-rpc and the geas side would be using gnurpc Grantbow (~grantbow@12-233-20-180.client.attbi.com) joined #gnuenterprise. derek: not really what i was wanting, but thanks ;) hi Grantbow hi. Now I'm really going to bed. ;-) hehe Seek3r (~dankuyken@vnnyca-1-g1-s1-46.vnnyca.adelphia.net) joined #gnuenterprise. Seek3r: good luck dragging derek away from watching soccer :) derek: is there any special reason navigator uses treecontrol instead of classic menus? in a main window ? madlocke (madlocke@pD9E0A578.dip.t-dialin.net) joined #gnuenterprise. siesel (jan@dial-213-168-89-114.netcologne.de) joined #gnuenterprise. morning hello madlocke,siesel hello... ariel_ (~ariel@ppp-217-133-245-161.dialup.tiscali.it) joined #gnuenterprise. siesel (jan@dial-213-168-89-114.netcologne.de) left irc: "later" Action: madlocke is away: I'm busy Action: btami is away: lunch siesel (jan@dial-213-168-89-94.netcologne.de) joined #gnuenterprise. alexey_ (~Alexey@techkran.vladimir.ru) left irc: "[x]chat" siesel (jan@dial-213-168-89-94.netcologne.de) left irc: "Real IRC clients know "to" is a preposition, not an adverb" jbailey (jbailey@CPE014260028338.cpe.net.cable.rogers.com) joined #gnuenterprise. ToyMan (~stuq@65.167.123.51) joined #gnuenterprise. Action: btami is back (gone 02:13:58) jbailey (jbailey@CPE014260028338.cpe.net.cable.rogers.com) left #gnuenterprise ("Client Exiting"). Action: madlocke is back (gone 02:53:33) jamest (~jamest@hobbes.math.ksu.edu) joined #gnuenterprise. siesel (jan@dial-194-8-205-46.netcologne.de) joined #gnuenterprise. hi all hello siesel siesel: is there any special reason navigator uses treecontrol instead of "classic" menus in a main window? hmmm, you should ask jamest and jcater on that topic. jamest: ? imho I like it, it has something of an explorer :) i asked derek before, but he was seeing match, and not answered Yes, England <-> Argentina is a must :) don' tell the result !!! i want to see at home there will be a reply on M1 i asked that navigator question, becouse eugeneai (~eugeneai@62.76.23.142) joined #gnuenterprise. some users (accountants) hate using mouse and a main window with menus is maybe simpler do you think of a menu bar or of a kind of 1. 2. 3. list in the main window? menu bar with hot keys and if I understand you correct, the main window will show the form right? yes um the tree can be navigated via the cursor keys and we're just trying things out to see what people think IIRC the text mode makes you pick numbers or some such but there is 2 windows i dont want text mode :) 2 windows? 1 navigator - 1 form yip as navigator can launch forms, reports, executables anything you want really I have mine here launching IIRC perl scripts just remember though we're still very much so playing so if you want to see something just describe it i like only 1 main window with main menubar (navigator) like the old navigator? i don't know old navigator ah it was a snigle window that contained a menu menu=menubar? no menubar :( just a menu in the screen the menubar isn't even working in the current one btw - i think what you ask for is a menu that changes the form and it's all in one window correct? yes this wouldn't work for me at all as users have multiple forms open at same time not to say we can't do it but it doesn't work in my case no no i want 1 main window with menubar(navigator) and forms,reports are the child windows in it oh interesting idea what would happen in cases i launch scripts? like excel opened with more then one xls a shell window inside navigator? it's an interesting idea may have to do some more playing btw navigator takes ui drivers so nothing prevents us from making something like that in it's own UIwxmdi.py jamest: I tried to embbed the forms windows in the second windows of navigator today. This could be possible, but it requires some changes to the forms code the entire UI system is under reworking we need to start a wishlist yes, mdi and figure out everything we would like/require as I really would like to get most the UI stuff nailed down this next release i have to run for a bit Action: jamest is installing new systems in grad offices this morning siesel: your the one working on on the GComm right? yes do you have any of it documented? Like what functions need to be provided by each of the drivers and such? I justed started documenting it :) there is a docbook proposal in cvs great.... our cellar is flooded.. I asking because I am building some code in php to support xml-rpc and soap in much the same way. if you already have things defined then I can look at doing mines much more like yours siesel: where abouts in cvs is it? I was looking around cvs last night. Im not sure whats what in there at the moment it is too early to describe how to add new drivers, because the number of supported drivers is too small at the moment its in docbook/Proposals/GComm/ so I'm not quite shure which functions should be made general available for all drivers and which are private functions. Im mostly concerned with the basic functions developers wanting to the the GComm are supposed to use. Like what they do to start the client, how they package up their params (selecting data types), and how they make the calls, and then how the return comes back. ditto with creating a server. how they add functions, how they are documented if I can make mine look like a port. so that it basicly matches, then it will be more generally consistant. Syntax will be a little diff because of the diffs between python and php ok, thats quite easy. the only difference between python and php implementation will be, that python supports a kind of introspection, so I can just call server = GComm.attach("xmlrpc",options) and then seriveA= server.request("ServiceA") and then serviceA.serviceForMe(param1,param2) I dont get that php can support introspection as well... but I dont follow your example ok, sorry, I try again. :-) eugeneai (~eugeneai@62.76.23.142) left irc: Excess Flood if you deal with python you can just request an service object and then all method call on that service object will be transparently executed on the server. eugeneai (~eugeneai@62.76.23.142) joined #gnuenterprise. you mean it does that in general, or with these xmlrpc/soap calls? StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) joined #gnuenterprise. It should do this in general using any protocoll which GComm provides as a transport layer. StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) left #gnuenterprise. ok. so its something that GComm accomplishes. I can do the same with php yeah, you can create service/proxy objects with php too, sorry, I forgot :) OK. lets step thru your example from above real quick. params = { 'host': 'myserver.mydomain', 'port': 8765, 'transport': 'https' } server = GComm.attach('xmlrpc', params) this is from the docbook a the moment the api is quite simple: client: 1. atach(interface,options) 2. request('servicename') 3. transparent calls with GComm your not using a web server (aka apache). your opening your own socket port? and python has its own web server lib or something? yes, an own socket port is created yes. alexey_ (~alexey@195.151.214.34) joined #gnuenterprise. OK. this is not something I can do with PHP. Since in PHP I am stateless. I have to wait for the web server to kick me into life no problem, this should work also. It just means that you have to parse the description file one time per method call ;) wait. I have another question now or you keep a preparsed file in a php file Action: siesel is waiting I see that GComm.attach needs to know what protocol its opening up for I figured it would do some kind of detection of the incoming request no, GComm.attach is for the client driver. for opening a client request I can see doing that... but for incoming I think we can detect what kind of call it is and process from that ah! there is a GComm.bind which creates a list of servers which serve methods, objects etc. defined in a GRPC file then that makes sense. GComm.attach would be needed by the client to open up its session with the server ok. lets stick to client for a min. I think we are getting somewhere ;-) GComm.bind would make not much sense for php I think it might... but lets stick to the client for a sec now I understand your example better php cannot do that syntax as you describe. server.DonutProvider.Factory.getDonutBakery('glazed yeast') I might be able to do something similar tho ok I think it in this way: 1. client calls gcomm_attach and gets an server object returned $server = GComm->attach('xmlrpc', params); $server = GComm->attach('xmlrpc', $params); 2. $server->request('Donutplace'); jcater (~jason@w202.z065105010.mem-tn.dsl.cnc.net) joined #gnuenterprise. 2b. the $server object builds an php object with eval() and returns it (possible?) yes.... but where did Donutplace come from? server.DonutProvider.Factory.getDonutBakery('glazed yeast') there is a description of all supported methods stored in an xml file. after my attach I can do something like $server->execute('DonutProvider.Factory.getDonutBakery','glazed yeast'); in case of xmlrpc and soap there is no need for parsing that file on the client. (for CORBA there seems to be that need) ra3vat (ds@ics.elcom.ru) got netsplit. you mean that description is on the server side right? yes, you could do the execute stuff, which would be much faster than the "direct access" stuff I thought off. yes, on the server side ok. hang with me on the client side first. I have plenty to understand about the server side I cant do the direct access with PHP, since I need to predefine the functions into the class. I POSSIBLY could do that by getting the WSML from the server and genning the wrapper functions. and for xml-rpc maybe doing introspection on the server to get some wsml, but thats too dependant on the serice haveing things setup properly for me so the execute style would be easier yes. thats what I thought off. kind of a bummer that ours wont match better tho... next. I see in this: server.DonutProvider.Factory.getDonutBakery('glazed yeast') your passing 'glazed yeast' as the params to that function ra3vat (ds@ics.elcom.ru) returned to #gnuenterprise. how are you deciding what type of param it is? do you just make a decision that its a string, and so package it up as one? yes. how can you always make the right descision? what if they wanted that packages as base64? or maybe you have server.DonutProvider.Factory.getDonutBakery(5) and you decide its an int, but the server expected it passed as a string btami (~tamas@ip102-205.ktv.tiszanet.hu) left irc: "Client Exiting" ok, at the moment that is the client programmers problem. There no solution for that in the driver now. :( hmmm. how does the client programmer control this? something like server.DonutProvider.Factory.getDonutBakery(base64_encode('glazed yeast')) ? in case of string and int, by passing the types string, int. and then you get this string which is base64'd. how do you detect that and then slap the around in in the xml packet? because a GTypeCast('base64','my test') will return an object with a type attribut. so the driver can handle it. but this part is not implemented at the moment. whats GTypecast ? OK!! thats what I was looking for so if they dont pass you an object, you auto-detect right. if they use this future GTypeCast, they will be giving you an object that you can then get the type and data from perfect that shoulds good I can implement this in PHP now... on the return data... StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) joined #gnuenterprise. I should not show my face do you auto convert all of its return into python vars for them? there should be an optional 'enforce RPC Definition check' mode on the client side which parses a GRPC file and cares for absolute correct types. not after what happened today we not only didn't kick ass we've also been ass-kicked god save the queen StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) left #gnuenterprise. siesel: I dont think requiring such a thing on the client is nessesary if we use GTypeCast do you auto convert all of its return into python vars for them? drochaid (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) joined #gnuenterprise. Seek3r: ok, i change it to very optional yes, the client looks at the return types (which are given in xmlrpc and soap) and converts them into python vars perfect eugeneai (~eugeneai@62.76.23.142) left irc: Excess Flood eugeneai (~eugeneai@62.76.23.142) joined #gnuenterprise. how do you pass multiple params to a function? server.DonutProvider.Factory.getDonutBakery('glazed yeast', 'white bread') or server.DonutProvider.Factory.getDonutBakery(Array('glazed yeast',whitebread)) the first way. should work in php too. not really (PHP 4<) ah. yes, but php4 isnt GPL compatible so I need to live with PHP3 I would have to use the second for it to work properly ok. So it has to be done the ARRAY way. now. how open are you to adjusting your end so that we can match more closely? StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) joined #gnuenterprise. a was reading a bit of your chat... as in using server.execute('DonutProvider.Factory.getDonutBakery',array('glazed yeast','whitebread')) quite open. (if it doesn't means changing heaps of code) siesel: you know that with python >2.2 an int type is an object, too? madlocke. cool. I really like python :) there is a new basic type that all classes should be derived from. it's name is object. unfortunately 'object' is used as variable in some places... :( but look: >>> a=1 >>> print isinstance(a,object) 1 madlocke: doh!... that could be worked out easily. if an object is detected, then it can look for some object var which the GTypeCast would always set basicly, if an object is passed, then it would make sure its a GTypeCast object StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) left #gnuenterprise. for other stuff like int, it would deal with it as nessesary Seekr: It don't have to be that way, you could just write many a method definition like execute($string,$par1=None,$par2=None,$par3=None,.........$par12=None...) a=1 type(a) should still return integer siesel: yes, but then we have to set an arbitrary limit on how many input params can be used by a service aka, how many $par* do we have be back in a few. gotta get my son up for school seek3r: IÄve to go too. But will back in ~4-5 hours l8r siesel (jan@dial-194-8-205-46.netcologne.de) left irc: "Real IRC clients know "to" is a preposition, not an adverb" StyXman (~mdione@modem117-adsl.capfed2.sinectis.com.ar) joined #gnuenterprise. alexey_ (~alexey@195.151.214.34) left irc: "Client Exiting" darn eugeneai (~eugeneai@62.76.23.142) left irc: "Client Exiting" another question. Am i right in thinking that GComm and GNUe RPC are two different things? No GComm was the original working name before any coding was done and now its called GNU RPC? or GNUe RPC? um depends on who you ask :) we don't actually have an official name yet some refer to it as GNU RPC some GComm some GNUe RPC Action: jcater is the latter I dont like GNUe RPC... because then it seems to be a GNU Enterprise thing... it should be more generic than that but it is GNUe it's part of GNUe Common if we can do this right, then that is only one implementation the "standard" or "proposal" can be called whatever but our implementation I' d think we would call GNUe-Something I am gonna try building a PHP version today with the xml-rpc driver ah cool I'd think "GNURPC" would be the name of the "standard" GNURPC or GComm... I guess gnurpc is fine I suppose I can rename my functions later if nessesary ok. Im gonna try building the client real quick and see how it goes Im gonna use this for the new web services I am developing for DotGNU Seek3r: you want to integrate this with GNUe Forms or is this a different solution? madlocke: completely different ok. this is just the RPC stuff ok that i understand it right. from what I understand GNUe Forms is just for converting sql results to forms in some automatic way GNUe's Interesting Communication Interface Gic-ee Seek3r: it's more than that oooo that's gicky it is rendering dialog definitions madlocke: like XUL? jamest: not bad and allows access via datasources to almost any input/output system Seek3r: forms can be extended via plugins XUL, with a purpose :) ariel_ (~ariel@ppp-217-133-245-161.dialup.tiscali.it) left irc: Read error: 110 (Connection timed out) Seek3r: so you can output to anything you want. you can generate html, use qt, use wxgtk, ... madlocke: how is the interface described? I like XUL because its a nice reusable xml file xul is binded to one widget set imho it's an xml definition, too madlocke: only because there is one implementation so far... but I dont think its tied to mozilla at all madlocke: can you paste a small example of the definition file? forms is able to output even to a curses termina