Monday, 2007-04-30

*** tpb has joined #tp00:00
*** ChanServ sets mode: +o tpb00:00
*** mithro has joined #tp00:02
nashrbr00:43
nashSort of like brb, but the reverse00:43
mithronot quite sure what the reverse of be right back is01:00
nashAm right back01:03
mithrowas left forward?01:05
mithrowonder if Lee will turn up today - with his thesis due and stuff01:06
nashWhen is yours?01:20
nashmithro: sqlalchemy.exceptions.InvalidRequestError: Cant get DBAPI module for dialect '<sqlalchemy.databases.sqlite.SQLiteDialect object at 0x84535ac>'01:31
mithrohrm01:37
mithroapt-get install python-sqlite?01:37
mithronash: i havn't seen that one before01:37
nash  File "/usr/lib/python2.4/site-packages/sqlalchemy/databases/sqlite.py", line 163, in __init__01:41
nash    self.supports_cast = (sqlite is not None and vers(sqlite.sqlite_version) >= vers("3.2.3"))01:41
nashAttributeError: 'module' object has no attribute 'sqlite_version'01:41
nashSo what do I need to upgrade then?01:41
nashmithro: ?01:43
mithronash: hrm01:43
mithrodpkg --list | grep sql01:43
nashbaboon:[/amd/baboon/local/nash/work/tranquillity/tpserver-py]% dpkg --list | grep sql01:43
nashii  libdbd-sqlite3-perl                 1.13-1.1                             Perl DBI driver with a self-contained RDBMS01:43
nashii  libhsqldb-java                      1.8.0.7-1                            Java SQL database engine01:43
nashii  libsqlite0                          2.8.17-2                             SQLite shared library01:43
nashii  libsqlite3-0                        3.3.8-1                              SQLite 3 shared library01:43
nashii  python-sqlalchemy                   0.3.6-2                              SQL toolkit and Object Relational Mapper for01:43
nashii  python-sqlite                       1.0.1-6                              python interface to SQLite 201:43
nashI can paste you more backtrace if you like01:43
mithrohrm....01:43
mithrotry "python-pysqlite2" ?01:44
nashWell it printed out an unbelievable amount of crap01:47
nashIs that normal?01:47
nashLike it just blew the guts out of a 2000+ line scrollback buffer01:48
mithronash: on first time you run - yes01:48
mithroit has to setup the database and it has very high level of debug on01:48
mithroplus tpserver-py spews a huge amount of debugging01:49
nashokay01:51
nashmithro: You said galaxie segfaulted?  It exits, but doesn't segfault for me - shoudl I be using the dev branch of tpserver-py?01:53
mithrothere is no dev branch of tpserver-py here01:53
mithroyeah - i get a segfault01:53
nash\hmmm01:53
mithroyour galaxy is probably empty?01:53
nashI wonder...01:54
mithroyou need to use tpserver-py-tool to populate it01:54
* nash mutters about ease of setup...01:54
mithrobut i thought it was segfaulting on the new account creation01:54
nashOkay - think I've found it01:55
nashI just added a couple of assers01:55
nashInteretingg they don't trip straigh away01:55
nashIs tpserver-py not flushing it's input straight away?01:55
nashThis is probably the relevant bit:01:56
nash/* FIXME: Need to check the access worked */01:56
nashwhere access == logging in01:56
mithronash: hrm? re: flushing01:58
nashI'm not getting the responses straight away - some wait until the next time I send01:59
mithronash: that is a bit weird02:01
mithronash: can you log a bug?02:01
nashAnyway - I've updated, jsut about to push02:02
nashmithro: Well I'll double check it's not my issue02:02
mithronash: i think i know the cause of the problem02:02
nashokay02:03
nashSo you want a bug?02:03
mithronash: can you put a very long pause between your sends and make sure it still occurs?02:03
CIA-30nash galaxie * r5be07fef77e2 /themeinterface.txt: Updates the themeinterface files02:03
CIA-30nash galaxie * rf0497a2897c5 / (8 files in 3 dirs): Merge with git+ssh://nash@tranquillity/home/nash/work/thousandparsec/eclient02:03
CIA-30nash galaxie * r884b0b0f785e / (14 files): Merge with git://git.thousandparsec.net/git/galaxie.git02:03
CIA-30nash galaxie * raff9d39b3fad / (Makefile tpe_comm.c): Fix segfault if login fails02:03
mithronash: yeah - i'm pretty sure I know what that problem is02:04
mithronash: btw cg is now (as of git 1.5 release) pretty unmaintained02:14
nashYeah, I recently found out... Still shouldn't be a major hassle for the short term - it's still 99% a wrapper anyway02:15
CIA-30mithro tpserver-py * r42f73782e82f /README: Small updates to the README file.02:15
CIA-30mithro tpserver-py * ra19823b610fa / (INSTALL config.py): Quick setup instructions for "inplace" operation.02:15
CIA-30mithro tpserver-py * r52c56a6ac512 /README: Finished the comment.02:15
CIA-30mithro tpserver-py * rcc4ff7feb9ab /.gitignore: Ignore the config file.02:15
mithronash: the git commands are getting very close to the cg-commants02:15
mithro1 assignment down, 2 assignments and 1 honours left02:16
nashmithro: Indeed.  Which means cg was probably on the right track in the first place ;-)02:17
mithronash: those instructions any better?02:18
nashAhh... mysql... still not a database after all these years...02:24
nashmithro: Still lacking things... like need to do all this other stuff in scratch pad.02:24
nashAlso a 'start a game' script would be extremely useful02:25
CIA-30mithro tpserver-py * r65080e7d2b8d /config.py: Revert the change to my local setup.02:25
CIA-30mithro tpserver-py * r7e7302e22eb4 / (INSTALL config.py config.py-template):02:25
CIA-30Move config.py to a template.02:25
CIA-30This should stop me from continually clobbering the SQLite default02:25
CIA-30configuration.02:25
* mithro becomes extremly lagged again02:26
mithronash: the stuff in scratchpad is only for developing02:30
nashokay02:30
nashSo how shoudl it work then?02:30
nashAlso... I just tried to use tpserver-py-tool02:30
nash(a most painful name)02:30
mithroapt-get install tpserver-py02:30
nash  File "/usr/lib/python2.4/site-packages/sqlalchemy/engine/__init__.py", line 113, in create_engine02:30
nash    return strategy.create(*args, **kwargs)02:30
nash  File "/usr/lib/python2.4/site-packages/sqlalchemy/engine/strategies.py", line 65, in create02:30
nash    raise exceptions.InvalidRequestError("Cant get DBAPI module for dialect '%s'" % dialect)02:30
nashsqlalchemy.exceptions.InvalidRequestError: Cant get DBAPI module for dialect '<sqlalchemy.databases.mysql.MySQLDialect object at 0x84b8c8c>'02:30
nashAnd for people who install from source?02:31
mithroor02:31
mithroeasy_install tpserver-py02:31
mithronash: ahh I proberly clobbered your config.py02:31
mithrohence the two extra pushes02:32
nashSo to recreate?02:32
nashmithro: It should really just work?02:32
mithrocg-update; cp config.py-template config.py02:32
nash2007-04-30 16:34:04,740 INFO sqlalchemy.engine.base.Engine.0x..ac {}02:33
nashTraceback (most recent call last):02:33
nash  File "./tpserver-py-tool", line 46, in ?02:33
nash    from config import netlib, db02:33
nashImportError: cannot import name netlib02:33
mithrohrm....02:35
mithroit appears I havn't updated tpserver-py-tool to work with the sqlalchemy changes :P02:35
mithroas I said - tpserver-py needs love02:44
nashOkay: So when I fail to login, is there any client parsable way of working out why?02:53
nashie "No such account" vs "Incorrect password"?02:53
mithronash: no - and probably for "security" there shouldn't be a different02:57
nashYep02:58
nashI know about account discovery02:58
nashStill makes it hard to work out what is going on - a classic security vs usability trade02:58
nashmithro: Okay - account creation... do I need to do any magic for that?03:05
* nash is going to do: May I login -> (if yes) -> Login -> if fails -> Create Account -> Login03:06
mithronash: you want to check the return from account creation03:19
nashWhy?  It tells you if the account exists?03:19
*** mithro has quit IRC03:34
*** mithro has joined #tp03:35
*** mithro has quit IRC03:56
*** nash has quit IRC04:07
*** mithro has joined #tp04:44
*** llnz has joined #tp05:25
mithrohey llnz05:28
mithroall handed in?05:28
llnzyeah05:29
mithrollnz: feels good?05:29
llnzyes05:29
llnzhow close are you to finished?05:30
mithro~2 weeks till I have to hand in my honours05:30
mithrothen i still have a couple of months of course work + exams left05:31
mithrofinish middle of this year05:31
llnzi have about a month of work, presentations and paper writing05:32
mithroend of july I'll be all finished05:33
mithrollnz: did you end up getting the post grad job?05:44
llnznope, noone did05:45
mithroouch, funding cuts?05:46
llnzyes05:46
CIA-30mithro scratchpad * r31d439f4edd4 / (temp temp2): Remove the temp files.05:51
mithrollnz: so what are your plans?05:52
llnzwriting paper and persentation and finding a job in May05:54
llnzstart work in June (or so)05:54
mithrowhat type of job you looking for? not going to work for yourself?05:54
llnzsoftware engineering or (linux) network admin05:55
llnzcurrently not going to work for myself, need an idea and someone to bankroll it06:01
mithrollnz: i was thinking more just odd network/sysadmin + software engineering contracts06:09
llnzmithro: hard to come across, small market06:09
mithrollnz: ahh okay, there seem to be enough around adelaide06:12
llnzmy current contact has finished, i should wander around to offer my services06:13
mithrollnz: might tide you over for a month or two06:15
llnztrue06:16
mithrodinner06:27
mithrobblr06:27
*** TBBle_ has joined #tp06:35
*** nash has joined #tp06:36
llnzhi TBBle_, nash06:38
nashheyo llnz06:39
nashthesis done?06:39
llnzyes06:46
Epyonnash, that "Art for MTSec page" you've sent me -- why not put it in the wiki?06:48
EpyonI can do it for you if I have the most recent version.06:48
nashEpyon: I wrote it up, and it was going up on the main webpage06:52
nashllnz: Congrats - happy with it06:52
llnzyes, it's finally done and it should be fine06:52
nashEpyon: however, you are welcom eot wiki it06:52
nashI don't think I've changet ut since I sent you a copy06:53
*** TBBle has quit IRC06:58
llnzhummm....06:58
llnzwhy am i getting "Invalid confirmation code. The code may have expired." when I try to confirm my email address06:59
llnz?06:59
llnz(in the wiki)06:59
nashllnz: You get one automatically when you first sign up07:01
nashThen if you click verify it invalidates the first07:01
llnzahh....07:01
llnzso i'm one message behind still07:01
nashprobably07:01
llnzahh, that worked07:04
Epyonwikified07:04
nashEpyon: Sweet07:04
Epyonnash, what are the requested sizes of the images?07:09
nashEpyon: I'm using 64x6407:10
nashHowever others may have other choices07:10
nashI think it's a good size07:10
EpyonUmmm.... you supplied FIVE missile types not SIX o.O07:11
nashEpyon: Because there is 5... I miscounted apparently when I said twelve (and hence 6 and 6)07:13
EpyonOkay, I'll correct it07:13
EpyonAre launch tubes and missile/torpedo racks separate components?07:13
Epyon11 launch tubes AND 11 racks?07:14
nashI did07:14
nashYes07:14
nashRacks hold, but don't fire07:14
Epyon7+10+6*6+2 = 55 images :P07:16
nashEpyon: Well I would imagine things like the tubes, and racks would be very similar except for either logo, or reusing the base/torp image07:17
mithronash: ping?07:17
nashmithro: pong07:17
Epyonyep, true07:17
mithro03:0607:17
mithromithronash: you want to check the return from account creation03:1907:17
mithronashWhy?  It tells you if the account exists?07:17
mithronash: because account creation can fail for a number of reasons07:17
nashmithro: Yep - what should it return in the various cases07:17
mithroOkay or Fail07:18
nashdoesn't tell you why?07:18
mithronash: the string does! but it's another one of those error code things07:19
nashString is useless07:20
mithronot to a user :)07:20
nashYeah, but the client doesn't know that it is useful to the user07:21
mithrowhy not? it's a failure to user input?07:21
nashNot necessarily: It may because the server wants the client to connect on a different port07:22
mithronash: then it should send a redirect frame07:22
nashI can hardcode that this failure means something, but not in a general case07:22
mithroin general - failures are (currently) design to be handed to a user07:23
mithrobtw i'm not why your push didn't generate multiple commit messages07:24
nashmithro: But I know specific case where a failure is not meant to be handled by the user - destroyed objects07:24
mithronash: yes - but that is the unushal case07:25
mithrobut the failure code stuff should make it a bit clearer?07:26
nashmithro: So in other words, the failure code is currently broken ;-)07:26
nashSeriously most error codes are bugs...07:26
mithronash: no, it's just not quite working in the way we thought07:27
nashmithro: I mean also in my client07:27
mithrobtw you could get a Fail frame in response to any operation07:27
nashyes07:28
mithronash: i guess you could say - any Fail which has error code != 4 (No such thing, The object/order/message does not exist) might be a good start07:29
nashmithro: Don't tell me - document it!07:30
llnztp04 would be a good time to rework the error codes07:30
nashllnz: I totally agree07:33
nashbbs07:33
nashback07:48
mithronash: you have commit access :P07:50
nashmithro: I thought I had before?07:50
mithronash: so you can document what you extract from me then :)07:50
nashAhh...07:51
nashNice punt07:51
mithronash: it seems obvious to me, so I don't see a need to document ;)07:51
nashlikely story07:59
nashhey - is there test/example code for tpcl - I want to start to implement it07:59
mithronash: only whats in the tpcl document - we really do need some test stuff08:01
nashokay08:02
nashIf I change the webpage - is it updated automatically?08:03
nashmithro: ?08:04
mithrollnz: does the script automatically checkout the web repo?08:04
llnzyes08:05
CIA-30nash web * r2caabcdd1046 /dev/documents.php: Add link for TPCL/NCL of dev document page.08:08
nashCool08:08
nashDon't see website update08:09
llnzgets updated everyt 15 minutes08:09
nashAhh08:09
llnz(unlike darcs, which was immediate)08:09
nashllnz: Why not a post-update hook?08:09
nashTo make it immediate?08:10
nashmithro: llnz: Are there any other tpcl examples in the servers for any of the rulesets?08:10
llnznash: because i haven't got that far08:11
llnzhummm....08:11
llnzyes, minisec has tpcl08:11
nashllnz: Fair enough...08:11
nashllnz: Also - cool - I'll pull them out as examples08:12
llnzall in minisec.cpp08:13
llnzin the property and component setup08:13
nashYep - got them08:18
nashCool08:18
nashthanks08:19
* llnz wanders off08:23
llnzlater all08:23
*** llnz has quit IRC08:24
nashAnyway - time for me to go too08:36
nashnight all08:36
mithrosee ya08:38
*** tuna-fish_ has joined #tp09:49
*** tuna-fish_ has left #tp09:49
*** dmpayton has joined #tp16:02
*** nash has quit IRC17:22
*** mithro has quit IRC18:36
*** nash has joined #tp18:37
*** mithro has joined #tp18:57
mithromorning people18:59
mithrowhat is up today?18:59
nashheyo mithro19:00
nashHow is life?19:00
mithronash: busy busy 2 weeks till i finish what has been suck my life away19:01
nashheh19:01
nashCould be worse... it could be 1 week away19:01
mithrowell I think it's closer to 1.519:02
mithro16th of May when I give the presentation19:02
nashokay19:03
mithrohave to hand up the final doco then too19:16
mithroso then I just have to catch up on all the course work I missed19:17
nashAhh yes... that is fun19:18
* nash remembers that19:18
nashbut most others are in the same boat19:18
mithronash: not really - i started my project mid-year, most start begining of the year19:19
nashOh okay19:19
nashSo you had the advantage last year?19:19
CIA-30mithro tpserver-py * rb085b3eb6e07 / (tp/server/bases/User.py tpserver-py-tool):19:21
CIA-30Hacked tpserver-py-tool so that it works again.19:21
CIA-30All the parts now work - but the virtual hosting is borked.19:21
mithronash: yeah, plus I got the very long summer break to work on the project19:21
nashdid you?19:21
mithrowhich means i've acheived 80% of my target rather then the ushal 30%19:21
mithronash: well, I'm the only person to produce to commercially produced PCB boards19:22
nashheh19:23
nashcool19:23
nashDystopicFro: BTW: You around?19:23
mithronash: i'll post pictures on my blog when it's all over19:24
mithrothe device is pretty cool19:24
nashCool19:24
nashWhat is it?19:24
mithroa 8 line telephone to computer interface to be sold commercially for around $US15019:25
nashSo does that include ADCs and the like?19:25
mithroit uses National Semiconductor CODEC's for all the ADC/DAC stuff19:26
mithrofound to be the cheapest way to do it19:26
nashSo it could be used to give you decent dial up speed?19:26
mithrofilters where just too expensive19:26
mithronash: no - the other way around, allow you to have a seperate telephone line in each room of your house19:27
nashSo 8 analogue lines in, to 1 digital upstream?19:29
mithroyes19:30
nashSo why couldn't I reverse that?19:30
nashi digital line in, to 8 upstream?19:30
mithronash: because interfacing to a POTS system is quite difference to pretending to be one19:30
nashAssuming the digital channel does it's own seperations19:30
nashRight19:30
mithroIE I can't detect a ringing signal, but I can generate one19:31
nashUmm.. isn't a ring like 60-90V ?19:31
mithronash: yes19:31
nashSo surely it's pretty easy to detect?19:31
mithroI produce a 80Vpp ringing signal19:31
nashright on spec19:32
mithronash: how, our DAC has a maximum of 1.2Vpp input19:32
nashOverload condition?19:32
nashA high voltage zener ?19:32
mithroyou need quite different "analog" interface19:32
mithroit wouldn't be a huge step to change it - you would only need to change the TIM module19:33
nashDid you name that one yourself?19:33
mithroTIM == Telephone Interface Module19:33
nashStill smells like a backronym19:33
mithroyou would want a PIM == POTS Interface Module19:34
mithrowhich I havn't designed - plus then you run into a whole bunch of evil AUSTEL regulations19:35
nashJust do what everyone else does a) ignore them or b) write in small text on the box 'for educational use only'19:35
nashAnd c) make sure it actually works... they do tend to follow 'no harm, no foul' rules19:36
mithronash: if I want to produce it commercially that isn't really an option19:36
nashAre you likely too?19:36
mithroJLP: ping?19:36
JLPmithro: pong19:36
mithroJLP: how did the RPM stuff go?19:37
mithroJLP: on SUSE build stuff19:37
mithroJLP: they are talking about it as the conference I'm currently at19:38
JLPmithro: http://software.opensuse.org/download/home:/JRepin/19:41
tpb<http://ln-s.net/JSZ> (at software.opensuse.org)19:41
mithroJLP: none of the python libraries yet?19:41
JLPnope not yet, i'm not even sure if i finished with cpp stuff19:42
JLPlooks like server is still missing19:42
mithroyeah - would be cool if we could get that working19:43
JLPyup, i can try over the holidays now, have two free days, starting tomorrow19:45
*** EdBoy has joined #tp19:53
EdBoydmpayton: boo19:53
*** EdBoy has left #tp20:07
DystopicFronash: around after a fashion, around in a more substantial fashion in a bit20:19
nashokay20:19
nashlet me know then ;-)20:19
*** mithro has quit IRC20:45
*** mithro has joined #tp22:46

Generated by irclog2html.py 2.5 by Marius Gedminas - find it at mg.pov.lt!