#exult@irc.freenode.net logs for 12 Apr 2002 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[00:24:53] <Darke> Hmm... try including `#include <iostream>` up the top, it looks not to be finding std::ios for some reason.
[00:26:14] <Kirben> no difference
[00:29:02] <Darke> Do a quick search replace on std::ios::pos_type and ios::pos_type and replace them with 'unsigned long'? There's only about 9 references in the both the UCCache files.
[00:30:08] * Darke could be barking up the wrong tree, but the only thing in common in those two lines is pos_type and `const unsigned int` and he doesn't think the latter is a problem.
[00:40:07] <Kirben> g++ -O2 -Wno-long-long -DHAVE_MIDI -DHAVE_SNPRINTF -Dsnprintf=_snprintf -DPENTAG
[00:40:07] <Kirben> RAM -DDEBUG -DSIZEOF_SHORT=2 -DSIZEOF_INT=4 -I./tools -I./viewer -I./files -I./a
[00:40:07] <Kirben> udio -I./headers -I./conf -I./sdl/include -c -o UCCache.o ./tools/UCCache.cc
[00:40:07] <Kirben> ./tools/UCCache.cc:275: macro `READ_RET' used without args
[00:40:07] <Kirben> ./tools/UCCache.cc:276: macro `WRITE_RET' used without args
[00:40:07] <Kirben> ./tools/UCCache.cc:287: macro `READ_PUSHR8' used without args
[00:40:09] <Kirben> ./tools/UCCache.cc:288: macro `WRITE_PUSHR8' used without args
[00:40:11] <Kirben> ./tools/UCCache.cc:291: macro `READ_PUSHR16' used without args
[00:40:13] <Kirben> ./tools/UCCache.cc:292: macro `WRITE_PUSHR1
[00:45:02] <Kirben> ./tools/UCCache.cc:292: macro `WRITE_PUSHR16' used without args
[00:45:02] <Kirben> ./tools/UCCache.cc:295: macro `READ_PUSHR32' used without args
[00:45:02] <Kirben> ./tools/UCCache.cc:296: macro `WRITE_PUSHR32' used without args
[00:45:02] <Kirben> ./tools/UCCache.cc:303: macro `READ_GENERIC0' used without args
[00:45:02] <Kirben> ./tools/UCCache.cc:307: macro `READ_GENERIC0' used without args
[00:45:02] <Kirben> ./tools/UCCache.cc:311: macro `READ_END' used without args
[00:45:04] <Kirben> ./tools/UCCache.cc:312: macro `WRITE_END' used without args
[00:45:06] <Kirben> ./tools/UCCache.cc:387: macro `WRITE_RET' used without args
[00:45:08] <Kirben> ./tools/UCCache.cc:396: macro `WRITE_PUSHR8' used without args
[00:45:10] <Kirben> ./tools/UCCache.cc:399: macro
[00:45:15] <Kirben> etc...
[00:45:34] <Darke> The quickest fix is to drop a '0' in the brackets to each call ('READ_RET(0)' and so on). And I'll clean up all the macro calls in UCOpcodes.h when I commit.
[00:46:11] * Darke sighs and wonders why gcc3 didn't complain to him about them. Unless they're not _technically_ bugs.
[00:53:54] <Kirben> ok all compiled for now
[00:58:55] <Darke> Thanks. I'd like to know why ios::pos_type wasn't working. Using unsigned long is noted specifically as being the Wrong Thing in my C++ std.lib book. Just curious, does std::streampos work instead?
[02:02:01] <Kirben> Yes that also fixes the first error
[02:06:45] <Darke> Thanks. I'll use that instead. I should poke around autoconf and see if I can get it to detect whether ios::pos_type is defined or std::streampos is.
[02:59:33] * Darke is off to test his KDE upgrade. He _might_ be back sometime soon. <grin>
[02:59:35] <-- Darke has left #exult ()
[03:53:07] <-- Kirben has left IRC (Read error: 104 (Connection reset by peer))
[04:46:45] --> Darke has joined #exult
[04:46:45] --- ChanServ gives channel operator status to Darke
[05:42:28] --> sb-x has joined #exult
[05:42:34] <sb-x> ?time
[05:42:34] <exultbot> It is now Fri Apr 12 05:42:34 2002 (GMT).
[05:42:53] <sb-x> Darke how is it?
[05:43:55] <sb-x> it = KDEXP 3000 ME SE (OSR 2)
[05:44:13] <Darke> Hello. 'It' is ok, I think.
[05:44:23] <sb-x> hi
[05:45:37] <sb-x> did you get the dev version of "KDEXP 3000 ME SE (OSR 2)" or the stable version?
[05:47:34] <Darke> There's such a thing as a 'stable version'?
[05:47:47] * sb-x laughs.
[05:47:56] <sb-x> How do you compare it to KDE2?
[05:49:10] <Darke> I'd love to, except it 'crashes' on startup. <grin> It appears when they say "don't compile it using gcc3.0, use the beta gcc3.1 if you must compile with gcc3", they mean it.
[05:50:57] <sb-x> :\
[05:51:09] <sb-x> So you kept your KDE 2 backup handy.
[05:51:29] * Darke is subscribed to the gcc-bugs ML. It appears there are still quite a few problems, although mostly estoric.
[05:52:08] <sb-x> Hmm
[05:53:02] <Darke> Yeah. KDE3 installs by default into the /opt/kde path, kde2 is installed by default into /opt/kde2. So all I need to do is reverse the declaration of /opt/kde2/bin or /opt/kde/bin in /etc/profile and it swaps which one boots.
[05:53:18] <Darke> s/reverse the/reverse the order of the/
[05:54:34] <sb-x> So people complained that things broke when KDE 2 installed into it's own directory, so they changed it back to /opt/kde for KDE 3?
[05:55:43] <Darke> It makes sense to me to drop it into another directory, I didn't care whether it was /opt/kde3 or /opt/kde. <grin>
[05:56:28] * sb-x agrees.
[05:56:36] <sb-x> thanks, maybe you'll get it to work later
[05:56:46] <sb-x> i have to finish some things and sleep
[05:56:52] <-- sb-x has left IRC ("X-Chat [1.6.4]")
[08:20:58] <-- Darke has left IRC (Remote closed the connection)
[08:22:41] --> Darke has joined #exult
[08:22:41] --- ChanServ gives channel operator status to Darke
[09:50:54] --> wjp has joined #exult
[09:50:54] --- ChanServ gives channel operator status to wjp
[09:50:57] <wjp> hi
[09:51:10] <Darke> Hi.
[09:51:34] <wjp> Darke: I forgot to ask wednesday: how did the interview go?
[09:52:45] <Darke> wjp: Thanks. The interview went well, but got the news today that I didn't get the job. <shrug> Such is life. <grin> At least my records are on file with them so that I can get called if they do need some emergency help.
[09:53:06] <wjp> :/
[09:54:00] <Darke> It was only a temp job, but it certainly would have helped get my foot in the door.
[09:54:56] <wjp> how old are you, btw?
[09:55:02] <wjp> (seems to have slipped my mind :-( )
[09:55:39] <Darke> Umm... 22 +/- 1, I keep forgetting. <grin>
[09:55:47] <wjp> hehe :-)
[09:55:53] * wjp is 22-1 ;-)
[09:56:04] <Darke> I _do_ know I was born in 1979. But that'a all. <grin>
[09:56:26] <wjp> that makes the -1 case pretty unlikely :-)
[09:56:36] * Darke grins. Yep.
[09:56:40] <wjp> so it's 22.5 +- 0.5 :-)
[09:57:01] <Darke> That sounds a bit better... <grin>
[09:58:46] * Darke is currently cursing gcc3.1cvs. It's got a funky configuration problem where it tries to ln -s, in the wrong directory and failing quietly. And the configuration files are a nightmare. <grin>
[09:59:01] <wjp> ugh
[09:59:02] <wjp> sounds like fun
[10:01:45] <Darke> Yep. Then there was the Makefile.in problem where it wasn't traversing down the directory tree far enough. It was getting passed a variable that looked like './../../version.c', and it needed one that looked like './../../../version.cc', so the line in the Makefile.in now looks like `thing=./.$(version_file)`. Which is about as unintuitive as you can get. <grin>
[10:02:27] <wjp> hehe :-)
[10:05:40] * Darke wonders at the 'italian translation' email on the ML... is it _really_ not that obvious that all the .flx files might not be strings?
[10:08:03] <-- royalsexy has left IRC (Read error: 110 (Connection timed out))
[10:09:14] --> royalsexy has joined #exult
[10:14:44] * Darke sighs. It's too complicated to fix manually, and there's no indication of which file is breaking silently. Oh well, I guess I'll have to wait a day or so to see if it's a 'known' problem.
[10:23:31] * wjp has to go again
[10:23:33] <wjp> bye
[10:23:39] <-- wjp has left IRC ("[x]chat")
[10:45:35] <-- Darke has left IRC (sterling.openprojects.net irc.openprojects.net)
[10:58:16] --> Darke has joined #exult
[10:58:37] --- ChanServ gives channel operator status to Darke
[10:58:52] --> Kirben has joined #exult
[10:58:52] --- ChanServ gives channel operator status to Kirben
[11:03:05] <-- royalsexy has left IRC (Read error: 110 (Connection timed out))
[11:06:27] --> royalsexy has joined #exult
[13:50:11] <-- royalsexy has left IRC (Remote closed the connection)
[13:50:45] --> royalsexy has joined #exult
[13:53:14] <-- royalsexy has left IRC (Remote closed the connection)
[13:53:44] --> royalsexy has joined #exult
[13:55:15] --- royalsexy is now known as flurotube
[14:11:25] --> Colourless has joined #Exult
[14:11:25] --- ChanServ gives channel operator status to Colourless
[14:11:48] <Colourless> hi
[14:20:01] <Darke> Hi.
[14:20:13] <Colourless> 'bout time :-)
[14:23:36] * Darke had his head in the autoconf docs, trying to locate something to test if a datatype rather then a function, exists or not. <grin>
[14:24:47] <-- flurotube has left IRC (Remote closed the connection)
[14:26:40] --> flurotube has joined #exult
[14:27:17] <-- flurotube has left IRC (Remote closed the connection)
[14:33:57] --> flurotube has joined #exult
[14:34:16] <-- flurotube has left #exult ()
[14:49:39] <Colourless> hey bunny rabbit, can conf be used to read a file from a flex?
[14:52:58] <Darke> Currently, no. Since, IIRC, the only access you can get to it's internal 'load a file' is through it's 'open a file' method. Which doesn't support flexes.
[14:53:13] <Colourless> That should be changed :-)
[14:53:42] * Colourless has grand visions of using conf to read small text based data files from flexes
[14:53:54] * Colourless of course is not say why he would want to do that
[14:54:28] <Darke> Actually, no. If you load the data from the flex into a std::string. You can pass them to Configuration::read_config_string().
[14:54:47] <Darke> But that's currently the only method. <grin>
[14:55:12] * Darke can guess why Colourless might want to do that...
[14:55:14] * Colourless has a look
[14:57:50] <Colourless> actually it appears that is what read_config_string is meant to do. Of course it duplicates the entire string, which is just a tad inefficent
[14:58:07] <Darke> You need to make sure you call Configuration::clear() before you call that function though if you're reusing a Configuration type. Since it relys on read_config_file to clear it.
[14:59:06] <Colourless> I wouldn't be reading from a file first. Of course reading from a flex isn't anything i plan on doing yet, i just want to know if it can be done... which it obviously can
[14:59:43] * Darke nods. One modification he's been considering to XMLEntity is that it can also load from a str::istream, but he's certainly going to wait until after 1.0 so he can't accidentally add any bugs. <grin>
[15:24:35] --> MeatShield has joined #exult
[15:25:44] --> wjp has joined #exult
[15:25:45] --- ChanServ gives channel operator status to wjp
[15:25:48] <Colourless> hi
[15:26:03] <wjp> hi
[15:26:08] <MeatShield> hey there.
[15:27:58] <-- MeatShield has left IRC (Client Quit)
[15:30:45] <Darke> Hi.
[15:41:06] * wjp is going to mess with the insides of his PC a bit...
[15:41:11] <wjp> bbl (hopefully ;-) )
[15:41:16] <-- wjp has left IRC ("[x]chat")
[16:04:09] <-- Kirben has left IRC ("System Meltdown")
[16:46:16] <Colourless> so Darke, did you read my update to opengl.txt and find what I wanted to use conf with a flex for? :-)
[19:07:40] --> Fingolfin has joined #exult
[19:07:52] <Fingolfin> yo
[19:07:58] <Colourless> hi
[19:09:20] <Colourless> have you by any chance been able to test out dcmpsound ?
[19:09:51] <Fingolfin> I have CVS/IRC/SSH for the first time this hour in the last 5 days ... no :-)
[19:10:08] * Fingolfin goes to update a big bunch of CVS repositories =)
[19:10:22] <Colourless> oh ok. hadn't seen you for... 5 days and wasn't sure :-)
[19:11:42] <Fingolfin> <g>
[19:24:00] <-- Colourless has left IRC (sterling.openprojects.net irc.openprojects.net)
[19:24:00] <-- Darke has left IRC (sterling.openprojects.net irc.openprojects.net)
[19:26:44] --> Darke has joined #exult
[19:27:48] --> Colourless has joined #exult
[19:30:59] <Fingolfin> hm, tools/ucfiddle.cc doesn't compile
[19:33:30] <Colourless> :-)
[19:33:41] <Colourless> i hear it's linux only
[19:33:47] <Fingolfin> now it does
[19:34:07] <Fingolfin> no std::ios::pos_type in my GCC version (similiar problem is being worked around in files/util.cc IIRC
[19:34:42] <Fingolfin> ARGH
[19:34:50] <Fingolfin> #define READ_RET(X) READ_GENERIC0(X) // 50
[19:34:53] <Fingolfin> but it's used as:
[19:34:57] <Fingolfin> READ_RET();
[19:35:07] <Fingolfin> -> why does that compile at all, anywhere, I wonder?
[19:36:32] <Colourless> is the ucfiddle compiling?
[19:36:44] <Fingolfin> now, yes
[19:36:57] <Fingolfin> but UCCache.cc isn't
[19:36:57] <Fingolfin> so I am working on that now
[19:37:08] <Colourless> :-)
[19:41:59] --> wjp has joined #exult
[19:41:59] --- ChanServ gives channel operator status to wjp
[19:42:03] <wjp> hi again
[19:42:12] <Fingolfin> hi willem
[19:42:12] <Colourless> wb
[19:42:24] <Colourless> you were gone... quite a while :-)
[19:42:31] <wjp> I know... :/
[19:42:38] <Colourless> what sugery did you perform to your machine
[19:42:46] <wjp> installing cd-writer
[19:42:48] <Colourless> surgery even
[19:43:00] <wjp> normally this wouldn't really be a problem...
[19:43:48] <wjp> but there were these tiny problems like already having used up all 4 IDE slots, the extra IDE card I have not (yet) being supported by Linux... fun stuff like that
[19:44:07] <wjp> oh, and don't forget about cables being too short... *grumble*
[19:44:25] <Colourless> fun fun fun :-)
[19:44:41] <wjp> then there was the part about needing ide-scsi support built into your kernel to be able to use cdrecord
[19:45:51] <Colourless> well then you should use an operating system that does support your ide card, and doesn't need a new kernel to use a cd-r :-)
[19:46:14] <wjp> well, it wouldn't have needed a new kernel if I hadn't stripped that part from it in the first place :-)
[19:46:20] <Colourless> hehe
[19:46:44] <Colourless> well, you have that option, and if i had it too, i probably would have done the same :-)
[19:46:54] <wjp> and I think that card is already supported in the latest version
[19:46:59] <wjp> :-)
[19:47:50] <wjp> now I'm stuck with one _tiny_ little glitch...
[19:48:06] <wjp> I can write CDs with it, but not read them
[19:48:21] <wjp> well, I _can_ read them, but then I have to reboot, and I won't be able to write again ;-)
[19:48:31] * wjp kicks ide-scsi
[19:48:41] <Colourless> hmm. that sounds like it could be a problem
[19:49:27] <wjp> it should technically be possible to read from it, but I just haven't figured out how :-)
[19:49:42] <Colourless> you think?? :-)
[19:56:38] <Fingolfin> well, with a big lense, and a lot of patience, I am sure wjp can read from his CDs
[19:56:54] <wjp> lol
[19:57:12] <Colourless> :-)
[20:24:04] <Colourless> hmm, strange... i ran a u8 screenshot through the dos 2xSaI scaler program and it doesn't make that much of a difference
[20:24:45] <wjp> how do you mean that?
[20:25:06] <Colourless> it's not smoothing out the lines very well
[20:27:18] <Colourless> Super2xSaI or SuperEagle would probably do a better job because they detect more patterns according to kreed.
[20:34:02] <Colourless> look at http://www.users.on.net/triforce/u82xsai.png to see what i mean. it doesn't seem to make as big a difference to u8 as it does to u7
[20:34:23] <wjp> hm, I see
[20:34:59] <Colourless> things like the eggs scaled quite nicely though :-)
[20:36:09] <Colourless> what has happened is it doesn't look like it detected the 2x1 stair stepping at all
[20:37:13] <wjp> do you have stand-alone versions for some of the other scalers?
[20:37:20] <Colourless> no i don't
[20:37:23] <Colourless> i've only got 2xsai
[20:37:57] --- wjp is now known as wjp|away
[20:37:58] <wjp|away> bbl
[21:40:52] --- wjp|away is now known as wjp
[22:01:06] <-- Fingolfin has left IRC ("42")
[22:39:34] --> Evil_Matt_O has joined #exult
[22:40:05] <Evil_Matt_O> ?seen fingolfin
[22:40:05] <exultbot> fingolfin left IRC around Fri Apr 12 22:01:06 2002 (GMT) ("42")
[22:40:21] <Colourless> [07:31] *** Fingolfin (maxhorn@p50890A22.dip.t-dialin.net) Quit ("42")
[22:40:23] <Colourless> [08:09] *** Evil_Matt_O (jb@1667079149-yippy.dsl.xmission.com) has joined #exult
[22:40:33] <Evil_Matt_O> drat!
[22:59:45] <-- Evil_Matt_O has left IRC ()
[23:03:24] <wjp> brb
[23:03:28] <-- wjp has left IRC ("[x]chat")
[23:04:00] --> wjp has joined #exult
[23:04:00] --- ChanServ gives channel operator status to wjp
[23:06:18] <Colourless> wb
[23:06:51] <wjp> thx :-)
[23:07:18] <Colourless> it's about time i left :-)
[23:07:26] <wjp> ok, goodnight
[23:07:33] <Colourless> bye
[23:07:35] <-- Colourless has left IRC ("no comment")