#exult@irc.freenode.net logs for 1 Feb 2013 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[00:00:06] <Dominus> k
[00:00:27] <Marzo> Then in shapes/shapewrite.cc line 427, replace this:
[00:00:31] <Marzo> WriteStr(out, usecode_name, true);
[00:00:37] <Marzo> With this:
[00:00:38] <Marzo> out << usecode_name << std::endl;
[00:01:09] <Eviltar> :) only trying to ease things, I take no offense at my way's not being suitable for you, ethical hedonism yo.
[00:01:37] <Dominus> :)
[00:05:43] <Dominus> marzo, still the same error
[00:06:00] <Marzo> O>o
[00:06:06] <Marzo> Send me the error
[00:06:56] <Dominus> http://pastebin.com/NBbHzWSS
[00:08:03] <Dominus> ah, we are now in utils.cc. didn't notice that it's now not uitls.h anymore (and before as well)
[00:08:36] <Marzo> I just noticed that too
[00:08:51] <Marzo> And in any case, something seems to be broken in this dev setup
[00:09:10] <Marzo> I just double-checked, and even the previous error should not be happening
[00:10:26] <Marzo> The function that the compiler is not finding is declared on <string> header
[00:10:50] <Marzo> Which is not only included in utils.cc, it is also included in utils.h!
[00:12:47] <Eviltar> is this xcode ? this might be relevant? http://stackoverflow.com/questions/11748504/trying-to-build-muparser-error-explicit-instantiation-of-stdbasic-ostream-b
[00:12:52] <Dominus> it's really strange it happens with llvm and not clang
[00:13:47] <Marzo> I just found that page too; but no, it is not xcode
[00:16:05] <Marzo> Dominus: out of curiosity, what version of llvm-gcc is this?
[00:17:08] <Dominus> i686-apple-darwin11-llvm-gcc-4.2 (GCC) 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2336.11.00)
[00:18:37] <Marzo> Can you do a clean build and post the compilation log? Just so I can check one thing
[00:18:56] <Dominus> marzo, I found the culprit and eviltar *did* send me to the right place
[00:18:59] <Marzo> (although I doubt it is this)
[00:19:08] <Marzo> Really?
[00:19:21] <Eviltar> the minmacosversion thing?
[00:19:42] <Dominus> I was still building for target OS X10.5 by defining export MACOSX_DEPLOYMENT_TARGET=10.5 and -mmacosx-version-min=10.5
[00:19:45] <Dominus> yes
[00:19:53] <Dominus> when I defined it for 10.8 it worked
[00:19:54] <Eviltar> yay
[00:20:00] <Marzo> Well, then you can undo everything I told you to do
[00:20:37] <Dominus> and still works
[00:20:53] <Dominus> the joys of using Apple :)
[00:21:04] <Dominus> not so far from mingw after all...
[00:22:44] <Eviltar> i found an old orange ibook with a dead hdd while i've been packing, I'm not even really sure where it came from, but I'm thinking of seeing how horrible it would be with debian mintppc
[00:23:17] <Dominus> kind of the reason I'm building the 32bit snapshots with clang against the old SDK 10.5 - building against higher SDKs but targeting 10.5 will not let Exult run on 10.5 :(
[00:23:25] <Dominus> Eviltar: don't bother :)
[00:24:06] <Dominus> thanks Marzo and sorry for driving you crazy
[00:24:29] <Marzo> Np
[00:24:44] <Marzo> Next time I will know what to ask :-p
[00:24:55] <Dominus> he he
[00:25:04] <Eviltar> 300 MHz? and 64mb of ram? sounds like a challenge, maybe gentoo
[00:25:05] <Dominus> I'll make a note in my build script
[00:25:40] <Dominus> Marzo, did you see the other warnings clang spat out in ucdisasm and newfilegump? http://pastebin.com/hWPgqzD8
[00:25:52] <Marzo> Dealing with them now
[00:29:46] <Dominus> I think this should remind me before doing bad SDK stuff again next time:
[00:29:48] <Dominus> # NEVER EVER MIX CURRENT LLVM-GCC + SDK 10.8 WITH MACOSX_DEPLOYMENT_TARGET=10.5 (and -mmacosx-version-min=10.5)
[00:30:35] <Dominus> Eviltar: you will only end up disappointed that you spent so much time on it :)
[00:32:05] <Eviltar> http://www.mintppc.org/content/user-experience-ibook-g3300-clamshell
[00:32:50] <Eviltar> i highly doubt this one has that much ram installed, but tbh I havent checked yet
[00:35:08] <Dominus> so, when my snapshot compile finally went through I'm going to bed...
[00:37:09] <Dominus> (another nice one I just stumbled over was an alias I had in my root called Developer to /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer (where the newer sDKs are stored) - somehow this made my ppc compile stumble, even though it uses /developer-old and not /developer
[00:37:40] <Dominus> the alias was new because I wanted to save me some writing
[00:38:31] <Dominus> but since the ppc toolchain used to be in /developer instead of developer-old, I can only presume that's where it faltered...
[00:38:35] <Dominus> oddities
[00:38:39] <Eviltar> maybe the hyphon is screwing it up, try to make the new one devnew or something
[00:39:12] <Dominus> very strange stuff :)
[00:39:31] <Eviltar> computers do screwy things sometimes
[00:47:20] <Dominus> good night
[00:47:30] <Eviltar> gn
[00:52:55] <Malignant_Manor> Marzo: You can roll up the carpet while descending.
[01:02:04] <-- Malignant_Manor has left IRC (Quit: ChatZilla 0.9.89 [Firefox 18.0.1/20130116073211])
[01:02:04] <Marzo> Malignant_Manor: thanks, fixed
[03:47:12] --> Malignant_Manor has joined #exult
[03:48:30] <Malignant_Manor> Marzo, why is .c_str() used on lines 79 and 83 in audio\midi_drivers\FluidSynthMidiDriver.cpp
[04:15:50] <-- Malignant_Manor has left IRC (Quit: ChatZilla 0.9.89 [Firefox 18.0.1/20130116073211])
[04:16:31] --> Kirben_ has joined #exult
[04:16:31] --- ChanServ gives channel operator status to Kirben_
[04:17:28] --> Baastuul_ has joined #exult
[04:19:20] <-- Kirben has left IRC (Ping timeout: 276 seconds)
[04:19:33] <-- Rottingbeef has left IRC (Ping timeout: 276 seconds)
[07:09:11] <-- RadoS has left IRC (Remote host closed the connection)
[07:45:42] <-- Kirben_ has left IRC (Ping timeout: 264 seconds)
[07:45:56] --> Kirben has joined #exult
[07:45:57] --- ChanServ gives channel operator status to Kirben
[08:21:37] <-- sh4rm4 has left IRC (Remote host closed the connection)
[08:25:34] --> sh4rm4 has joined #exult
[10:10:12] <-- nutron has left IRC (*.net *.split)
[10:10:12] <-- Eviltar has left IRC (*.net *.split)
[10:13:13] --> nutron has joined #exult
[10:13:13] --> Eviltar has joined #exult
[12:22:52] <Dominus> still more than 50 bugs in the tracker :)
[12:23:51] <Dominus> but impressive work. From december 2011 to december 2012 ~70 commits. From december 2012 to february 2013 almost 200 :)
[12:31:37] <Dominus> I think apart from the top 5 bug reports (and only the top three really important) and maybe the blacksmith schedule, we are done for a release
[12:37:59] --> RadoS has joined #exult
[12:41:21] <-- Kirben has left IRC (Ping timeout: 276 seconds)
[13:04:54] --> TheCycoONE has joined #exult
[13:17:06] <-- sh4rm4 has left IRC (Remote host closed the connection)
[15:03:12] <-- TheCycoONE has left IRC (Read error: Connection reset by peer)
[15:10:38] --> TheCycoONE has joined #exult
[17:14:55] --> sh4rm4 has joined #exult
[18:43:43] <Marzo> Dominus: On examining wuc/wud/ucxt, I am seeing that the ASM output of ucxt is quite different from that of wud
[18:44:17] <Marzo> On fixing it (details don't much matter), I came across again at the unknown audition opcode
[18:44:45] <Marzo> Now, the implementation in Exult gives the same results as the original
[18:45:02] <Marzo> But looking at the usecode, I can kind of guess what it was supposed to do
[18:45:09] <Marzo> (but never worked)
[18:46:44] <Marzo> Should I try implementing it?
[18:54:32] --> Malignant_Manor has joined #exult
[18:56:25] <Dominus> Marzo: I have no opinion concerning *aynthing* usecode related. I never used any of those tools and only wrote just a tiny bit for the SF island patch
[18:56:34] <Malignant_Manor> What is it supposed to do.
[18:56:48] <Marzo> I am thinking more as in "fixing" how Exult runs it
[18:57:00] <Marzo> But now that I am looking more closely, it seems just broken usecode
[18:57:26] <Marzo> It seems like it should be some sort of default action -- if no answer has been found, run it
[18:57:45] <Marzo> Otherwise, skip to the location it points
[18:58:13] <Marzo> And now that I think of it, it seems that Exult would do this with the correct usecode
[18:58:33] <Malignant_Manor> I don't really understand. Do you have an example?
[19:00:28] <Marzo> Think of it being like a default in a switch
[19:01:25] <Marzo> It is not needed in UCC, as it does a better job at implementing conversation loops
[19:04:43] <Malignant_Manor> I remember the output of conversations by utxc as horrible.
[19:05:23] <Malignant_Manor> I haven't really looked at the original code for it.
[19:05:30] <Marzo> It will marginally improve with what I am doing right now, but a good conversation format will require substantial changes
[19:05:50] <Malignant_Manor> Are there any problems with the original game that this would fix?
[19:06:52] <Malignant_Manor> If it is a short fix, it doesn't really matter either way as long as it doesn't introduce bugs.
[19:06:55] <Marzo> Now that I thought more, no -- fixing it would require a usecode fix, not an opcode fix
[19:26:21] <Malignant_Manor> Marzo is removing the class qualifier okay for these? http://pastebin.com/edKM4Pp5
[19:27:06] <Marzo> Probably
[19:27:10] <Malignant_Manor> It's supposed to be unnecessary and some compilers will see it as an error according to cppcheck.
[19:27:28] <Marzo> Yeah, I remember seeing it
[19:27:55] <Marzo> If things work without them, I think it is safe
[19:29:10] <Malignant_Manor> Well, I'm not sure how to test all these besides the fact that it compiles.
[19:30:00] <Marzo> Hrm. I just realized I did a bad decision when implementing a call indirect opcode
[19:30:50] <Marzo> Oh, well -- I will add a new one and make UCC never use the former again
[19:35:31] <Malignant_Manor> Cyclopes at 1221, 237 will teleport through walls when they wake up.
[19:36:43] <Marzo> Put it on the tracker, will look at it when I finish what I am doing
[19:50:43] <-- Dominus has left IRC (Read error: Connection reset by peer)
[19:51:14] --> Dominus has joined #exult
[19:51:15] --- ChanServ gives channel operator status to Dominus
[19:53:46] --> DominusExult has joined #exult
[19:53:47] --- ChanServ gives channel operator status to DominusExult
[19:56:12] <-- Dominus has left IRC (Ping timeout: 276 seconds)
[19:56:13] --- DominusExult is now known as Dominus
[20:08:09] <-- Malignant_Manor has left IRC (Quit: ChatZilla 0.9.89 [Firefox 18.0.1/20130116073211])
[20:21:32] <-- Dominus has left IRC (Read error: Connection reset by peer)
[20:22:05] --> Dominus has joined #exult
[20:22:05] --- ChanServ gives channel operator status to Dominus
[21:18:38] <-- Marzo has left IRC (Ping timeout: 252 seconds)
[21:49:38] --> Marzo has joined #exult
[21:54:33] <-- Dominus has left IRC (Read error: Connection reset by peer)
[21:55:03] --> Dominus has joined #exult
[21:55:04] --- ChanServ gives channel operator status to Dominus
[22:09:09] <-- TheCycoONE has left IRC (Quit: And then there were n-1)
[22:27:03] --> Kirben has joined #exult
[22:27:04] --- ChanServ gives channel operator status to Kirben
[22:53:49] <Dominus> so, I'm off for about a week. my snapshots should be built anyway unless something breaks them :)
[22:54:12] <-- Dominus has left IRC (Quit: good night)
[23:44:36] <-- Baastuul_ has left IRC ()
[23:55:42] --> Rottingbeef has joined #exult