#pentagram@irc.freenode.net logs for 26 Jan 2004 (GMT)

Archive Today Yesterday Tomorrow
Pentagram homepage

[00:57:59] --> Kirben has joined #pentagram
[00:57:59] --- ChanServ gives channel operator status to Kirben
[01:48:19] <-- shinji-kun has left IRC (""Baka!" - Asuka-chan")
[04:50:09] --> sbx has joined #pentagram
[05:11:58] --> EsBee-Eks has joined #pentagram
[05:11:59] <-- sbx has left IRC (Read error: 104 (Connection reset by peer))
[06:52:37] <-- EsBee-Eks has left IRC ()
[08:09:28] * wjp wakes up
[08:09:30] <wjp> morning
[09:19:35] --> watt has joined #pentagram
[09:19:42] <watt> hey all.
[09:44:59] <wjp> hi
[09:46:46] <watt> I see you an Colourless have been busy
[09:46:53] <wjp> :-)
[09:47:01] <wjp> just a bit :-)
[09:49:04] <watt> is "--enable-ttf" all that is required to use new ttf 's?
[09:51:38] <watt> oh.. duh.. I need to actually get SDL_ttf
[10:15:11] <watt> ack segfault!
[10:35:53] <watt> it might help if the font is in data... trying again
[10:59:52] <wjp> yes, since there's no proper way to load the font yet, there's also no proper error checking on loading it :-)
[11:01:05] <wjp> oh, I see I made CVS use VeraBd.ttf
[11:01:19] <wjp> didn't realize I committed that
[11:02:29] <wjp> (for the logs, VeraBd is from the Bitstream Vera fonts, btw)
[11:05:00] <watt> Very cool stuff.
[11:29:12] --> Colourless has joined #Pentagram
[11:29:14] --- ChanServ gives channel operator status to Colourless
[11:29:36] <Colourless> hi
[11:29:47] <watt> hiya
[11:40:11] <watt> so.. I was playing around and made a "RotationalMoverProcess" which allowed keyboard movement (using a turn left or right and then walk method as opposed to direct movement) and noticed that only one Process could control Avatar's movement at any given time
[11:40:37] <watt> due to the waitFor method I believe...
[11:41:08] <Colourless> there should only really be 1 process that control avatar's movement, and that will take in all inputs and work out what to do with it
[11:41:33] <watt> offered some interesting results.. I could only move with keyboard until I collide with something, then I could only move with mouse until I collide again..
[11:42:53] <Colourless> mouse movement should override keyboard input
[11:43:28] <watt> yeah.. so I'm not thinking along the lines of subclasses that the AvatarMoverProcess asks what to do and if it doesn't get anything then it moves to the next one.. when it recieves nothing from all of them.. then it stands still
[11:43:38] <watt> right.. I figured that much
[11:44:56] <Colourless> since the same command can be bound to multiple devices you should probably use counters for command states
[11:45:21] <Colourless> so if you held both, you'd need to release both for the command to actually stop
[11:46:22] <Colourless> of course that can cause problems sometimes if not all of the keyup's are sent
[11:49:00] <watt> yeah.. I'm not to keen on the idea of Avatar running until he drowns.
[11:49:34] <Colourless> so, it's up to you
[11:49:35] <watt> but still.. counters shouldn't be hard to implement
[11:49:53] <Colourless> people should know that holding down 2x of the same command isn't a good idea
[11:53:02] <wjp> hi Ryan
[12:59:08] <-- ragzter has left IRC ("Reboot; jag återkommer om en stund!")
[13:12:03] <wjp> hm, CVS is pretty fast again luckily :-)
[13:15:09] <wjp> I guess if SDL_ttf is used in Makefile.mingw I should add some basic error checking
[13:18:01] <Kirben> Any other extra files needed ? just crashes.
[13:18:42] <wjp> VeraBd.ttf in the data directory
[13:19:19] <wjp> latest cvs (as in 2 minutes ago) shouldn't crash anymore if it's missing, btw
[13:23:27] <Kirben> Where do I find 'VeraBd.ttf' ?
[13:29:33] <Kirben> ah a gnome fonts package, so users will have to get those separately.
[13:29:50] <Kirben> hmm current cvs seems to be broken:
[13:29:51] <Kirben> gumps/widgets/TextWidget.cpp: In member function `bool
[13:29:51] <Kirben> TextWidget::setupNextText()':
[13:29:52] <Kirben> gumps/widgets/TextWidget.cpp:98: `tx' undeclared (first use this function)
[13:29:52] <Kirben> gumps/widgets/TextWidget.cpp:98: (Each undeclared identifier is reported only
[13:29:53] <Kirben> once for each function it appears in.)
[13:29:55] <Kirben> gumps/widgets/TextWidget.cpp:98: `ty' undeclared (first use this function)
[13:47:41] <wjp> Kirben: VeraBd.ttf is part of the Bitstream Vera fonts
[13:47:53] <wjp> it's freely distributable for us
[13:48:10] <wjp> (although I think we need to include a readme file for it or something)
[13:49:03] <wjp> gah, that's a stupid bug
[13:49:40] <wjp> fixed now
[13:52:59] <wjp> Kirben: if you want to distribute the font, put the files from http://www.math.leidenuniv.nl/~wpalenst/vera.zip in the data directory
[13:53:31] <Kirben> ok will do.
[13:56:43] <wjp> I'll probably get around to making fonts configurable soon
[14:56:52] <-- Kirben has left IRC ("System Meltdown")
[16:18:35] <-- watt has left #pentagram ()
[16:44:48] * wjp sighs and _really_ fixes the crash when VeraBd.ttf isn't found now :-)
[16:44:57] <wjp> only took me three tries too
[16:45:20] <Colourless> :-)
[16:46:29] * wjp remembers something and switches to #exult
[17:30:22] --> watt has joined #pentagram
[17:35:57] <wjp> hi again
[17:37:55] <watt> hiya.. yeah.. played some windows games.. got bored.. I'm back
[18:59:01] <wjp> ok, U8 indeed seems to be using DOSLatinUS (CP437) or DOSLatin1 (CP850)
[18:59:20] <wjp> (both are the same on the characters U8 has in its fonts)
[19:00:01] <wjp> interesting how they treat the german-double-s in those
[19:00:48] <wjp> in CP437, 0xE1 is in a row of greek characters (so it would be a beta)
[19:01:04] <wjp> in CP850, most of the greek characters are gone, but 0xE1 is still there
[19:02:07] <wjp> yay, found a map from CP850 to unicode :-)
[19:06:01] <wjp> I'll adjust it for the use of 0x40/'@' in U8 and commit it
[19:06:09] <wjp> won't hurt to have that around
[19:48:55] * wjp hacks in that table to see if it indeed "works"
[19:49:19] <wjp> those @'s being replaced by bullets is kind of nice :-)
[19:51:57] <wjp> and french and german are working now too
[20:32:57] * wjp sighs... left a hardcoded width=180 somewhere in TextWidget.cpp
[20:33:07] <wjp> and here I was wondering why some long conversation options weren't showing up
[20:34:49] <Colourless> :-)
[21:08:54] --> shinji-kun has joined #pentagram
[21:25:26] <wjp> hm, interesting... comment in dosbox cvs commit: "[....] cos pharlap (crusader) wants it..."
[21:25:51] <wjp> apparently someone is looking at getting crusader to work
[21:58:11] <-- Colourless has left IRC ("casts invisibility")