00:00:02 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.24-a0-284-g9fb9971 (34) 01:38:20 Unstable branch on crawl.develz.org updated to: 0.24-a0-284-g9fb9971 (34) 01:39:34 -!- amalloy_ is now known as amalloy 01:55:27 -!- TAS-2012v is now known as TAS_2012v 02:03:49 -!- TAS-2012v is now known as TAS_2012v 02:09:53 Windows builds of master branch on crawl.develz.org updated to: 0.24-a0-284-g9fb9971 02:59:31 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-284-g9fb9971 03:55:01 Fork (stoatsoup) on crawl.kelbi.org updated to: 0.21.2-955-g4933f74f56 05:11:44 -!- TAS-2012v is now known as TAS_2012v 05:26:35 Pinkbeast: account names like that one which are basically just ads are no longer permitted on CKO 08:53:00 -!- amalloy is now known as amalloy_ 11:23:13 Stable (0.23) branch on underhound.eu updated to: 0.23.1-50-ga50397a6bc 11:39:49 Pinkbeast: how does one build ipbt? 11:40:14 (I'm not using jettyplay because I don't have old java around and I'm not signing up with oracle to get it) 11:46:08 ebering: ./configure; make ; make install # I feel like this is a bit Capt Obvious and you might be asking another question 11:46:46 Pinkbeast: times they have a changed 11:47:01 it's autoconf now 11:47:43 Oh, hang on, this source archive is from 2017 11:48:38 ... but same procedure with ipbt-20180824.dc543d3.tar.gz 11:49:01 ah 11:49:24 the git repo does not have the nicities of the package 11:50:37 Yes. I think Simon has... something that builds the package for distribution from the git repo, but what that is isn't obvious. I'll ask him tomorrow, if you like. 11:51:51 I'll build froom the tarball (and hope it works on osx) 11:52:10 my options are this or jettyplay which, as mentioned requires OldJava 11:52:48 If it doesn't build on OS X I imagine that can be fixed. 11:53:25 (also when running it you tend to always want -U with DCSS recordings) 11:54:04 It has built 11:58:00 ... does it work? 11:58:26 gotta get a ttyrec first :) 11:59:33 it's running 12:01:08 it works! 12:01:22 yay! Until you get a hugeterm ttyrec and don't notice 12:02:13 thankfully hugeterm players seme to be gooing extinct 13:05:34 Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-284-g9fb9971 (34) 13:28:01 Pinkbeast: https://i.imgur.com/3b1qJm6.gif the end result 13:42:39 that's a gorgeous terminal 13:44:06 ^ 13:44:21 ebering: please takes notes on how you get such beautiful terminal for future highlighters 13:48:12 * Pinkbeast makes the bold_brightens_foreground face 14:00:26 alexjurk1: I use these colours http://csclub.uwaterloo.ca/~ebering/crawl/rl-solarized.txt 14:00:37 and menlo regular 14:03:20 I'll make a notes page tho 14:10:23 http://www.chiark.greenend.org.uk/~damerell/temp/crawl.png # I don't really hold with pastel terminals :-) 15:53:50 -!- amalloy_ is now known as amalloy 17:21:29 -!- aidanh_ is now known as aidanh 17:37:02 Has anyone had any issues building dgamelaunch in Debian? Following: https://crawl.develz.org/wiki/doku.php?id=dcss:server:setting_up_dgamelaunch_and_webtiles. 17:37:09 cc: error: lex.yy.c: No such file or directory 17:37:09 cc: fatal error: no input files 17:38:54 pwnmonkey: Suggest you pastebin a complete transcript. 17:39:30 output of make: https://pastebin.com/4j5QZNFa 17:42:27 flex isn't installed 17:42:43 is bison not a replacement? 17:43:29 bison creates y.tab.c. flex creates lex.yy.c 17:43:42 bison is a parser generator, flex is a lexical analyzer generator 17:43:50 gotcha 17:45:01 Do you happen to know the package name on Debian? It's not just "flex" 17:46:40 It's "flex" here 17:46:55 and it's flex on mint/ubuntu so I'd expect it to be on debian as well 17:47:32 hmm interesting, still getting the same error. Do I need to autogen again or something? 17:47:38 yes 17:56:09 Even after autogen.sh still getting the same issue. Does this maybe have to do with it not finding ncurses? (libncursesw5-dev confirmed installed.) libncursesw5-dev 17:56:09 https://pastebin.com/ThWQdCgc 17:56:52 it won't have written out proper makefiles because of that error 17:59:52 Hmm, I wonder why it cant find ncurses. 18:02:22 do you have a config.log? 18:04:04 yeah but it's very very long 18:04:36 "/usr/bin/ld: cannot find -lcursesw" 18:05:01 longer than my terminal history so I can't copy it from putty 18:06:21 "Not using ncursesw, not using color." 18:07:35 Do you have libncursesw5 as well libncursesw5-dev 18:08:06 libncursesw5 is already the newest version (6.1+20181013-2). 18:08:07 libncursesw5-dev is already the newest version (6.1+20181013-2). 18:08:32 I think we need to see the config.log, can't you scp it off? 18:08:38 Yeah I will 18:13:13 https://pastebin.com/atpwZkwM 18:14:12 /usr/bin/ld: /usr/lib/gcc/x86_64-linux-gnu/8/../../../x86_64-linux-gnu/libfl.so: undefined reference to `yylex' 18:14:28 bleh. it shouldn't be including -lfl there 18:14:54 It did here 18:15:26 Please try installing libfl-dev 18:15:37 yeh, it's safe as long as flex's runtime doesn't invoke yylex() itself. which apparently used to be true but isn't in whatever version they have installed. 18:15:41 libfl-dev is already the newest version (2.6.4-6.2). 18:15:55 that said, a dummy yylex() in the test file would work here 18:17:36 pwnmonkey: What about "flex" 18:17:47 (I mean, the package of that name) 18:18:30 Mind you libfl-dev Depends on flex so it jolly well ought to be installed 18:19:34 flex is already the newest version (2.6.4-6.2). 18:22:15 ls -l /etc/alternatives/cc 18:22:25 lrwxrwxrwx 1 root root 12 Apr 11 11:54 /etc/alternatives/cc -> /usr/bin/gcc 18:23:00 You've got a much more recent gcc. I think geekosaur has a better understanding of what's going on here... 18:23:51 I did notice that autogen was trying cc -V 18:23:57 not cc -v, so yeah. 18:26:53 basically linking in -fl requres that yylex() be defined, but the generated configure test file doesn't 18:27:21 hilfy «crawl:master» Z$ nm --dynamic /usr/lib/x86_64-linux-gnu/libfl.so | grep yylex 18:27:22 U yylex 18:27:29 geekosaur: OK but why did it _ever_ work? 18:27:43 that's why I said some earlier version of flex didn't do that. 18:27:45 I did a make clean, is there an equivalent autogen clean? 18:28:19 back iin the lex days, -ll included only a couple of utilities used by a generated yylex, not calls to it 18:29:00 heh, it also defines main() these days. that's definitely different 18:30:10 pwnmonkey: Is either copy of you still here? 18:30:42 also it'd have worked with static linking since that should omit its main() 19:22:21 yeah sorry 19:22:24 Pinkbeast 19:22:30 had to step out for a minute. 19:29:07 Suggest you run "./configure LEXLIB=" 19:34:01 kk 19:35:17 https://pastebin.com/d5wnKwvU 19:35:53 and output of make: https://pastebin.com/YArRVSyW 19:48:25 the hacky way around this is CC='cc -Dyywrap=1' 19:48:49 the correct way is to not have -lfl in the configure check for ncurses, since it's not relevant there 19:49:00 but that means editing configure.ac 19:51:36 hm, not sure that works especially if it's generating GNU style C 19:52:09 yeah hacky way doesn't work 19:52:20 make: cc -Dyywrap=1: Command not found (if that's what you meant) 19:52:40 oh 19:52:42 i messed up 19:53:01 "./configure LEXLIB= CC='-Dyywrap=1'" 19:53:03 er 19:53:07 "./configure LEXLIB= CC='cc -Dyywrap=1'" 19:53:20 ah 19:53:21 note the space after the first = 20:05:00 yeah seems to be the same issue: https://pastebin.com/Dhp5U1PT 20:08:55 not quite the same issue, and one of the things I'd suspected might go wrong 20:10:22 heh.just realized the flex stuff is slightly broken anyway. it does try to check for it not being found, but the check is wrong 20:12:06 and I just spotted why this is broken. and am not sure the obvious workaround will work 20:12:11 haha interesting, I'm in web dev so all this makefile and cc stuff is over my head. 20:12:40 hm, actually it's ancient, not fixing a different bug 20:13:25 move line 18 of configure.ac to after line 85 20:13:44 and rerun the autogen thing, without the LEXLIB or CC parameters 20:19:14 https://pastebin.com/5ujuum3H 20:20:13 heh. same issue, different program which doesn't even care. :/ 20:21:42 this is a pain. 20:22:02 The entire purpose of this machine is for a crawl server, is there a specific debian version that would resolve this? 20:22:41 I don't know what version might have a flex that doesn't include that 20:25:29 debian has a package called "flex-old" 20:25:34 I might try that 20:26:45 but it doesnt look like there is a libfl "old" 20:27:39 for that one you remove libfl-dev 20:27:45 well I seemed to have a least changed the error 20:27:48 flex-old includes it 20:27:49 yeah I did just to try 20:28:19 hey it worked :) 20:28:24 or rather, didnt error on make 20:28:46 I had forgotten to add -pthread back after my last autogen 20:31:07 no issue tracker on dgl :/ 20:32:37 |amethyst, are you tracking any of this? 20:38:47 * geekosaur patches the wiki page to note the flex-old package, for now 20:41:43 which won't help anyone trying to build on e.g. fedora; ideally configure.ac and the makefile should be reworked to do this correctly (LEXLIB only used in the dgamelaunch link) 20:51:45 cool yeah, I can open the issue if you want as well. Thanks for the help 20:57:06 Hello! 20:57:51 Is it intended behaviour that monsters can teleport in the gauntlet? 22:18:57 :I 22:21:17 hi gammafunk 22:21:27 I/ 22:21:30 Er 22:21:46 o/ 23:39:24 I seemed to have found another build issue with "sudo /home/crawl-dev/dgamelaunch-config/bin/dgl update-trunk" https://pastebin.com/x2yw6PW0 23:44:24 you need to install it in the chroot 23:49:12 hm, maybe not. you are also doing it via sudo, but you have t installed in *your* user's python modules, not the ones root uses 23:50:44 sudo apt-get install python-yaml 23:50:53 may have to rerun that in the chroot as well