#bitfighter IRC Log

Index Search ←Prev date Next date→

IRC Log for 2011-05-18

Timestamps are in GMT/BST.

00:44:00sam686 has joined
00:44:00ChanServ sets mode +v sam686
00:54:00sam686 Quit (Ping timeout: 258 seconds)
00:54:00sam686 has joined
00:54:00ChanServ sets mode +v sam686
01:26:00watusimoto Quit (Ping timeout: 260 seconds)
01:31:00raptor Quit (Remote host closed the connection)
02:26:00Helmic has joined
02:29:00karamazovapy Quit (Read error: Connection reset by peer)
02:29:00[1]sam686 has joined
02:29:00sam686 Quit (Ping timeout: 258 seconds)
02:29:00[1]sam686 is now known as sam686
02:33:00Helmic1 has joined
02:33:00Helmic Quit (Disconnected by services)
02:33:00Helmic1 is now known as Helmic
02:37:00[1]sam686 has joined
02:39:00sam686 Quit (Ping timeout: 258 seconds)
02:39:00sam686 has joined
02:39:00ChanServ sets mode +v sam686
02:40:00[2]sam686 has joined
02:40:00sam686 Quit (Disconnected by services)
02:40:00[2]sam686 is now known as sam686
02:41:00ChanServ sets mode +v sam686
02:42:00[1]sam686 Quit (Ping timeout: 258 seconds)
03:24:00sam686 has left
03:26:00sam686 has joined
03:28:00Helmic Quit (Ping timeout: 246 seconds)
03:29:00raptor has joined
03:29:00ChanServ sets mode +o raptor
03:29:00Helmic has joined
04:08:00karamazovapy has joined
04:13:00sam686:31] CTCP/VERSION request from raptor : version request?
04:13:00raptoryes - it tells me what IRC client you are using
04:13:00raptori asked because you Hydra folks seem to get a lot of ping timed outs...
04:13:00raptorasked = checked
04:13:00sam686Konversation 1.3.1 (C) 2002-2010 by the Konversation team. ok i see what you use..
04:13:00raptoryep
04:14:00sam686it's becuase i had a lot of internet disconnects today due to re-wiring the cables and cords.
04:14:00raptoroh ok
04:16:00sam686why is the word Hydra colored yellow for me?
04:17:00raptorprobably because your IRC client has an ego :)
04:18:00sam686bitfighterLogBot doesn't respond to version request..
04:19:00raptorhmmm... i thought it did
04:19:00[raptor VERSION]
04:19:00[raptor PING]
04:19:00raptorresponds to PING
04:20:00sam686your client won't respond to "userinfo""
04:20:00raptori don't have that option - is that WHOIS?
04:22:00sam686one thing i dislike from Hydra is it's very difficult to copy test in the chat test, all there is are "Copy Buffer", its difficult to select, and there is no copy selection.
04:22:00sam686i have to open chat log to copy, Hydra can log to file.
04:22:00raptorlike i say - konversation is the best client i've ever used
04:23:00raptorbut you'd have to install KDE for windows to use it
04:23:00sam686how big is KDE in download size / install size?
04:24:00raptornot sure - you can start by getting the installer: http://windows.kde.org/download.php
04:24:00raptorits a small launcher to download the rest
04:24:00raptorif you only select the app that you want, it should tell you the overall size
04:24:00raptorbut i'd expect it to be big
04:25:00raptoryou might want to try others - irssi (if you're crazy), chatzilla, mIRC
04:26:00raptorooo, i found this: http://www.silverex.org/news/
04:26:00karamazovapydoes anyone here have mumble installed?
04:27:00raptori think konversation was based on Xchat
04:27:00raptornever heard of mumble
04:27:00karamazovapywant to help me test?
04:27:00raptorwhat OS?
04:27:00karamazovapyhttp://mumble.sourceforge.net/
04:27:00karamazovapyany
04:27:00raptorsure
04:27:00raptorone moment
04:27:00karamazovapyI threw a server up on my vps - I want to see if it works
04:29:00sam686why download a 200 MB KDE for 1 MB konversaions? There are lots of IRC clients that are about 1 MB in size.
04:30:00raptorhaha
04:30:00raptoryeah, thought it would be high
04:30:00raptorit uses kdelibs and family
04:30:00raptorwhich is a whole desktop environment
04:30:00raptorok karamazovapy, i installed mumble
04:30:00karamazovapyadd 206.217.217.165 as a server
04:31:00karamazovapyonce you run through the startup crap
04:31:00sam686there is an option of writing custom IRC client, or using telnet as IRC (sounds crazy, having to input ping every minute)
04:31:00watusimoto has joined
04:33:00raptori connected
04:53:00karamazovapyguess I'll test later
04:55:00raptor Quit (Read error: Operation timed out)
05:07:00raptor has joined
05:07:00ChanServ sets mode +o raptor
05:07:00raptorok i'm back
05:07:00raptorapparently restarting my sound system shutdown the network subsystem, too
05:12:00raptorhey skype works now - i had to gut the pulse audio subsystem from my OS
05:12:00raptorand rely on good ol' ALSA
05:17:00karamazovapy...neat...?
05:17:00raptorso maybe mumble will work now?
05:31:00raptorkaramazovapy: so mumble 1.1 works ok
05:31:00raptorbut your server is 1.2 so i can't connect
05:43:00raptorsam686: did you find a better IRC client?
05:43:00sam686no..
05:43:00sam686haven't checked much...
05:44:00sam686i am mostly fine what i have.
05:53:00karamazovapyraptor - Error during SSL handshake: error:1408A10B:SSL routines:SSL3_GET_CLIENT_HELLO:wrong version number [13]
05:54:00raptoris that something you are seeing on your side?
05:54:00karamazovapythat's from the server log
05:54:00karamazovapydoesn't give any hints as to why your other connections wouldn't allow voice chat though
05:54:00raptorprobable from when i tried to connect with 1.1
05:54:00karamazovapyyeah
06:02:00raptorgood night folks
06:02:00raptor Quit (Remote host closed the connection)
06:05:00raptor has joined
06:05:00ChanServ sets mode +o raptor
06:06:00raptorone last message for watusimoto: the game server server is running at 80% RAM usage because of a squeezebox server and backend database taking 60MB of RAM
06:06:00raptorjust FYI
06:06:00raptorok good night
06:06:00raptor has left
08:29:00watusimoto Quit (Ping timeout: 250 seconds)
08:59:00sam686 has left
13:14:00Helmic has left
15:31:00raptor has joined
15:31:00ChanServ sets mode +o raptor
16:47:00watusimoto has joined
17:45:00watusimotobefore I broke it again, I had all editor items working at least to some extent
17:46:00raptoruh oh
17:46:00watusimotoso I'll be doing a little more rearranging, then i want to check what I have in
17:46:00watusimotostill needs more work, but it's clear that the new form works
17:47:00watusimotoand, I think, is better from a maintenance and logical flow point of view
17:47:00watusimotomore code reuse, closer ties between the editor and the game, etc.
17:48:00watusimotoI'd also like to consider using an XML like format for the level files
17:48:00raptoras long as maintenance is easier
17:48:00raptorugh, XML
17:48:00raptorhowabout JSON :)
17:48:00watusimotothat would make it easier to add new attributes, and make it easier to reuse level reading/writing code between different objects
17:48:00raptorXML means human readable
17:48:00raptorok XML
17:48:00watusimotowhatever
17:49:00watusimotohuman readable is important
17:49:00raptoryeah, so XML is the way to go
17:49:00watusimotowe we might want to start thinking about what that would look like
17:49:00raptorunless we do it in python?
17:49:00raptorthat's somewhere in the middle...
17:49:00watusimotoyou mean make the levels be in python?
17:49:00watusimotothey could be in lua
17:50:00watusimotoevery level could be a self-contained levelgen
17:50:00raptorhaha
17:59:00raptorugh, XML means XPATH
18:12:00watusimotomaybe there's something else
18:13:00raptori'm just trying to think of something that is simple enough to parse and put into our object model without too much layering/abstraction
18:16:00raptorOGDL requires whitespace
18:16:00raptornot sure i like that idea
18:21:00raptorthis looks promising: http://stackoverflow.com/questions/170686/best-open-xml-parser-for-c
18:21:00raptorrapidXML
18:21:00raptorwe don't need a lot of XML features
18:22:00raptorwe may not even need to use attributes
18:24:00watusimotohttps://secure.wikimedia.org/wikipedia/en/wiki/YAML
18:24:00watusimotocriteria I see are 1) easily parsable (i.e. library support) and 2) human editable
18:24:00raptorme too
18:26:00raptorYAML requires whitespace justification to be equal among siblings
18:27:00raptori don't know about you, but whitespace-dependent languages drive me up the wall
18:27:00raptorwe would also be requiring level creators to make sure their whitespace is in line properly with YAML
18:38:00watusimototrue
18:39:00watusimotoone advantage of xml is we might be able to bend our levels into something parsable (and displayable) via svg renderers
18:39:00raptorexplain
18:39:00watusimotoit would be cool to look at a level in a browser and at least see the general form of it
18:39:00raptorOOOoooo
18:39:00raptoryeah
18:39:00raptori forgot about that
18:39:00watusimotosince svg == xml (sort of), there may be a clever way
18:40:00watusimotoparsing isn't really a big deal since we'll use a library
18:40:00raptoryeah, i'm thinking xml is the way to go
18:40:00watusimotoI mainly want to be able to write attributes in different orders
18:40:00raptorthat rapidxml library looks promising: it's small and fast (they claim approaching the speed of c++ strlen)
18:40:00watusimotoso I can make the level reading/writing more modular
18:41:00watusimotospeed isn't even really an issue given the size of our files; that said faster is ebtter than slower
18:41:00raptoryes, that's true - i'm just thinking about sam's server with hundreds of levels, and k's server with levels that are crazy complex
18:42:00watusimotobut we only read them one by one
18:43:00watusimotoexcept at startup when we scan them all
18:43:00raptoryes, that
18:43:00watusimotolooking at pugixml
18:43:00watusimotohttps://pugixml.googlecode.com/svn/tags/latest/docs/quickstart.html
18:43:00watusimotoI think it would be helpful to take a simple level and mock it up in xml to see what it would look lik
18:44:00watusimotoand how much extra crap we'd need to add to make it renerable in svg readers
18:44:00raptorthe source is twice the size of rapid xml
18:44:00watusimotoI just chose it arbitrarily
18:48:00watusimotorapidxml looks good too
18:54:00Helmic has joined
18:58:00raptori'm putting together a test level that i will convert to a schema of sorts
19:07:00raptori'm always indecisive when it comes to using a tag or an attribute
19:09:00watusimotoI'm not sure...
19:09:00watusimotoI've managed to avoid xml this long...
19:11:00raptorfor a barrier, the points are in groups of two, right?
19:13:00watusimotox,y
19:13:00watusimotoyes
19:14:00watusimotoI see what you meant: <team>1</team> vs. <team = 1>
19:14:00watusimoto<point>x,y</point> v. <point x=44, y=66>
19:14:00raptoryep
19:15:00raptori'm about half done with an example
19:15:00raptorthen you can look over it and see if i missed anything or make anything more concise
19:15:00watusimotowell, it will give us a point of departure for our discussion
19:21:00raptorspeed zone is start (x,y) end (x.y) speed?
19:37:00raptorok
19:37:00raptori'm confused
19:37:00raptoris processLevelLoadLine() not used anymore?
19:37:00raptorhow are the objects loaded from the file now?
19:52:00sam686 has joined
19:52:00ChanServ sets mode +v sam686
20:06:00raptorok ready for my first iteration watusimoto?
20:06:00raptorhere is the original level file: http://96.2.123.136/upload/test.level
20:07:00raptorhere is the XML conversion: http://96.2.123.136/upload/test.level.xml
20:09:00raptora good read when thinking about cleanup up the data as far as using attributes/elements: http://www.w3schools.com/DTD/dtd_el_vs_attr.asp
20:09:00sam686in a current old loader, I added some form of adding some arguments (SnapEnable and R10) in any order (numbers nust stay in order) in SpeedZone::processArguments
20:15:00sam686really want to group polywall into polywalls? why quote numbers? <polywalls><polywall><point x="1" y="2"> ... </polywall></polywalls>
20:18:00sam686why have "minplayers" twice? <minplayers>4</minplayers> might be better with <minplayers = 4 /> or similar..
20:18:00sam686does description allow multiple lines?
20:18:00raptorto answer all of your questions
20:19:00raptorwe can do anything you want :)
20:19:00raptori do suggest raeding this first: http://www.w3schools.com/DTD/dtd_el_vs_attr.asp
20:23:00sam686One problem is having to keep compatible with older versions..
20:24:00raptorolder versions of XML or of the .level format?
20:24:00sam686maybe both, mainly .level
20:25:00raptoryes, we'll have to do an abstraction for reading in older levels, but we would only write to the new XML
20:25:00raptoras far as xml goes: using attributes like this: <minplayers count="4" />
20:26:00raptoras the advantage of being smaller code
20:26:00raptorhowever, using elements like this: <minplayers>4</minplayers>
20:26:00sam686i don't know if it make a difference..
20:26:00raptorhas the advantage of faster parsing
20:27:00raptorsee the section entitled "Avoid using attributes?" in that link i posted, above
20:27:00sam686what happens if quotes is missing? like in <point x=-1 y=-3 />
20:27:00raptorthen it isn't well-formed XML
20:27:00raptoranother bad point to attributes
20:28:00raptorso this may be better: <point><x>-1</x><y>-3</y></point>
20:29:00sam686XML looks like it have 2 drawbacks, 1. having to use same attribute name twice, 2. having to use quotes on everything including numbers.
20:29:00sam686well, its either 1, or 2, that might be a problem..
20:29:00raptoryes
20:29:00raptorbut the advantage of being easily understandable
20:30:00raptorand it doesn't really matter programming it - it only matters when manually editing the fiel
20:31:00raptorthat's why i'm thinking this format is better: <point><x>-1</x><y>-3</y></point>
20:32:00sam686another problem is how to warn users the level is corrupted, currently the level just goes blank if corrupted (missing gameType? ).
20:33:00raptorthat isn't XML specific is it?
20:34:00sam686in XML, could be easier make a mistake (missing </x>)
20:34:00sam686it world be difficult to find the error when the level won't load.
20:34:00raptorusually most text editor tell you if it is missing
20:35:00raptorXML is a very structurally rigid specification
20:35:00raptorso it is easy to tell errors
20:35:00raptor(if your editor has a built in XML parser)
20:37:00sam686windows don't have a built-in default XML editor, only notepad. Though many internet browser might find errors
20:37:00raptoryeah - any browser will tell you right where the error is - besides, anyone serious about text editing should not use notepad :]
20:39:00sam686XML may take longer to start hosting, as it have to look at all possible levels for title, minplayers, maxplayers.
20:40:00raptoryes, that's what we figure
20:41:00raptorbut it will allow for greater flexibility when writing the files, and order won't matter anymore
20:43:00sam686some XML errors won't be found in any XML error checker, such as putting <polywall>...</polywall> inside <barrier>, or missing <x>4</x>
20:43:00raptornot if we use a DTD
20:44:00raptorwhich states the exact structure a specific XML file must meet
20:45:00raptorfor isntance
20:46:00raptorhere is the DTD for that XML file i created: http://pastie.org/1923615
20:47:00raptorall you have to do is load the DTD with a parser, and if the XML doesn't match the rules, it is invalid
20:47:00raptorbut that may be overkill for what we need
20:48:00sam686what happens when you put <minplayers>zz</minplayers>, that might not be caught my error checking?
20:49:00raptoryes that is true
20:49:00raptorXML is a structural format
20:49:00Helmic has left
20:49:00raptorthat would have to be caught by the code
20:50:00sam686might make it harder to convert XML to older .level format (if ever want to run level in older version)
20:50:00raptorvery true
20:50:00raptori'm thinking we would never do that though
20:51:00raptorbeca
20:51:00raptorbecause we only move forward!
20:51:00raptor:)
20:53:00sam686Many errors (not enough points, not a number when number required) will probably have to be checked in level loading and somehow display / log warnings?
20:53:00raptoryes - which is more than what we do right now
20:55:00sam686no matter what, using something in newer version that don't exist in older version won't work properly (polywalls not supported in 014 / 013), same goes with XML or not (.level).
20:56:00raptoryep
21:51:00raptor Quit (Remote host closed the connection)
22:47:00watusimotohow are the objects loaded from the file now? ==> still using processlevelLoadLine()
22:52:00watusimotoeach object will now have a compainion function, currently called toString() that handles the output end of things
22:53:00watusimotosort of like pack/unpack update
22:57:00[1]watusimoto has joined
22:59:00watusimoto Quit (Ping timeout: 258 seconds)
22:59:00[1]watusimoto is now known as watusimoto
23:31:00raptor has joined
23:31:00ChanServ sets mode +o raptor
23:34:00ThePing has joined
23:34:00ThePing has left
23:48:00sam686 Quit (Read error: Connection reset by peer)
23:49:00sam686 has joined
23:49:00ChanServ sets mode +v sam686

Index Search ←Prev date Next date→

These logs were automatically created by BFLogBot on irc.freenode.net.