[00:30] Last message repeated 1 time(s). uday (uday@host-64-110-96-159.interpacket.net) left irc: [x]chat chillywilly (danielb@d126.as29.nwbl0.wi.voyager.net) joined #gnuenterprise. ra3vat (ds@195.239.66.3) joined #gnuenterprise. ra3vat (ds@195.239.66.3) left irc: Ping timeout for ra3vat[195.239.66.3] chillywilly (danielb@d126.as29.nwbl0.wi.voyager.net) left irc: Read error to chillywilly[d126.as29.nwbl0.wi.voyager.net]: EOF from client reinhard (rm@N802P014.adsl.highway.telekom.at) joined #gnuenterprise. neilt (neilt@dialup-63.208.64.233.Dial1.Chicago1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ hello all hello jamest (jamest@fh-dialup-201168.flinthills.com) joined #gnuenterprise. reinhard: you here? morning jamest morning i have a question you have a robots.txt file on the web site ok, I'm getting ready to go to grandparents for the day ok, robots.txt i keep submitting the web site to search engines and getting no response it excludes all robots what! do you mean to do that no can i fix it do you mind if i fix it sure, I'm there now though you want me to fix it ok, please remove it sure crap that looks that it may have been part of the nuke install as the timestamp is the same as the nuke files as for geas segfaulting I put a message in connections.c getFullClassDefinition right before the return call it's printed I put a message in schema.c schema_get_class (which is what is called by connections.c getFullClassDefinition) right before its return call it's printed so I assume something in there isn't liked by orbit but I don't have a clue about orbit also I get get schema.py to run ServerError: Class language::help was not found Traceback (most recent call last): File "./schema.py", line 253, in ? show_fields( c ) File "./schema.py", line 116, in show_fields show_help_info( f.classname, f.name) File "./schema.py", line 67, in show_help_info obj = get_help_object1( name, field ) File "./schema.py", line 61, in get_help_object1 if len(obj_list.objects) != 0: UnboundLocalError: Local variable 'obj_list' referenced before assignment ba i really need to run as wife is getting kids ready to go by herself jamest: i have solution to last and I'm the one in a hurry to leave when you get back so it's looking bad :) l8r damn missed the fun just when you turn away for a minute ... ;) dnWifey (derek@cpe-24-221-112-50.az.sprintbbd.net) joined #gnuenterprise. Nick change: dnWifey -> dnSoccer mdean (mdean@arc11x93.kcnet.com) joined #gnuenterprise. ToyMan (stuq@c5300-1-ip165.albany.thebiz.net) joined #gnuenterprise. later all neilt (neilt@dialup-63.208.64.233.Dial1.Chicago1.Level3.net) left irc: mdean (mdean@arc11x93.kcnet.com) left irc: Ping timeout for mdean[arc11x93.kcnet.com] mdean (mdean@arc11x41.kcnet.com) joined #gnuenterprise. ra3vat (ds@195.239.66.10) joined #gnuenterprise. hello all hey ra3vat any news? um news is reinhard agreed to do all the work for no pay Nick change: dnSoccer -> derek just those in the log last night :) reinhard: hello derek: that's old news ra3vat: ah yes i remember you manually extended sys.path for python you think there is any chance that you find out how python sets its path? reinhard: i'm trying. have no previous knowledge about that reinhard: may it be coupled with libpython2.0.a issue i had? ra3vat: that was? ra3vat: missing libraries when linking? something like that and i was needed to reconfigure python to get libpython2.0.a that works oh you built python from source or installed binary? ToyMan (stuq@c5300-1-ip165.albany.thebiz.net) left irc: Client Exiting i just took another libpython2.0.a ah ok so probably not your distro's fault then :) btw i found this Start Python and type import sys print sys.path This will give a list of directories where Python looks for modules. The file netcdfmodule.so must be in one of them. If it isn't, then you must either copy it, or add the directory where it resided to the search path. There are many ways of doing that: the environment variable PYTHONPATH, a file with the extension .pth containing the path in /usr/local/lib/python1.5/site-packages (or something similar, depending on your Python installation), a Python startup script that adds a directory to sys.path, and possible many other solutions. -- i could imagine that for example a python startup script is not executed for embedded interpreter btw may i be curious and ask what ra3vat stands for? it doesn't seem to be related to your real name? it's my real name in another great community - amateur radio :) my callsign ah and how is this built up? random letter number combination? or does it mean something? prefix preassigned for countries number gives region v assigned for our local region (say town) in russia at assigned continuosly and given to me Action: reinhard sometimes wonders how many things there are in the world he knows nothing about :) it's worldwide unique key like say libuuid :) :) sept, 11 neilt noticed his ham radio station some action happened in our country in 1990 1991 thought me that ham radio deserves to be kept forever yes i understand that i live very near swizerland and during ww2 some people here could listen to swiss radio which was actually prohibited by death penalty AFAIK and they heard quite different news as from "official" radio (swizerland was neutral in ww2) so i have a few spare from avonics, submarine and old police in soviet time short wave radio was only way to hear something not from our officials yeah i figured that did you learn english in school? then we forgot about it when tanks entered in Moscow in 1991 everyone took them from shelf yeah it was school where i started to learn english :) so you went to school after 1991 i guess i guess english was not tought in soviet schools? foreign languages always was tought in schools there was no way to use it ah so no one really knows it after school my mother told me they didn't learn english here in ww2 because england and u.s. were "our enemies" in ww2 i completed school in ancient 1982 so i'm probably the oldest bastard here :) hmmm not sure about neilt jamest and me are 31 neilt is older but i don't know how much lately he mentioned that he has grandchildren spent 2 years in 1983-1985 in army yuck and it was real possibility to take party in war in afganistan at that time strange how history repeats itself yeah as everyone even after fisrt grade of universities was taken into army brb ra3vat (ds@195.239.66.10) left irc: Ping timeout for ra3vat[195.239.66.10] Mr_You (rouzer@mankind.boredom.org) joined #gnuenterprise. ra3vat (ds@195.239.66.44) joined #gnuenterprise. hehe.. its funny to read someones online dream log.. ra3vat (ds@195.239.66.44) left irc: Ping timeout for ra3vat[195.239.66.44] Mr_You (rouzer@mankind.boredom.org) left irc: bleah neilt (neilt@dialup-63.208.67.236.Dial1.Chicago1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ neilt (neilt@dialup-63.208.67.236.Dial1.Chicago1.Level3.net) left irc: Ping timeout for neilt[dialup-63.208.67.236.Dial1.Chicago1.Level3.net] neilt (neilt@dialup-63.208.67.43.Dial1.Chicago1.Level3.net) joined #gnuenterprise. #gnuenterprise: mode change '+o neilt' by ChanServ!s@ChanServ Nick change: neilt -> kb9rre kb9rre: i guess you read the logs? my ham call sign kb9rre: the new parser already finds syntax errors the old one didn't notice cool purchasing.gcd:136: unknown class 'money' currency::money *price; ^ money is a type and not a class so it should be currency::money price; the old one was finding that i specifically made a check of that hmmmmmm sure? it's in cvs version of purchase.gcd this line idl has typedef right? nickr: i think yes but not sure nickr: yes reinhard: good catch money public price (field) Reference: load class currency::money where currency::money.sys_id = this._price that would not work certainly bbl Nick change: reinhard -> rm-away ra3vat (ds@195.239.66.7) joined #gnuenterprise. ra3vat (ds@195.239.66.7) left irc: [x]chat corba in perl rules that is all nah, if you going to do CORBA use structured cobol Har. I just wish ORBit had an objc binding :( me too Although whats cool about objc is that it could have a dynamic binding thing so you could have a thing that loads the idl and creats a class for it that you can subclass at runtime muah this is sort of what perl does. chillywilly (danielb@d164.as14.nwbl0.wi.voyager.net) joined #gnuenterprise. ra3vat (ds@195.239.66.7) joined #gnuenterprise. kb9rre here is ra3vat. good night dear friend. i'm glad to meet you first time. yur signal are five nine here in vladimir. my name is dmitry. back to you. kb9rre here is ra3vat. :) ? ra3vat this is kb9rre, how are you. you are 5 by 8 here in chicago. over :) heh smells like ham in here chillywilly: hello hi nickr: you can smell the channel? uday (uday@host-64-110-96-159.interpacket.net) joined #gnuenterprise. Nick change: kb9rre -> kb9rrz chillywilly: no offense or anything, but on occation you totally miss whats going on its been so long i fogot forgot :) nickr: whatever nickr: it was a joke, I think you're the one who missed it kb9rrz: yeah, me too :) uday (uday@host-64-110-96-159.interpacket.net) left irc: [x]chat Nick change: kb9rrz -> neilt nickr: and also I know that's neilt is being goofy and doing C.B. talk or something with ra3vat chillywilly: no its NOT CB! nickr: so what do I not get? oy what is it then? I said or something neilt: did you get my joke? nickr: nope. sorry chillywilly: ham radio all hail nickr licensed ham radio nickr (nick@sdsl-64-32-227-138.dsl.nyc.megapath.net) left irc: [x]chat neilt: oh, ok well I never did ham radio its the same difference as script kiddies and hackers eh? :) anyone can do CB it takes a pro to be a ham so now I am a script kiddie? no, not referring to you neilt: sure you weren't just explaining the difference between cb and ham chillywilly: chill, it was an example, you do know the difference between script kiddies and hackers, right neilt: umm, no I need guidance from you elders because I am just too young and foolish to know any better chillywilly: yes, you are a script kiddie in ham radio ;P chillywilly: will saying its CB is a first class insult to a ham that's the only thing I could think of as that's the only thing I ever used, I am not a hammer or whatever lol, you sure aren't not that that is necessaryily bad only used the C.B. couple of times too...just something not very interesting to me funny think is i think i have only used it a couple of times thing i registered and took the test, to prevent my freedom from being taken away maybe that's all it usually takes to pick up the "lingo" it used to be that you could talk to anyone in the world on ham radio and not pay any fee i'm script kiddie in ham radio too. i'm only proud i know some great guys who are hackers in there but governments are taking away that privilage and forcing people to use cell phones ack in a few more years the freedom to use the air waves will be gone well that's crap just like Free Software/Open Source will be squashed by SSSCA and legislation like that sort-of yes we can't let it happen you should fight them to the death everytime a new cellular frequency is opened up they remove some frequency from amateur radio yet amateur radio is credited with helping in almost every major disaster in the world we need to GPL our ham radio licences hmmm, and I wonder who is pushing for all of that guess what happens whem there is not power or telephone a ham sets up a protable radio and relays all traffic for emeergency recrue crews and families cool I know I would find your soft spot somewhere ;) just one the freedoms i dont want to lose one of many think how threatening it is, i can put a radio transmitter in my house that allows me to talk to anyone in the world yep of course they also have to have a radio and the atmosphereic conditions have to be right we need less control and more freedom but there are always relays and networks recent events are gonna screw up a lot of things neilt: i read your mail about xml-rpc and i think that xml-rpc is not what we are looking for er? SOAP == xml-rpc++ or so I have been told i have not looded at soap but xml-rpc is out IMHO for geas just ask mdean Action: chillywilly nudges mdean Action: mdean reminisces about Monty Python.... i would suggest reading gnue-discuss mail list email i wrote and reply if there are other thoughts well, did anyone read what we I was talking about doing with this Gcomm thingy that jamest brought up? I will send a mail about it to here and to DotGNU developers's list Nick change: rm-away -> reinhard neilt: wrt neil_test.gcd here being gnue-discuss i have major revission to neil_test.gcd that i am testing now ah the current cvs version of this file contains a lot of stuff we decided to not support any longer yes i know well it was a good thing as it was a good test for the error catching of the new parser :) I wonder if there's anywhere we can get the actual ODL syntax hmmm crap reinhard: will i checked it in, but forgot to compile it first so i hope it works methods_python.c:131: warning: no previous prototype for `python_execute_function' methods_python.c: In function `python_execute_function': methods_python.c:140: `obj' undeclared (first use in this function) methods_python.c:140: (Each undeclared identifier is reported only once methods_python.c:140: for each function it appears in.) methods_python.c:142: `methodname' undeclared (first use in this function) methods_python.c: In function `load_python_binding': methods_python.c:524: warning: assignment discards qualifiers from pointer target type bad neilt :P nevermind neilt: np i'll give you 20 seconds to fix it j/k lol Action: reinhard can easily play the big masta Action: reinhard is working on code that is not even linked into the executable :) jeeeze you had me going there dres (dres@4.18.171.42) joined #gnuenterprise. that code is my test code that was not supposed to be checked in for restructuring methods to use real arguments reinhard: fixed thanks neilt: that was done in record time :) mdean (mdean@arc11x41.kcnet.com) left irc: Ping timeout for mdean[arc11x41.kcnet.com] neilt: you just pointed me to a bug in the new parser code i defined that a symbol starts with a letter and then has 1 or more letters or digits instead of 0 or more so it didn't recognize "a" as a symbol cool where do i have an "a" parameter in test method ahh ttp://www.object-relational.com/articles/c++_and_object-relational_mapping.html er http://www.object-relational.com/articles/c++_and_object-relational_mapping.html reinhard: does you parser check for the maximum size of field name, class name, module name not yet i updated my test gcd's again and added a fail gcd file you may add anything you wish here i am not using it cool btw other than the old parser the new doesn't give up a file at first error cool that was on my list of things to fix so if you have say a typo in a typename I would be ok if you switched gcdVerifier over to use the new parser you have a class with chra test<30>; then the rest of the file is parsed ok and only this line is ignored cool does it report the error? nope, why gcdverifyer will be the first ting to work with the new parser we all know its an error :) lol cool neilt: i must admit that it seems to take longer than i expected reinhard: i mean if you want to go ahead and change config to compile it go ahead according to murphy's law plus because of real work issues i had lately i had not so much time :(( but i still belive it was right to do it as now every change in the parser we can think of seems so easy yes i think a lot of things will be better which before it seemed impossible now like typedef (alias) is already prepared and just has to be plugged in so to say cool plus i could learn a lot in doing this code shouldn't geas report the error and exit? if there's some parsing error no why not? report the error and continue on it's like if the c compiler would exit on your first error someday the parser will not run only at startu startup but also there will be a idl method to say "please reload gcd files" well I mean it parses and then displays all errors and exts at least i dreamt that last night ;) you dreamt about it? j/k you never want a server to exit hehe didn't you say that you dreamt about something gnue releated the other day? unless you explicitly tell it too chillywilly: imagine you just changed a gcd file in a company with 60 users well this is at startup I was thinking it's already ridiculous to tell them to log out of gnue while you restart geas the running case is different gcd updates will have to be doable while geas is running -- somehow but then the thought of having a bug in a gcd file reinhard: I know we wanna do dynamic updates and not getting geas to start again till it is fixed is not good so if you have a typo in a gcd file all other classes and all other fields in this class work as normal only the buggy field is not visible the errors are reported, itcontinues to parse, but classes with errors are not updated ? the fields where the errors are are not updated everything is updated except the error and geas continues on actually what you said but on field level not on class level if the errors is a class it is ignored exactly if the error is a field it is ignored ra3vat (ds@195.239.66.7) got netsplit. if the error is a methods it is ignoired if the error is that the gcd file doesn't start with the word "module" then the whole file is ignored we need to do a diff thing too as you don't want it to updatewhat has already been updated right? I guess that can be done later or something because in reality you will use gcdverifier to check the gcd stuff before you start geas neilt: sure but you may hack something up quick and not use it and you want to know if it is fscked up but don't forget even classes working with gcdverifier could bomb because of dependencies on other files if you copy the gcd file from one directory tree to another reinhard: that would be rare sure well actually i know myself because you should run gcdverifier on the same tree as geas is using when you're under pressure you sometimes don't test before you run ;) in fact gcd verifier should default to hat that ya never know what is going to happen, gcdverifier is just a tool to use to make sure your files are good right? i think gcdverifier should be extended files, syntax, dependencies, fields, methods etc well actually it is already more than a verifier gcd verifier actually parses the files using the same parser as geas it should be usable as a tool to nicely display your classes what it does now anyway no GUI dependiencies no not gui text mode maybe sorta like \d does in psql schema.py already does that, why does gcdverifier need to do this ra3vat (ds@195.239.66.7) got lost in the net-split. i would rather see us improve schema.py for this gcdverifier can do it without geas running as it also tests corba functionality well ok i am ok with that gcdverifier already displays that informations i sorta liked the fact that it does yes so do i but not interactively just to stdout omg you misunderstood me i meant not interactively i mean just like it does now cool only maybe a little better formatted yes i agree to better format as long as it is a complete format and with a switch to make it possible to create a dia file of it j/k the switch for a dia file is not hard to do i said i would do this if someone sent me the dia xml format this would take about 2 days to complete I cna send you one neilt: you sure? not just the manual think about the problem how to arrange the classes so that the arrows are well recognizeable chillywilly: i mean the exact xml we would generate for gcd's and not spaghetti thats an algorithm from knuth neilt: please don't tell me you have read knut h i think its directed graphs taocp yes, if its the same neilt: I have a UML diagram of a design, would that be useful? neilt: i have vol. 1 at home chillywilly: thats the problem, i dont have time to figure it out neilt: or should I make a UML dia gram of one of the gcds i am on page 30 or something like that i need someone that knows exactly what we need to do and can write it down neilt: and i don't have the time to read on reinhard: i have used knuth, but not read it through and through ah hmmm you know there's a gnu project for a MIX simulator? i have the fact that it is pascal and not C s/have/hate/ or assembler ra3vat (ds@195.239.66.7) joined #gnuenterprise. I wonder is dia has a DTD neilt: other question i didn't even think about the possibility that we could want two modules in one file does that really make sense? probably not, but on the other hand why have the limitation just in case the test files is probably the only place it makes sense ok i'll change it chillywilly: btw fyi an error message of geas looks like gcd_fail_test.gcd:44: unknown class 't1' fail::t1 [] a_list; ^ and this is probably one that neilt didn't want (?) gcd_fail_test.gcd:36: parse error, expecting `';'' type t1 ^ thats right that should pass no, that should not pass hence its in the file gcd_fail* reinhard: check the comments in those files ok both are correct to fail but you missed one line 43 or 45 should also fail i didn't post all my messages ahh i get for example this one: gcd_pass_test.gcd:78: unknown class 'class2' mod1::class2 [] a_list_field; ^ and class2 is really not defined afaict oooo, thats not good and for fail_test i get gcd_fail_test.gcd:36: parse error, expecting `';'' type t1 ^ gcd_fail_test.gcd:44: unknown class 't1' fail::t1 [] a_list; ^ gcd_fail_test.gcd:50: unknown class 't1' fail::t1 * a_ref; ^ btw missing ";" are very dangerous in your case the complete definition of type t1 is ignored ahh so we need another test type so the errors are not related fixed gcd_pass i think i need some sleep tomorrow is family day fixed gcd-fail see you on monday probably reinhard: later later all reinhard (rm@N802P014.adsl.highway.telekom.at) left irc: Those who say it can't be done should never stop those who are actually doing it. ra3vat (ds@195.239.66.7) left irc: [x]chat neilt (neilt@dialup-63.208.67.43.Dial1.Chicago1.Level3.net) left irc: later nickr (nick@sdsl-64-32-227-138.dsl.nyc.megapath.net) joined #gnuenterprise. mdean (mdean@arc10x80.kcnet.com) joined #gnuenterprise. chillywilly (danielb@d164.as14.nwbl0.wi.voyager.net) left irc: Ping timeout for chillywilly[d164.as14.nwbl0.wi.voyager.net] ajmitch (ajmitch@p20-max7.dun.ihug.co.nz) got netsplit. nickr (nick@sdsl-64-32-227-138.dsl.nyc.megapath.net) got netsplit. jamest (jamest@fh-dialup-201168.flinthills.com) got netsplit. derek (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit. dres (dres@4.18.171.42) got netsplit. Topic changed on #gnuenterprise by ChanServ!s@ChanServ: GNU Enterprise : http://www.gnue.org [If no one is home email info@gnue.org] #gnuenterprise: mode change '-o gnuebot' by ChanServ!s@ChanServ #gnuenterprise: mode change '+o gnuebot' by ChanServ!s@ChanServ ajmitch (ajmitch@p20-max7.dun.ihug.co.nz) returned to #gnuenterprise. jamest (jamest@fh-dialup-201168.flinthills.com) returned to #gnuenterprise. derek (derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise. dres (dres@4.18.171.42) returned to #gnuenterprise. nickr (nick@sdsl-64-32-227-138.dsl.nyc.megapath.net) returned to #gnuenterprise. --- Sun Sep 23 2001