| *** tpb has joined #tp | 00:00 | |
| *** ChanServ sets mode: +o tpb | 00:00 | |
| *** tuna-fish has quit IRC | 00:11 | |
| *** tuna-fish has joined #tp | 00:17 | |
| *** whacky has quit IRC | 00:29 | |
| *** llnz has joined #tp | 00:37 | |
| *** tansell-laptop has joined #tp | 01:43 | |
| tansell | back | 02:04 |
|---|---|---|
| tansell | ezod, I'm assuming you have gone? | 02:04 |
| tansell | Greywhind, sure we can chat about it | 02:08 |
| Greywhind | cool | 02:09 |
| tansell | Greywhind, we can do that now if you want | 02:09 |
| Greywhind | hmm... it's a bit late | 02:09 |
| Greywhind | 2:10 AM here | 02:09 |
| tansell | Greywhind, didn't think you would still be here :) | 02:09 |
| Greywhind | could we talk tomorrow evening? | 02:09 |
| Greywhind | maybe sometime around midnight my time? | 02:09 |
| tansell | that is about 2 hours earlier then now? | 02:10 |
| tansell | I might be around - best bet is to just keep poking me | 02:11 |
| Greywhind | ok | 02:14 |
| *** mithro has quit IRC | 02:20 | |
| *** nash has quit IRC | 02:25 | |
| *** mithro has joined #tp | 02:33 | |
| *** JLP_ has joined #tp | 02:44 | |
| *** JLP has quit IRC | 02:45 | |
| cherez | mithro: You around? | 02:51 |
| mithro | cherez: kinda | 02:54 |
| cherez | Is there any document of what libtpproto2-py needs done? | 02:55 |
| CIA-65 | llnz tpserver-cpp * rc152ee8cb9d7 /tpserver/ (playerconnection.cpp playermanager.cpp playermanager.h): | 02:55 |
| CIA-65 | Refactored the Player::findPlayer to be a big nicer. | 02:55 |
| CIA-65 | Shifted password auth to login() in PlayerConnection, removed lots of | 02:55 |
| CIA-65 | exception handling for the normal or expected cases. | 02:55 |
| CIA-65 | llnz tpserver-cpp * r87f4e13aac40 /tpserver/objectview.cpp: Fixed another missing bracket warning, this time in ObjectView | 02:55 |
| tansell | cherez, not really | 02:56 |
| tansell | I can tell you the point libtpprotot2-py is at | 02:58 |
| cherez | Well, I think it would help to have some idea when I write the proposal. | 02:58 |
| tansell | I would recommend doing the research to find out | 02:59 |
| cherez | That was my plan already. | 02:59 |
| tansell | libtpproto2-py was up to the stage where I could use it with tpclient-pytext | 03:00 |
| tansell | If I remeber correctly I could download objects in both tp04 and tp03 | 03:00 |
| cherez | libtpproto-py implements everything correctly? | 03:00 |
| tansell | correctly is a loaded word :) | 03:01 |
| tansell | there are no tests for libtpproto-py | 03:01 |
| cherez | Within the specification? | 03:01 |
| cherez | Mmm. | 03:01 |
| cherez | Well enough that the clients work? | 03:01 |
| tansell | there are some start of tests for libtpproto2-py | 03:01 |
| tansell | yes - enough that the clients work | 03:01 |
| cherez | Well, it's late now; I'll read through the libraries tomorrow. | 03:02 |
| cherez | Night, sirs. | 03:02 |
| tansell | libtpproto-2py uses the xml protocol definition | 03:03 |
| tansell | while libtpproto-py is hard coded | 03:03 |
| llnz | tansell: tpserver-cpp is in reasonable state for release, shoud I go ahead and release 0.6.1? | 03:05 |
| tansell | I think we have a reasonable binary for windows working | 03:05 |
| llnz | so is that yes? | 03:11 |
| *** peres has quit IRC | 03:16 | |
| *** konr has joined #tp | 03:17 | |
| tansell | yes | 03:17 |
| tansell | I think so | 03:17 |
| tansell | I'll do the debs tonight if you release before I finish work | 03:18 |
| llnz | cool | 03:21 |
| konr | Hi there! I'm interested in joining you guys for the SoC 2009 by working with bots and AI. If I got correctly, you have two working bots that use different heuristics and it you'd like to have new ones, am I right? | 03:23 |
| konr | s/it// | 03:23 |
| konr | s/s\/it\/\//s\/it \/\/s | 03:24 |
| tansell | konr, we would prefer you to improve the existing bots | 03:26 |
| tansell | doing things like, increasing the speed, adding support for new rulesets, etc | 03:26 |
| konr | tansell, interesting! what is the most promissing of the two? | 03:27 |
| tansell | daneel-ai at the moment | 03:29 |
| konr | wow, made in python | 03:30 |
| tansell | konr, why is that wow? | 03:31 |
| tansell | Python is the natural choice for AI programming | 03:31 |
| konr | tansell, I am currently messing around with python - and trying to create a bot too! - so that makes daneel much more interesting :) | 03:32 |
| konr | but the bot i was planning would just use some classification/outlier algorithms to filter content for my websight! | 03:34 |
| konr | I'm learning AI this semester - it's so exciting! | 03:34 |
| *** jainbasil has joined #tp | 03:39 | |
| konr | tanselll, hmmm, is there any particular thing that is highly needed for daneel and doable in 3-months? | 03:40 |
| *** jainbasil has quit IRC | 03:40 | |
| tansell | konr, try looking at the ideas page | 03:40 |
| tansell | there are some ideas there | 03:40 |
| konr | tansell: at tp.net's "ideas for programmers" wiki page? It just says that there are many possible improvements and then proceeds by giving some examples of bots and AI. Could these suggestions be in other page? | 03:47 |
| tansell | this is a probably a good page to start at - http://www.thousandparsec.net/wiki/Daneel-ai | 03:49 |
| tpb | Title: Daneel-ai - Thousand Parsec Wiki (at www.thousandparsec.net) | 03:49 |
| konr | Thanks! | 03:49 |
| CIA-65 | llnz tpserver-cpp * r23d4bfb759ac / (ChangeLog NEWS configure.ac): Prep for 0.6.1 release | 04:14 |
| *** Epyon has quit IRC | 04:17 | |
| *** peres has joined #tp | 04:20 | |
| *** mithro has quit IRC | 04:34 | |
| *** mithro has joined #tp | 04:48 | |
| CIA-65 | llnz web * rbb0819ba7ac9 / (3 files in 3 dirs): Added tarball and news item for tpserver-cpp 0.6.1 release. | 05:19 |
| *** gab32 has joined #tp | 05:30 | |
| tpb | New news from http://www.thousandparsec.net/tp/rss.php: TPServer-cpp 0.6.1 released. | 05:32 |
| *** Gaurav__ has left #tp | 05:33 | |
| *** Gaurav__ has joined #tp | 05:35 | |
| *** Greywhind has quit IRC | 05:36 | |
| *** Gaurav__ has quit IRC | 05:55 | |
| *** Gaurav__ has joined #tp | 06:15 | |
| *** Gaurav__ has quit IRC | 06:30 | |
| * llnz wanders off | 06:31 | |
| llnz | later all | 06:31 |
| *** llnz has quit IRC | 06:31 | |
| *** JLP has joined #tp | 06:34 | |
| *** JLP_ has quit IRC | 06:34 | |
| *** Greywhind has joined #tp | 06:51 | |
| *** gab32 has quit IRC | 07:02 | |
| JLP | good morning everyone | 07:02 |
| tote | morning :) | 07:03 |
| *** skiffcz-gsoc has joined #tp | 07:11 | |
| *** JLafont has joined #tp | 07:13 | |
| skiffcz-gsoc | oh, crap, so after wandering into the land of references, Im back to pointers. honeymoon is over | 07:47 |
| Vadtec | mornin | 08:00 |
| *** whacky has joined #tp | 08:09 | |
| JLP | Vadtec (and others writing proposals), in the proposal do mention your IRC nickname (so we can connect proposals to nicks here) and do mention where are you from (at least the timezone) | 08:24 |
| JLP | now that there is only one and Vadtec mentioned it when he submited it it is easy to connect with nickname, but later when there are couple of them it gets confusing for us | 08:25 |
| tote | does tpclient-pywx keep an open connection to the server? or does it reconnect on every turn? | 08:28 |
| skiffcz-gsoc | JLP: affirmative | 08:28 |
| JLP | tote: hm can't say for sure, but i think it also connects when adding/removing/editing orders | 08:30 |
| JLP | i think it is more, when needed | 08:30 |
| tote | okey so open a connection for adding an order, and then closing it is not a bad thing? | 08:31 |
| *** tuna-fish has quit IRC | 08:33 | |
| *** bddebian has joined #tp | 08:35 | |
| tote | is it always true that everyone in a game can see all objects? | 08:36 |
| bddebian | Heya | 08:38 |
| JLP | tote: in current version of protocol, yes, in version 4 the visibility and history of it will be added | 08:39 |
| JLP | bddebian: ahoy | 08:39 |
| bddebian | Hi JLP | 08:41 |
| skiffcz-gsoc | tote: Ive found something interesting regarding player connection | 08:41 |
| skiffcz-gsoc | Connection This class is the current network connection to a player. Is controlled and contained by the Network class. Holds details on the type of network connection and what version of the protocol the client is running, and creates the correct type of frame. Looks after the players socket and is responsable for sending and receiving Frames. Suports IPv6. | 08:41 |
| skiffcz-gsoc | its from server cpp documentation | 08:42 |
| *** whacky has quit IRC | 08:42 | |
| skiffcz-gsoc | so you may want to peek into this class or objects using it, if server actually closes the socket or not .. I guess the same approach will be used on client side | 08:43 |
| tote | skiffcz-gsoc: thanks will look into it | 08:43 |
| skiffcz-gsoc | the info that its held by the same class as C sockets (Network) would imply its created each time socket is opened .. but I havent looked into the Network class yet, as I have lot of my own business right now | 08:46 |
| tansell | JLP, https://build.opensuse.org/project/show?project=home%3Amithro%3Athousandparsec%3Adeb | 09:16 |
| tpb | <http://ln-s.net/33ba> (at build.opensuse.org) | 09:16 |
| *** Gaurav__ has joined #tp | 09:39 | |
| *** peres has quit IRC | 09:45 | |
| *** verhoevenv has joined #tp | 09:50 | |
| *** mithro has quit IRC | 09:52 | |
| *** mithro has joined #tp | 09:59 | |
| *** Black-Panther has joined #tp | 10:05 | |
| Black-Panther | Hi everyone! | 10:05 |
| Black-Panther | I'm really interessted in being one of your GSoC students! Who may I talk to? | 10:06 |
| JLP | Black-Panther: welcome | 10:07 |
| Black-Panther | hi! | 10:07 |
| Black-Panther | May I ask you a few questions? | 10:07 |
| JLP | Black-Panther: have you already read thru the gsoc/developers documentation | 10:07 |
| JLP | Black-Panther: sure | 10:08 |
| Black-Panther | I read those docs which are linked on http://socghop.appspot.com/org/show/google/gsoc2009/thousandparsec | 10:08 |
| tpb | <http://ln-s.net/3235> (at socghop.appspot.com) | 10:08 |
| JLP | Black-Panther: cool | 10:09 |
| Black-Panther | but I don't remember a documentation though | 10:09 |
| JLP | Black-Panther: so did any task from the idea list catch your interest? | 10:09 |
| Black-Panther | ... | 10:09 |
| Black-Panther | yes | 10:09 |
| Black-Panther | to be honest, there are two ;) | 10:09 |
| JLP | Black-Panther: nothing wrong with that | 10:09 |
| JLP | Black-Panther: you can always submit two applications :) | 10:10 |
| Black-Panther | maybe more too ;) | 10:10 |
| Black-Panther | so... my strength are C++ and webprogramming... | 10:10 |
| JLP | yeah, but it's good to focus on not too many and get those as polished as possible | 10:10 |
| Black-Panther | sure | 10:10 |
| Black-Panther | for this reason, I wanted to talk to somebody | 10:11 |
| Black-Panther | to know on what to focus | 10:11 |
| Black-Panther | I read about a web-interface you'd like to have | 10:11 |
| Black-Panther | now I'd have some technical questions | 10:12 |
| Black-Panther | (only to know if I'm capable to complete my work) | 10:12 |
| Black-Panther | how does the PHP-Script would interact with the program? | 10:12 |
| Black-Panther | is there an api specified, or is there to write a new protocol? | 10:13 |
| Black-Panther | should I use an existing one? | 10:13 |
| Black-Panther | ... | 10:13 |
| Black-Panther | ;) | 10:13 |
| JLP | there is some PHP protocol library in our git repository | 10:13 |
| JLP | so you would probably use that and improve and extend it | 10:14 |
| Black-Panther | so the comunication between server and webclient exists already (basic things)? | 10:14 |
| tote | between server and webclient no, but between server and a client yes :) | 10:15 |
| Black-Panther | what I meant was, if there are already PHP-routines which do establish a connection to a server? | 10:16 |
| JLP | Black-Panther: well i don't know how far php library is, so some stuff is probably already there and some not | 10:16 |
| JLP | Black-Panther: best you take a llok at code and probably mithro knows more if he's still around | 10:16 |
| * mithro should be in bed | 10:17 | |
| mithro | like 3 hours ago | 10:17 |
| tote | :) | 10:17 |
| Black-Panther | ^^ | 10:17 |
| mithro | Black-Panther: if you are thinking of writing a web client in php I would really urge against it | 10:17 |
| mithro | we do have a very limited protocol library written in php | 10:17 |
| mithro | but it's mainly for the metaserver | 10:17 |
| mithro | a Python based web client makes a load more sense because a python library already exists | 10:18 |
| tote | http://xvid.se/tp.jpg <- this is the problem im having atm, the connections between the webinterface and the server, it will be a connection per client | 10:18 |
| mithro | and is well test | 10:18 |
| Black-Panther | I see... | 10:18 |
| mithro | used by 3 existing clients | 10:18 |
| mithro | tpclient-pywx, tpclient-pyogre and daneel-ai | 10:18 |
| mithro | tote: yes, you are correct you'll need one connection per client | 10:19 |
| tote | there isnt a way to have only one connection against a server, and send user specific commands? | 10:19 |
| tote | mithro: okey | 10:19 |
| mithro | tote: as a web client is just like every other client | 10:19 |
| mithro | my suggestion would be to have a background daemon which syncs a mysql database with the tpserver-cpp via the tp protocol | 10:19 |
| mithro | and then have the web side read from the mysql database | 10:20 |
| tote | hmm intresting | 10:20 |
| Black-Panther | should work, I guess | 10:20 |
| mithro | as remeber downloading the universe is rather expensive | 10:20 |
| mithro | actually, you could just use libtpclient-py's prexisting cache mecanism - but that requires you to load the whole cache into memory for each operation | 10:21 |
| mithro | s/operation/web request/ | 10:21 |
| Black-Panther | but if the webclient has only to read from a MySQL database, then it makes sense to write it in PHP | 10:21 |
| mithro | unless you use something like fast cgi | 10:21 |
| mithro | Black-Panther: it makes sense to write it in a langauge we can maintain into the future :) | 10:21 |
| Black-Panther | sure | 10:21 |
| Black-Panther | so Python would be your choise? | 10:22 |
| tote | Black-Panther: why does anything makes sense to write in php? :P | 10:22 |
| mithro | plus writing the sync side still needs to be done in python | 10:22 |
| mithro | tote: the tp website is something which makes sense to write in php | 10:22 |
| mithro | it's a glorified server side include :) | 10:22 |
| tote | haha sure if you put it like that :) | 10:22 |
| Black-Panther | :D | 10:22 |
| tote | mithro: ah well thanks for your input, gonna think about it, expect a proposal in a couple of hours :) | 10:23 |
| Black-Panther | what would you like to have done using C++? | 10:24 |
| mithro | JLP: can you add the following to the ideas page | 10:24 |
| mithro | * Convert the existing wiki+website into something like iki wiki | 10:24 |
| mithro | cavets - this requires writing a bunch of scripts to do data conversion | 10:24 |
| mithro | and reworking the website html/javascript and converting the current php | 10:25 |
| tote | what data needs converting mithro? | 10:25 |
| mithro | tote: wiki is mediawiki so that needs to be converted | 10:25 |
| mithro | and we want to preserve history | 10:26 |
| tote | ok | 10:26 |
| mithro | we want to move to a new system which is wiki based - but uses a git backend | 10:26 |
| mithro | anyone interested in the singleplayer scenario's idea? | 10:26 |
| mithro | I think that would be a great project for someone | 10:26 |
| mithro | JLP: we should do a news post about things we are intersted in having happen | 10:27 |
| Black-Panther | is it possible to write it in c++, then yes! | 10:27 |
| mithro | IE the singleplayer scenario | 10:27 |
| mithro | Black-Panther: nope, mainly Python | 10:27 |
| mithro | Black-Panther: C++ stuff is focused around tpserver-cpp | 10:27 |
| Black-Panther | hmmm... what would you like there to be done? | 10:28 |
| mithro | unittesting is a big one in my eyes | 10:28 |
| mithro | more persistence backends | 10:28 |
| mithro | lots of general cleanup (but this needs to be better specified in the proposal) | 10:29 |
| Black-Panther | but for unittesting someone should be quite familiar with the existing code...? | 10:29 |
| mithro | pulling a lot of the code which is duplicated in each ruleset into a common library they can use | 10:29 |
| mithro | Black-Panther: somewhat familiar - but there is time to do so | 10:29 |
| mithro | s/do so/become so/ | 10:30 |
| Black-Panther | tp isn't a small project though... | 10:30 |
| mithro | improving the risk ruleset would be a good project | 10:30 |
| Black-Panther | and by improving you mean what? | 10:30 |
| mithro | unittesting is a big one in my eyes <- unittesting the tpserver-cpp core | 10:30 |
| mithro | Black-Panther: adding different extra rules | 10:31 |
| mithro | IE Ever played risk 2010? | 10:31 |
| mithro | maybe add dynamically changing links between countries | 10:31 |
| Black-Panther | sadly not... only the normal risk | 10:31 |
| Black-Panther | but I played many games like it ;) | 10:31 |
| mithro | Black-Panther: is there anything about risk you have said "ooh, it would be so much better if" | 10:32 |
| mithro | now is your chance | 10:32 |
| mithro | but the risk ruleset isn't risk in the normal sense either | 10:32 |
| Black-Panther | if I could change some rules, I would change a lot ;) | 10:32 |
| mithro | Black-Panther: btw - you should checkout the singleplayer stuff and try a risk game locally | 10:32 |
| mithro | Black-Panther: this is about incremental development - not starting from scratch | 10:32 |
| mithro | git clone git://git.thousandparsec.net/git/tpclient-pywx.git | 10:33 |
| mithro | git clone git://git.thousandparsec.net/git/daneel-ai.git | 10:33 |
| mithro | git clone git://git.thousandparsec.net/git/tpserver-cpp.git | 10:33 |
| mithro | build tpserver-cpp | 10:33 |
| Black-Panther | what's that? | 10:33 |
| mithro | and run tpclient-pywx and you should be able to play singleplayer | 10:33 |
| Black-Panther | ah, ok | 10:33 |
| mithro | works on most linux's and mac's | 10:33 |
| mithro | we should have windows binaries and a release by the end of the week | 10:34 |
| tote | mithro: you dont happen to know if there is a libtool package for debian for the new version required by tpserver-cpp? :) | 10:34 |
| Black-Panther | perfect... I'm linux only on VM | 10:34 |
| mithro | tote: there should be | 10:34 |
| mithro | Black-Panther: I hope that is Linux with a Windows VM :P | 10:34 |
| Black-Panther | not really... | 10:34 |
| tote | mithro: all i got from squeeze is 1.5.26-4 :/ | 10:34 |
| tote | so i cant build tpserver-cpp | 10:34 |
| mithro | you'll be required to do most of your development on Linux (or atleast test throughly on LInux) | 10:35 |
| mithro | as your mentors will be using that to check your code | 10:35 |
| mithro | tote: llnz uses debian | 10:35 |
| mithro | it should just work | 10:35 |
| mithro | you did a fresh checkout like above right? | 10:35 |
| Black-Panther | If I get accepted, then I'll install one ;) | 10:35 |
| Black-Panther | but till now the VM is enough | 10:35 |
| mithro | not using the older packages from the web? | 10:35 |
| mithro | I just built tpserver-cpp packages for debian 4.0 and 5.0 without problems | 10:36 |
| tote | mithro: yep | 10:36 |
| mithro | what was the problem again? | 10:36 |
| tote | one moment | 10:37 |
| mithro | tote: http://download.opensuse.org/repositories/home://mithro://thousandparsec://deb/Debian_5.0/ | 10:37 |
| tpb | <http://ln-s.net/33d1> (at download.opensuse.org) | 10:37 |
| mithro | i'll import them into our normal repository tommorrow | 10:37 |
| tote | mithro: http://rafb.net/p/SKDb7j24.html | 10:38 |
| tpb | Title: Nopaste - No description (at rafb.net) | 10:38 |
| tote | and configure exits with error: config.status: error: cannot find input file: Makefile.in | 10:39 |
| mithro | btw - the killer way to get into GSoC is to submit patches before GSoC starts :) | 10:40 |
| mithro | (real patches - not I want to get into GSoC patches) | 10:40 |
| verhoevenv | Tsk. :) | 10:40 |
| mithro | tote: problem a | 10:40 |
| mithro | libtoolize: unrecognized option `--recursive' | 10:40 |
| mithro | Try `libtoolize --help' for more information. | 10:40 |
| mithro | problem b | 10:41 |
| mithro | configure.ac:63: required file `./ltmain.sh' not found | 10:41 |
| mithro | hey verhoevenv | 10:41 |
| mithro | verhoevenv: I'm excited to see you get more contributors to daneel-ai thanks to GSoC :) | 10:41 |
| mithro | verhoevenv: hopefully we'll get some decent applications for it | 10:42 |
| Landon | mithro: would you say putting tech tree support in the wxwidgets client is a whole summer project? | 10:42 |
| tote | mithro: ye probably :) | 10:42 |
| verhoevenv | mithro: Yeah, would be nice | 10:42 |
| mithro | Landon: it could be | 10:42 |
| mithro | tote: what happens if you change the libtoolize line in autogen.sh | 10:43 |
| mithro | to | 10:43 |
| mithro | libtoolize --automake --copy --force --ltdl | 10:43 |
| Landon | ok, I'll mark that as yes :P since I'm still pretty unfamiliar with the client code | 10:43 |
| mithro | ? | 10:43 |
| mithro | Landon: it depends on how you go about it | 10:43 |
| CIA-65 | mavrinac gentoo-overlay * rc89ab9da191e /games-server/tpserver-cpp/ (ChangeLog Manifest tpserver-cpp-0.6.1.ebuild): Version bump tpserver-cpp to 0.6.1. | 10:43 |
| mithro | Landon: I wouldn't say it's a high priority however | 10:43 |
| mithro | Okay for those who are listening - here is what I'm excited about | 10:44 |
| Landon | are there any client improvements that are higher priority? I'll see if I can't get two proposals out before the end | 10:44 |
| mithro | 1) Helping out with the 3d client | 10:44 |
| mithro | 2) Helping out with daneel-ai | 10:44 |
| tote | mithro: seems to be working now | 10:44 |
| mithro | 3) A web client (but I need to see a *very* strong application) | 10:45 |
| mithro | 4) proper testing of libtpproto2-py | 10:45 |
| verhoevenv | Eh? What is libtpproto2-py? | 10:45 |
| verhoevenv | tp04? | 10:45 |
| mithro | verhoevenv: it's a rewrite of libtpproto which uses the protocol.xml rather then been hand written | 10:46 |
| verhoevenv | Ah, kay. | 10:46 |
| verhoevenv | Good. | 10:46 |
| mithro | verhoevenv: there will be very small changes to libtpclient-py when it gets ported | 10:46 |
| mithro | but I doubt it'll be much of a problem for you | 10:46 |
| mithro | as you smartly choose to use libtpclient-py which both tpclient-pywx and tpclient-pyogre now use | 10:46 |
| mithro | so if I break that, I break three people :) | 10:46 |
| verhoevenv | It was the obvious choice, really. :) | 10:47 |
| mithro | I think everyone is moving or already moved to using the protocol.xml file | 10:47 |
| mithro | libtpproto2-py uses it | 10:47 |
| mithro | libtpproto-java does | 10:47 |
| CIA-65 | mavrinac web * r1784bdbcead3 /downloads/downloads.xml: Update linux-gentoo version of tpserver-cpp to 0.6.1. | 10:47 |
| mithro | libtpproto-rb | 10:47 |
| mithro | libtpproto-rb does as well | 10:48 |
| mithro | Landon: If you are intersted in Python and client side stuff, the 3d client is probably the way to go | 10:48 |
| mithro | I can't think of any big high priority project for tpclient-pywx | 10:49 |
| mithro | apart from maybe a battleviewer | 10:49 |
| Landon | ok, didn't realize the 3d client was in python | 10:50 |
| mithro | Landon: yeah - only way it was able to be done in 3 months :) | 10:50 |
| Landon | hehe | 10:50 |
| *** peres has joined #tp | 10:54 | |
| mithro | Landon: the 3d client is an example of what a top student can do in the 3 gsoc months | 10:55 |
| mithro | A strong app for a low priority project will win over a weak app for a high priority project | 10:55 |
| mithro | we are also very open to ideas from students | 10:55 |
| mithro | as it shows that the student has taken enough time to evaluate where the project is defficent and figure out a way to solve it | 10:56 |
| mithro | Landon: if you really want to work on tpclient-pywx I would recommend a project to extend the single player support to include scenarios and multiplayer | 10:57 |
| mithro | tote: send tp-devel and email about your experiance with --recursive and see if lee has a way to fix it both both cases | 10:57 |
| Landon | mithro: I'll keep it in mind | 10:58 |
| mithro | Landon: if you have time submit multiple proposals | 10:59 |
| mithro | we have had a couple of cases where two students have submitted for the same thing | 10:59 |
| tote | mithro: send tp-devel? | 10:59 |
| Landon | yup | 10:59 |
| mithro | tote: s/and/an/ | 10:59 |
| mithro | tote: tp-devel mailing list | 10:59 |
| tote | aha | 10:59 |
| mithro | Landon: where both proposals where very strong - but of course we can only accept one student to do it | 10:59 |
| tote | just gonna step out a sec, do it when i get back | 10:59 |
| mithro | Landon: if they had a second strong proposal they would have been likely to get in on that instead | 11:00 |
| mithro | anyway I should be in bed | 11:00 |
| mithro | any last questions before I go? | 11:00 |
| mithro | you have 5 minutes | 11:00 |
| *** Gaurav__ has quit IRC | 11:00 | |
| skiffcz-gsoc | mithro: I saw your list of "what am I excited about", and it quiet makes us cpp people and not-so-much pythonists a sad pandas :) | 11:01 |
| mithro | the sooner you submit applications the sooner you can get feedback and improve them | 11:01 |
| mithro | skiffcz-gsoc: well I have a biased towards the python side of the project | 11:01 |
| mithro | A strong app for a low priority project will win over a weak app for a high priority project <------- that goes for C++ too | 11:02 |
| mithro | A project to extend risk is pretty good | 11:02 |
| mithro | A project to help unittest the server is good | 11:02 |
| mithro | A proposal to improve tpserver-cpp including a bunch of smaller related projects is good | 11:03 |
| mithro | I'm not really "excited" about them, but I still think they are important :) | 11:03 |
| verhoevenv | Urgh. If you dare extend risk, then also extend the daneel-ai rulefile. Don't leave that to the poor ai developer. :( | 11:03 |
| ezod | working on MTSec is a good one | 11:03 |
| mithro | verhoevenv: ha :) | 11:03 |
| mithro | ezod: yeah - I had forgotten about that | 11:03 |
| mithro | working on MTSec is a really good project too | 11:04 |
| mithro | and if ezod is up then it means I should have been in bed many many hours ago | 11:04 |
| mithro | ezod: hey :) | 11:04 |
| mithro | ezod: btw - get the invite? | 11:04 |
| ezod | mithro: hallo - up quite late again eh? i've been up 2 hours now, and i slept in :) | 11:04 |
| ezod | yes i did | 11:04 |
| verhoevenv | He should have been in bed 3 hours ago.. Well, that's what he said an hour ago. :) | 11:05 |
| mithro | verhoevenv: yeah that is about right | 11:05 |
| mithro | ezod: spent most of the night building tpserver-cpp debs | 11:05 |
| mithro | oh ezod was a c++ man | 11:05 |
| mithro | if you have c++ questions proke him | 11:05 |
| mithro | that is a combination of poke and prod :P | 11:06 |
| mithro | anyway I'm really gone now | 11:06 |
| ezod | later :) | 11:06 |
| skiffcz-gsoc | gn | 11:06 |
| mithro | submit early, submit often! :) | 11:06 |
| * ezod did both c++ and python for gsoc last year | 11:06 | |
| skiffcz-gsoc | ezod: ok, ill try to unwrap my thoughts a bit, please bare with me, because Im not much familiar with possibilities of python | 11:09 |
| skiffcz-gsoc | lets say, that if I want to build a game on top of tp | 11:10 |
| skiffcz-gsoc | standalone, one package game, with client, inbuilt server, AI | 11:10 |
| skiffcz-gsoc | focused primarily on win platform | 11:10 |
| skiffcz-gsoc | my first choice would be cpp based pieces, the server, some cpp client, obviously cpp protocol, cpp bots | 11:11 |
| skiffcz-gsoc | put this together, throw some media files on top, compile, ship | 11:11 |
| * JLP comes back from a quick nap | 11:12 | |
| ezod | skiffcz-gsoc: sounds ambitious for 3 months | 11:13 |
| skiffcz-gsoc | nono, i dont mean it as a proposal :) | 11:13 |
| ezod | ah | 11:14 |
| ezod | ok then | 11:14 |
| skiffcz-gsoc | but trying to see it from a view of someon wanting to build an actual game using the tp framework | 11:14 |
| ezod | well, here's what i'd do, if i wanted to do what you just described | 11:15 |
| *** Gaurav__ has joined #tp | 11:15 | |
| ezod | we are currently working on releasing the new tpclient-pywx with singleplayer (within the week, i hope) | 11:15 |
| ezod | the win32 version in particular will ship as a bundle with tpserver-cpp and daneel-ai, and basically only include minisec and risk, for which only the latter is supported in daneel-ai | 11:16 |
| ezod | so | 11:16 |
| ezod | i could do something like: | 11:17 |
| ezod | 1. write the ruleset for tpserver-cpp for the game i want to create | 11:17 |
| ezod | 2. make custom media files, branding, etc. for tpclient-pywx | 11:17 |
| ezod | 3. write daneel-ai rules or a new ai client for my ruleset | 11:17 |
| ezod | 4. bundle and release as a single game | 11:18 |
| ezod | no need to reinvent the wheel :) | 11:19 |
| ezod | although it is definitely possible to implement your own server, client, and/or ai if there's some reason to | 11:20 |
| ezod | with the existing protocol and client libraries you can work in a variety of languages interchangeably (with more or less success depending on the maturity and features of said libraries) | 11:21 |
| *** Black-Panther has quit IRC | 11:22 | |
| skiffcz-gsoc | hm, I still somehow feel wierdness around installing python along and ave an interpreted client :) (possible cause - my mind is not free, lack of experience, etc.) | 11:23 |
| ezod | skiffcz-gsoc: py2exe, if you're referring to win32 | 11:24 |
| ezod | and on linux, python apps are perfectly fine | 11:24 |
| skiffcz-gsoc | ah, py2exe, It didnt occurred to me .. lack of experience then :D | 11:25 |
| verhoevenv | For development, Python works remarkably fine on Windows. | 11:29 |
| verhoevenv | For deployent, py2exe also works remarkably fine. | 11:29 |
| verhoevenv | Although that's only what I've heard, I never used py2exe. | 11:29 |
| *** BarryCarlyon has joined #tp | 11:30 | |
| skiffcz-gsoc | verhoevenv: I guess in my case its the prejudice about interpreted languages as good dev tools, but .. once I had to make a dynamic page. It was simple calculator. Ive written cgi-script in C.... so yeah, Im this kind of person ... | 11:31 |
| ezod | verhoevenv: i'm about to use py2exe on daneel-ai today | 11:32 |
| ezod | skiffcz-gsoc: i was always a big C coder | 11:32 |
| ezod | then about 4 years ago i found python | 11:33 |
| ezod | it is what programming ought to be | 11:33 |
| verhoevenv | It really doesn't make much difference. | 11:33 |
| verhoevenv | You just skip the manual compilation step. | 11:33 |
| ezod | yeah, that and 98% of the boring work | 11:34 |
| ezod | i do stuff in python in 5 minutes that would take days in C | 11:35 |
| verhoevenv | Well yeah, that's the difference between Python and C. | 11:36 |
| verhoevenv | If you have a basis in Python, I highly recommend the Python Challenge (google around for it) | 11:36 |
| skiffcz-gsoc | bub but .. the compiler doesnt apply its magic on your code, and and .. its not magical! ..You guys cause my world to tremble. Well, but if there is another world, why not :) | 11:36 |
| verhoevenv | There's still a compiler. It just works more in the background. Don't worry too much. :) | 11:37 |
| verhoevenv | It's a lot more magical even. :P | 11:37 |
| ezod | python performance hit is almost negligible in most cases because generally all the really intensive stuff is coded at the back end in C | 11:37 |
| ezod | and you can do that yourself too - i do all the time | 11:37 |
| ezod | oh yeah, python challenge, i did that a while back | 11:39 |
| skiffcz-gsoc | ok, ive did the 2^38 in google calculator, now lets dig up my old c++ caesar decipherer .. it must be here somewhere ... | 11:42 |
| skiffcz-gsoc | j/k ... thanks for the tip, Ill try it out once I finish the proposal thing and the mess around it ;) | 11:43 |
| BarryCarlyon | Good Local time everyone, just hunting thru the project ideas for various groups and saw the web based client for tp, though it rather interesting and a challenge so thought I would drop in and say hi! | 11:44 |
| BarryCarlyon | So | 11:44 |
| BarryCarlyon | Hi! | 11:44 |
| ezod | hi :) | 11:44 |
| JLP | BarryCarlyon: ahoy | 11:44 |
| *** Patrick` has joined #tp | 11:45 | |
| JLP | Patrick`: wb | 11:45 |
| Patrick` | guess which random bloke talked me into not being a moron | 11:45 |
| Patrick` | you might have met him :) | 11:45 |
| JLP | Patrick`: hm, let me think ... | 11:45 |
| ezod | Patrick`: kamaelia added to my gentoo overlay, if you care :) http://tinyurl.com/ezodovl | 11:45 |
| tpb | Title: mavrinac.com :: gentoo overlay (at tinyurl.com) | 11:45 |
| Patrick` | nice | 11:46 |
| Patrick` | swing by the IRC channel, it's pretty idle but MS is happy to hear from users | 11:46 |
| BarryCarlyon | I applied for kamaelia last year tis quite good | 11:46 |
| verhoevenv | skiffcz-gsoc: you're missing the point of the python challenge I think ;) | 11:47 |
| Patrick` | BarryCarlyon: I did it 2 years ago, that's what got us started | 11:47 |
| BarryCarlyon | Really cool! | 11:47 |
| skiffcz-gsoc | verhoevenv: nah, I was just kidding ;) | 11:47 |
| skiffcz-gsoc | verhoevenv: Ill try it for sure, when this is over ;) | 11:48 |
| skiffcz-gsoc | verhoevenv: thank you for tip ;) | 11:48 |
| *** peres has quit IRC | 12:13 | |
| *** whollycow has joined #tp | 13:26 | |
| *** mhilmi has joined #tp | 13:51 | |
| tote | anyone except mithro that knows how the cache in libtpclient-py works? cant seem to find any docs for it | 14:27 |
| Patrick` | hah, I finally get myself set up and of course wx is GTK only :/ | 14:53 |
| *** whollycow has joined #tp | 14:56 | |
| skiffcz-gsoc | Patrick`: X11 | 14:56 |
| *** whollycow has quit IRC | 14:57 | |
| *** whollycow has joined #tp | 14:59 | |
| Patrick` | skiffcz-gsoc: what about it? | 14:59 |
| whollycow | (pardon my joining and leaving, irc client was being flaky) | 14:59 |
| skiffcz-gsoc | Patrick`: uh, nothing, I totally misread it, trash it ;) | 15:00 |
| Patrick` | neat | 15:00 |
| Patrick` | I was getting paranoi | 15:00 |
| Patrick` | I finally restore my linux install to do some work, and it's only going to stick out like a sore thumb and call in a squillion packages I ain't got | 15:00 |
| *** tuna-fish has joined #tp | 15:01 | |
| Patrick` | or, uh, three | 15:01 |
| skiffcz-gsoc | :) | 15:01 |
| Patrick` | hey | 15:11 |
| *** Noughmad has joined #tp | 15:12 | |
| Patrick` | since I'm only going to be doing python dev at this point and using the cpp server for comparism of features etc, I think I'll be ok with a pre-packaged 0.4.1 right? | 15:14 |
| Patrick` | or is it virtually irresemblable to the current repository | 15:14 |
| Patrick` | http://packages.thousandparsec.net/key.asc 404s | 15:16 |
| *** Noughmad has quit IRC | 15:16 | |
| *** Noughmad has joined #tp | 15:16 | |
| *** llnz has joined #tp | 15:17 | |
| Patrick` | but it doesn't really matter, the packages just won't install on a non-supported ubuntu distro | 15:19 |
| Patrick` | heh, looks like I'll be running the server on windows - the irony! | 15:19 |
| skiffcz-gsoc | eer | 15:19 |
| Patrick` | it's not a case of technically possible | 15:20 |
| skiffcz-gsoc | just FYI, the demasorvers are running now | 15:20 |
| Patrick` | I just don't want faffing around with linux and building and packages to eat my "SoC evening" | 15:20 |
| Patrick` | yeah, I've been on the demo server before briefly, but I wanted to poke/configure it a bit more | 15:20 |
| skiffcz-gsoc | ah | 15:21 |
| skiffcz-gsoc | well, I managed to compile the 0.6.0 server on fresh ubuntu in some half an hour or so | 15:21 |
| Patrick` | it seems as though building it myself will actually be the least faffy way to do it | 15:22 |
| skiffcz-gsoc | it really isnt any big pain | 15:22 |
| Patrick` | yeah, but there's just os much faff | 15:23 |
| Patrick` | it's an iterative process to get the build scripts to fail at successively later points | 15:23 |
| Patrick` | "oops, I need to install a custom readline library" | 15:23 |
| Patrick` | my fault for going with 8.10, no packages for it | 15:26 |
| *** Gaurav__ has quit IRC | 15:32 | |
| *** tuna-fish has quit IRC | 15:34 | |
| Patrick` | is there someone here who's run the python server? | 15:38 |
| Patrick` | the build is calling for tp.netlib >2.2, but I have installed python-tp-netlib 2.4.something | 15:39 |
| Patrick` | erm, not the build, just running the thing | 15:40 |
| skiffcz-gsoc | nope, cpp server only here | 15:45 |
| skiffcz-gsoc | sorry | 15:45 |
| llnz | Patrick`: mithro who developed it should be around in 2-3 hours | 15:46 |
| skiffcz-gsoc | mithro went to bed like 4 hours ago, so he wont be able to help you for some time | 15:46 |
| Patrick` | I have uni work | 15:47 |
| Patrick` | this is my only time to work on a proposal | 15:47 |
| Patrick` | but I already look like a huge noob | 15:47 |
| Patrick` | so I should go | 15:47 |
| tote | llnz: you have any knowledge on how the cache in libtpclient-py should be used? | 15:47 |
| Patrick` | everything seems incredibly fragile! | 15:47 |
| llnz | tote: no | 15:48 |
| llnz | Patrick`: the tpserver-py does run and it had a ruleset written for it last year | 15:48 |
| Patrick` | yeah, I'm trying to run it now | 15:48 |
| llnz | Patrick`: it probably isn't as well documented as it should be | 15:48 |
| Patrick` | it's dependant on a net library that needs to be installed from a deb, but isn't being recognised | 15:49 |
| Patrick` | and the process for building my own is an exercise in actually making me not want to be here | 15:49 |
| Patrick` | plus if I was doing a python project, I'd pretty much be asleep when mithro was up so it wouldn't be that great | 15:49 |
| llnz | installing for packages should work, but most people runs run from source | 15:49 |
| Patrick` | yes, but even running the server script is whining about a missing package | 15:50 |
| Patrick` | or a missing import | 15:50 |
| Patrick` | or something | 15:50 |
| llnz | JLafont: are you around? | 15:50 |
| Patrick` | I'll come back to it later | 15:50 |
| Patrick` | I need a break before i punch something | 15:50 |
| Patrick` | haha! | 15:55 |
| Patrick` | syntax error in the requirements script | 15:55 |
| Patrick` | *fix* | 15:55 |
| llnz | cool | 15:55 |
| Patrick` | it's Not Just Me Being Thick | 15:55 |
| *** shrink has quit IRC | 15:56 | |
| *** shrink has joined #tp | 15:56 | |
| Patrick` | do kindly ignore the petulant whining in the last page | 15:57 |
| Patrick` | I still haven't got it to work, butnow I'm actually doing something that *feels* useful | 15:58 |
| *** Gaurav__ has joined #tp | 15:59 | |
| ezod | we should probably make libtpproto-py a submodule of tpserver-py for inplace, like tpclient-pywx and daneel-ai | 16:05 |
| Patrick` | yes | 16:08 |
| Patrick` | in this instance, it's trying to import libtpproto-py files from "tp.netlib" | 16:08 |
| Patrick` | but tp is a subdir of the folder where the requirements script is :) | 16:09 |
| ezod | i would have thought it would continue on to site-packages if it didn't find it there | 16:09 |
| Patrick` | nope | 16:10 |
| ezod | but my assumptions about python importing have been burning me lately :( | 16:10 |
| Patrick` | manually running the import command works in every folder except there | 16:10 |
| Patrick` | yeah | 16:10 |
| Patrick` | that was one of the first things I got stung by | 16:10 |
| ezod | well tpserver-py is running for me | 16:11 |
| ezod | i symlinked netlib from my libtpproto-py checkout | 16:11 |
| Patrick` | yeah, that'd work as well | 16:12 |
| Patrick` | I'd rather fix the script, it'll get me back into practice | 16:12 |
| ezod | i'll see about getting that submodule in properly later | 16:12 |
| ezod | note that when installed tpserver-py will work fine (as it also lives in site-packages then) | 16:13 |
| Gaurav__ | hi llnz :) | 16:13 |
| llnz | hi Gaurav__ | 16:13 |
| Gaurav__ | just going through the persistence code | 16:14 |
| Patrick` | yeah | 16:14 |
| *** shrink has quit IRC | 16:15 | |
| *** shrink has joined #tp | 16:15 | |
| CIA-65 | mavrinac tpserver-py-dronesec * rdb0315868ea4 /setup.py: Removed tpserver-py-turn from scripts as it does not exist. | 16:35 |
| tote | user accounts, are those server specific or game specific? | 16:43 |
| ezod | there's a distinction? | 16:46 |
| tote | hmm, i guess not, nvm :) | 16:49 |
| ezod | at least i don't think there is currently | 16:50 |
| ezod | Patrick`: looks like dronesec branch of tpserver-py is what you want to use btw | 16:52 |
| ezod | it isn't just a ruleset branch - it has that submodule, among other things | 16:52 |
| Patrick` | so it's likely that any future work would build on from that if it was gonna get merged | 16:54 |
| ezod | yeah | 16:54 |
| Patrick` | now, I've not used git before, but I'm aware of how it differs from svn/what makes it awesome | 16:55 |
| Patrick` | so I'm thinking it could look in the main repository and not find that branch - does it live with someone? | 16:55 |
| Patrick` | aha, no, it's on the wiki | 16:56 |
| Patrick` | mwaha | 16:58 |
| Patrick` | ok, so this was a previous SoC and it's a unique ruleset as well as some other changes | 16:58 |
| Patrick` | oh I tell you what, dronsec reminds me of an indie game I played last month called dyson | 16:59 |
| Patrick` | it wasn't deep, it was realtime | 16:59 |
| Patrick` | but it was art. | 16:59 |
| Patrick` | JLafont: heard of it? | 17:00 |
| ezod | Patrick`: git checkout -b dronesec origin/dronesec | 17:04 |
| Patrick` | yeah, no, I'm in it now | 17:04 |
| ezod | oh ok | 17:04 |
| Patrick` | but cheers | 17:04 |
| *** Gaurav__ is now known as Gaurav__sleeping | 17:05 | |
| *** bcarlyon has joined #tp | 17:07 | |
| *** bcarlyon is now known as BCarlyon | 17:07 | |
| *** Gaurav__sleeping has quit IRC | 17:11 | |
| Vadtec | hi all | 17:18 |
| * BCarlyon waves | 17:20 | |
| Vadtec | JLP / llnz: either of you active? | 17:21 |
| llnz | yes | 17:21 |
| Vadtec | hi | 17:21 |
| JLP | Vadtec: barely | 17:21 |
| Vadtec | :o | 17:21 |
| Vadtec | i have a question | 17:21 |
| Vadtec | is it ok if I post my application in the wiki so I can track changes to it? | 17:21 |
| JLP | Vadtec: yup | 17:22 |
| llnz | sure | 17:22 |
| JLP | Vadtec: youple of people did it the same way last year | 17:22 |
| Vadtec | yeah | 17:22 |
| Vadtec | thats why i was asking | 17:22 |
| Vadtec | i didnt know if the staff had placed them there or not | 17:23 |
| Vadtec | i apologize for the lack of flow in my proposal, im not very good at writing them | 17:25 |
| JLP | Vadtec: no worries, we'll help you beat it into shape :) | 17:28 |
| JLafont | alrighty, back from work | 17:34 |
| JLafont | Patrick`: No I have not, lemme google it | 17:34 |
| Patrick` | JLafont: very pretty, worth an hour of your time | 17:37 |
| Patrick` | http://www.dyson-game.com/ | 17:38 |
| tpb | Title: Dyson (at www.dyson-game.com) | 17:38 |
| *** Azkar has joined #tp | 17:38 | |
| *** Jens_Lauble has joined #tp | 17:42 | |
| Jens_Lauble | irc://irc.freenode.org/#tp | 17:43 |
| llnz | welcome Jens_Lauble | 17:44 |
| *** Jens_Lauble has left #tp | 17:45 | |
| Vadtec | ugh, i have algebra AND english to finish tonight | 17:46 |
| Vadtec | at least algebra is only a few problems...they just take 1.5 pages to work | 17:47 |
| Vadtec | ok, im getting this algebra outta the way, cause the english is due by 2345 tonight | 17:50 |
| JLP | nice Eugene Tan and TP mentioned in the news about GSoC → http://www.zdnetasia.com/news/software/0,39044164,62052477,00.htm?scid=rss_z_nw | 17:50 |
| tpb | <http://ln-s.net/33mB> (at www.zdnetasia.com) | 17:50 |
| ezod | cool | 17:51 |
| llnz | very cool | 17:52 |
| Patrick` | cool | 17:54 |
| *** nash has joined #tp | 18:02 | |
| *** Epyon has joined #tp | 18:14 | |
| *** peres has joined #tp | 18:41 | |
| *** Noughmad has quit IRC | 18:48 | |
| *** Azkar has quit IRC | 18:50 | |
| *** whollycow has quit IRC | 19:38 | |
| *** mhilmi has quit IRC | 19:41 | |
| *** mhilmi has joined #tp | 19:42 | |
| *** Greg____ has joined #tp | 19:56 | |
| *** mithro has quit IRC | 20:02 | |
| *** mithro has joined #tp | 20:24 | |
| *** shenki has quit IRC | 20:33 | |
| *** Greg____ has left #tp | 20:37 | |
| tote | mithro: you there? how would i use the cache in libtpclient-py? | 20:54 |
| tansell | tote, you mean directly? | 20:59 |
| tansell | rather then a mysql db? | 20:59 |
| tote | yes directly and just store it in memory | 21:04 |
| tansell | see libtpclient-py/tp/client/cache.py | 21:06 |
| verhoevenv | Or see it used in daneel-ai. | 21:07 |
| tansell | tote, if you did end up using it directly you probably want to optimise it so you don't have to load the whole thing into memory | 21:07 |
| tote | why wouldnt i want to load the whole thing into memory? | 21:11 |
| llnz | tote: it could be quite large | 21:21 |
| llnz | but most AI methods require (or run better) with all infomation in memory anyway | 21:22 |
| verhoevenv | I dunno. It can't be a lot larger than what you transfer between server and client. | 21:26 |
| verhoevenv | And you don't want that to be of the order of megabytes I'd assume... | 21:26 |
| *** shenki has joined #tp | 21:26 | |
| verhoevenv | Night all. | 21:29 |
| *** verhoevenv has quit IRC | 21:29 | |
| tote | when i join the demo game with tpclient-pywx, does it download the whole universe then? | 21:29 |
| llnz | tote: it does | 21:29 |
| tote | and if im doing a webclient, i would be able to just let one users connection do that, and the other users can get the universe data from a cache? | 21:31 |
| shenki | llnz: have you stumbled across the shave patch? would it help if i re-tried posting it to the list? | 21:31 |
| llnz | shenki: you could either give me a url to the repository it's in, or post the patch directly to me | 21:32 |
| llnz | tote: no, each user has a different view of the universe | 21:32 |
| tote | llnz: hmm, thats a problem then :) | 21:34 |
| shenki | llnz: http://gitorious.org/projects/tpserver-cpp/repos/mainline/logs/pretty-build | 21:36 |
| tpb | <http://ln-s.net/33q-> (at gitorious.org) | 21:36 |
| llnz | shenki: can i give git that url? | 21:36 |
| shenki | (i wonder if the message never made it to the list because the patch was too large) | 21:36 |
| shenki | llnz: the git url is git://gitorious.org/tpserver-cpp/mainline.git | 21:37 |
| shenki | you want the 'pretty-build' branch | 21:37 |
| llnz | ok, i'll give that a go tonight | 21:40 |
| shenki | cheers | 21:41 |
| tote | llnz: so if 10 users are connected to the webclient, and there is a new turn, they all update the universe, so i need to keep a cache for every user or else there would be lots of traffic between gameserver and webclient? | 21:41 |
| llnz | yes | 21:41 |
| shenki | llnz: instead of grabbing the branch, you could also just cherry-pick 063eb15241960f9cb44696fad8691435b1538b71 | 21:42 |
| shenki | as it's only one commit | 21:42 |
| llnz | by using the cache, you reduce the traffic between the webclient and the server | 21:42 |
| llnz | shenki: will do | 21:42 |
| cherez | What the blazes is a semi-signed integer? | 21:44 |
| cherez | mithro: You are mad with power! | 21:45 |
| llnz | cherez: it's an unsigned integer that the maximum value (ie, -1) is special | 21:46 |
| cherez | Ah, I see. | 21:46 |
| shenki | so you can't really store 2^32 values, it's 2^32-1? (the top one reserved to represent -1?) | 21:47 |
| llnz | yes | 21:47 |
| shenki | interesting | 21:47 |
| * shenki has not heard of such a beast before | 21:47 | |
| shenki | is it a c++ thing? | 21:47 |
| llnz | shenki: nope, mithro made it up | 21:47 |
| cherez | It's a libtpproto thing, it seems. | 21:47 |
| shenki | haha | 21:48 |
| shenki | cherez: +1 to mithro being mad with power :) | 21:48 |
| cherez | Sweet. | 21:48 |
| llnz | in tpserver-cpp there is a #define for the value of it | 21:49 |
| llnz | i think it's in libtpproto-cpp as well | 21:50 |
| tote | llnz: yes, thanks, gonna rethink a small part of my proposal :) | 21:50 |
| llnz | tote: it does complicate some things, but makes some things easier | 21:51 |
| tote | indeed :) | 21:54 |
| nash | llnz: I will note however ALL the user of semi-signed int could be replaced by MAX_UINT, and it would make a hell of a lot more sense, without having to come up with new terms to describe an error code. | 22:04 |
| llnz | nash: true, i didn't suggest it | 22:05 |
| nash | llnz: And it also means you don't have this magical ambiguity over data-types on the wire... but that's another issue | 22:07 |
| llnz | tpserver-cpp and libtpproto-cpp treat Sint32 as uint32 and Sint64 as uint64 | 22:08 |
| llnz | no real ambiguity | 22:08 |
| nash | yes, they should... just the protocol spec makes it ambiguous. | 22:10 |
| *** peres has quit IRC | 22:18 | |
| *** bddebian has quit IRC | 22:57 | |
| *** rsisnett has joined #tp | 23:13 | |
| *** Gaurav__ has joined #tp | 23:13 | |
| rsisnett | Hi, Could anyone help me with a weird error in libtpproto while compiling parsek? | 23:15 |
| llnz | rsisnett: i assume you mean libtpproto-cpp? | 23:16 |
| rsisnett | yes, I indeed | 23:16 |
| rsisnett | yes indeed* | 23:16 |
| llnz | i can try to answer you question (since i wrote libtpproto-cpp), but I will be away for an hour or so soon | 23:17 |
| rsisnett | ok hope it's simple | 23:21 |
| rsisnett | rsisnett mainwindow.cpp:125:cannot allocate an object of abstract type 'LoggerWiidget': | 23:21 |
| rsisnett | loggerwidget.h:30: note: because the following virtual functions are pure within 'LoggerWidget': | 23:22 |
| rsisnett | logger.h:26: note: virtual void TPProto::Logger::error(char*, ...) | 23:22 |
| rsisnett | logger.h:33: note: virtual void TTProto::Logger::warning(char*, ...) | 23:22 |
| rsisnett | I know what that means, but why is this happening? | 23:22 |
| llnz | it might be that the interface (ie, the baseclass) changed | 23:22 |
| llnz | the loggerwidget logs messages from the library | 23:22 |
| llnz | impement the missing methods | 23:23 |
| llnz | s/impement/implement/ | 23:23 |
| rsisnett | mmm ok | 23:25 |
| rsisnett | thank you | 23:25 |
| llnz | bbl | 23:41 |
| *** llnz has quit IRC | 23:42 | |
| *** whacky has joined #tp | 23:47 | |
| *** konr has quit IRC | 23:58 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
