#exult@irc.freenode.net logs for 19 Jun 2013 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[03:50:46] <-- Marzo has left IRC (*.net *.split)
[03:50:47] <-- Eviltar has left IRC (*.net *.split)
[03:51:57] <-- sh4rm4 has left IRC (*.net *.split)
[03:52:48] --> Marzo has joined #exult
[03:52:48] --> Eviltar has joined #exult
[03:53:43] --> sh4rm4 has joined #exult
[03:54:00] <-- Marzo has left IRC (*.net *.split)
[03:54:07] <-- Eviltar has left IRC (*.net *.split)
[03:55:39] --> Marzo has joined #exult
[03:55:39] --> Eviltar has joined #exult
[06:09:59] <-- nutron has left IRC (Read error: Operation timed out)
[06:12:17] --> nutron has joined #exult
[07:17:25] --> Colourless has joined #exult
[07:17:25] --- ChanServ gives channel operator status to Colourless
[08:11:23] <-- Lightkey has left IRC (Ping timeout: 240 seconds)
[09:51:06] --> ShamblerDK has joined #exult
[10:17:29] <-- sh4rm4 has left IRC (Ping timeout: 240 seconds)
[10:22:36] --> sh4rm4 has joined #exult
[11:31:02] <-- Marzo has left IRC (Ping timeout: 256 seconds)
[12:11:57] --> TheCycoONE has joined #exult
[12:27:25] --> Marzo has joined #exult
[12:34:47] <-- TheCycoONE has left IRC (Ping timeout: 256 seconds)
[12:37:48] --> TheCycoONE has joined #exult
[12:46:24] <-- Marzo has left IRC (Ping timeout: 256 seconds)
[12:57:32] <-- TheCycoONE has left IRC (Ping timeout: 252 seconds)
[13:14:15] --> Marzo has joined #exult
[13:21:19] <-- Marzo has left IRC (Read error: Connection timed out)
[13:22:47] --> Marzo has joined #exult
[13:23:06] --> TheCycoONE has joined #exult
[13:32:59] --> TheCycoTWO has joined #exult
[13:35:23] --> TheCycoTHREE has joined #exult
[13:36:20] <-- TheCycoONE has left IRC (Ping timeout: 246 seconds)
[13:36:59] <-- TheCycoTHREE has left IRC (Read error: Connection reset by peer)
[13:38:23] <-- TheCycoTWO has left IRC (Ping timeout: 240 seconds)
[13:41:19] --> TheCycoONE has joined #exult
[15:17:53] <-- Marzo has left IRC (Ping timeout: 245 seconds)
[15:43:49] --> Marzo has joined #exult
[16:17:36] <-- Marzo has left IRC (Ping timeout: 256 seconds)
[16:21:57] --> Marzo has joined #exult
[18:02:45] <-- TheCycoONE has left IRC (Ping timeout: 248 seconds)
[18:03:54] --> TheCycoONE has joined #exult
[19:36:50] <-- Marzo has left IRC (Ping timeout: 246 seconds)
[19:58:35] --> Lightkey has joined #exult
[20:35:59] --> Marzo has joined #exult
[21:13:04] <-- TheCycoONE has left IRC (Read error: Connection reset by peer)
[21:15:41] --> i30817 has joined #exult
[21:16:13] <i30817> Usecode compiler is not building on Ubuntu Saucy on launchpad.
[21:17:38] <Lightkey> !!
[21:18:09] <Lightkey> i30817: do have Portuguese (not BR_pt) LucasArts games by any chance?
[21:19:17] <i30817> Never did, i only use english version, who would play comedy games not on the original language anyway if you knew it.
[21:19:21] <i30817> Error occurs on both 64 bits and 32 bits
[21:19:22] <i30817> make[4]: Entering directory `/build/buildd/exult-1.5.0+svn20130618~saucy1/usecode/compiler'
[21:19:24] <i30817> /bin/bash ../../ylwrap ucparse.yy y.tab.c ucparse.cc y.tab.h `echo ucparse.cc | sed -e s/cc$/hh/ -e s/cpp$/hpp/ -e s/cxx$/hxx/ -e s/c++$/h++/ -e s/c$/h/` y.output ucparse.output -- bison -y -d -v
[21:19:25] <i30817> updating ucparse.output
[21:19:27] <i30817> updating ucparse.hh
[21:19:28] <i30817> g++ -DHAVE_CONFIG_H -I. -I../.. -I./../../headers -I./.. -I./../../files -I./../.. -O2 -Wno-long-long -O2 -c -o ucparse.o ucparse.cc
[21:19:30] <i30817> /bin/bash ../../ylwrap uclex.ll lex.yy.c uclex.cc -- flex
[21:19:31] <i30817> make[4]: *** No rule to make target `ucparse.h', needed by `uclex.o'. Stop.
[21:19:54] <i30817> Maybe the reason for this error is that byson or flex were updated?
[21:19:54] <Dominus> i30817: yeah, sounds like the problem in the bug tracker we haven't figured out yet
[21:20:28] <i30817> I didn't pin version on the control file; so this is probably happening on most Saucy compiles.
[21:20:39] <i30817> version of anything that is.
[21:21:18] <i30817> But it's only on Saucy.
[21:21:22] <Dominus> https://sourceforge.net/p/exult/bugs/1761/
[21:22:15] <Dominus> maybe you can add something to it.
[21:22:27] <Dominus> we really have no clue what is going wrong there
[21:23:14] <i30817> Idk, i'd guess it's a fuckup on upstream. Maybe a backwards compatibility grammar bug.
[21:23:56] <i30817> I don't know how to work with lex.
[21:24:56] <i30817> ucparse.h obviously isn't there. Maybe it was autogenerated empty on previous versions and now isn't?
[21:25:19] <Dominus> did you read the bug report?
[21:25:49] <i30817> yeah, it's the same error.
[21:25:58] <Dominus> for some reason you end up with ucparse.hh instead of ucparse.h
[21:26:31] <i30817> Oh.
[21:27:15] <i30817> The sed line?
[21:27:17] <i30817> /bin/bash ../../ylwrap ucparse.yy y.tab.c ucparse.cc y.tab.h `echo ucparse.cc | sed -e s/cc$/hh/ -e s/cpp$/hpp/ -e s/cxx$/hxx/ -e s/c++$/h++/ -e s/c$/h/` y.output ucparse.output -- bison -y -d -v
[21:29:29] <Dominus> I can't immediately tell you the correct line or what is different here
[21:30:46] <Dominus> on the way to bed in a few... But we have discussed it unfruitful before in here. you can search around the logs at about the time of the bug creation
[21:30:54] <i30817> sed -e s/cc$/hh/ <- replace all 'cc' at the end of file with hh (not sure if end of line goes bye-bye)
[21:31:40] <i30817> s/cpp$/hpp/ replace all cpp at the end of line (etc)
[21:32:19] <wjp> you're talking about code generated by automake now I think?
[21:32:49] <wjp> it's likely just a matter of replacing ucparse.h in Makefile.am by ucparse.hh, but dunno
[21:33:08] <wjp> of course doing that unconditionally will break things for other automake/ylwrap/whatever versions
[21:33:19] <wjp> the trick is figuring out what works everywhere
[21:33:38] <wjp> and we haven't taken the time to look at the particular combination of tools in your distro yet
[21:36:03] <i30817> It's autogenerated? I have no idea really. Seems like some rename is not doing its job correctly to me. If the file uparse.hh exists, but sed replaced all references to it by uparse.h ...
[21:37:11] <i30817> Strange it happens only on a new distro version (considering the tools also update on the others)
[21:37:51] <wjp> 23:32 <@wjp> it's likely just a matter of replacing ucparse.h in Makefile.am by ucparse.hh, but dunno
[21:37:54] <wjp> try that
[21:41:55] <wjp> (that's usecode/compiler/Makefile.am, by the way)
[21:43:31] <i30817> I see, there is even a note there:
[21:43:33] <i30817> AM_YFLAGS = -d -v # Want ucparse.h.
[21:43:34] <i30817> (whatever AM_YFLAGS d and v are)
[21:44:56] <wjp> flags for bison
[21:45:16] <wjp> generate header and verbose, resp.
[21:47:10] <i30817> Ah yes:
[21:47:11] <i30817> http://www.gnu.org/software/automake/manual/html_node/Yacc-and-Lex.html
[21:47:18] <i30817> AM_YFLAGS is usually used to pass the -d option to yacc. Automake knows what this means and will automatically adjust its rules to update and distribute the header file built by ‘yacc -d’4. What Automake cannot guess, though, is where this header will be used: it is up to you to ensure the header gets built before it is first used. Typically this is necessary in order for dependency...
[21:47:19] <i30817> ...tracking...
[21:47:21] <i30817> ...to work when the header is included by another file. The common solution is listing the header file in BUILT_SOURCES (see Sources) as follows.
[21:47:47] <i30817> What a mess
[21:49:57] <i30817> Guess the 'correct' solution is to add
[21:49:58] <i30817> BUILT_SOURCES = uparse.h
[21:50:00] <i30817> ?
[21:53:31] <i30817> 'ucparse.yy' shouldn't it be 'ucparse.y' instead?
[21:54:23] <i30817> Ah no, it has got to be the hh form because it's c++ i guess.
[21:54:38] <i30817> So yy is right.
[21:54:40] <wjp> so... did you already try replacing ucparse.h by ucparse.hh at the bottom of Makefile.am?
[21:55:33] <i30817> Yes, going to try that now; but i'm not really building locally and that source file is in your repo; so i'll need to make a patch file to upload to launchpad.
[21:55:49] <i30817> If it works, could you edit that in upstream?
[21:55:50] <wjp> why not build locally?
[21:55:52] <wjp> no
[21:56:04] <wjp> 23:33 <@wjp> of course doing that unconditionally will break things for other automake/ylwrap/whatever versions
[21:56:07] <wjp> 23:33 <@wjp> the trick is figuring out what works everywhere
[21:56:41] <wjp> doing experimentation like this offline is really painful; I'd really recommend doing it locally
[21:57:35] <i30817> Really think it wouldn't work anywhere? I can't see why. Accepting .h file as a alias for a .hh file is the real bug anyway.
[21:58:03] <wjp> no, other platforms generate ucparse.h
[21:58:05] <wjp> and not ucparse.hh
[21:58:09] <i30817> ugh
[21:58:48] <wjp> which is why it works for all other platforms :-)
[21:58:52] <i30817> Why? It's a cc file.
[21:58:59] <wjp> ask the bison people
[21:59:03] <wjp> or the automake people
[21:59:04] <wjp> dunno
[21:59:06] <i30817> uh, .yy file
[21:59:23] <wjp> or ylwrap
[21:59:34] <wjp> or whichever of these tools is doing this particular bit of renaming
[22:00:13] <wjp> you seem to be under the impression there are standard conventions for these kinds of naming :-)
[22:00:33] <i30817> Yes... used to java.
[22:01:55] <i30817> Ok, it will be a patch file then. If you ever 'fix' it upstream it will only be already applied.
[22:02:23] <i30817> grumble, sparse clone, blah
[22:02:31] <wjp> no guarantees this will work, though
[22:07:29] <-- sh4rm4 has left IRC (Ping timeout: 240 seconds)
[22:10:49] --> sh4rm4 has joined #exult
[22:14:09] <i30817> Sigh, have to create a new ssh key for my launchpad account (nuked computer)
[22:15:25] <i30817> Really, was i supposed to know that the key wouldn't be in the gnupgp dir...
[22:29:59] <-- i30817 has left IRC (Ping timeout: 260 seconds)
[22:48:46] --> i30817 has joined #exult
[23:01:30] <-- i30817 has left IRC (Quit: ChatZilla 0.9.90 [Firefox 21.0/20130512193848])
[23:08:07] <-- ShamblerDK has left IRC (Remote host closed the connection)