| 00:19.34 | *** join/#brlcad stas (~stas@188.24.35.114) |
| 00:42.40 | CIA-128 | BRL-CAD: 03Cprecup 07http://brlcad.org * r3509 10/wiki/User:Cprecup: first step: compiling and installing BRL-CAD issues |
| 01:37.00 | *** join/#brlcad stevegt_1 (~stevegt@cislunar.TerraLuna.Org) |
| 01:56.00 | starseeker | cristina: you have the freetype dev packages? |
| 01:56.57 | starseeker | you may need to clear your CMake build directory as well and re-run CMake |
| 05:13.57 | *** join/#brlcad stevegt_ (~stevegt@c-69-181-134-76.hsd1.ca.comcast.net) |
| 07:48.59 | *** join/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 08:25.27 | *** part/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 08:25.55 | *** join/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 08:42.17 | *** part/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 08:46.06 | *** join/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 09:14.20 | *** join/#brlcad kane_ (~Mesut@dslb-084-063-252-087.pools.arcor-ip.net) |
| 11:29.38 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) |
| 11:56.30 | cristina | starseeker: yes, I have the following packages installed: freetype2-demos, libfreetype6, libfreetype6-dev, libftgl-dev, libftgl2, libxft-dev, libxft2. I've even tried the whole building with a new checkout (in case I didn't clear everything that needed to be cleared) |
| 12:10.11 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 12:22.49 | starseeker | cristina: ok, the next step is to try make VERBOSE=1 and look at the actual command line that's failing |
| 12:25.47 | starseeker | cristina: there's also something strange about *needing* the system libs, if you've set -DBRLCAD_BUNDLED_LIBS=Bundled |
| 12:29.38 | starseeker | you're doing the following?: |
| 12:29.54 | starseeker | svn co https://brlcad.svn.sf.net/svnroot/brlcad/brlcad/trunk brlcad |
| 12:29.57 | starseeker | cd brlcad |
| 12:32.18 | starseeker | cmake . -DCMAKE_BUILD_TYPE=Debug -DBRLCAD_BUNDLED_LIBS=Bundled |
| 12:32.26 | starseeker | make |
| 12:52.27 | *** join/#brlcad kane_ (~Mesut@dslb-084-063-252-087.pools.arcor-ip.net) |
| 13:15.15 | *** join/#brlcad Neil___ (~chatzilla@117.229.107.155) |
| 13:20.18 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 13:22.07 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 13:24.29 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 13:33.48 | CIA-128 | BRL-CAD: 03starseeker * r50353 10/brlcad/trunk/ (4 files in 4 dirs): Beef up distclean rule |
| 13:49.21 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 13:54.39 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 14:26.45 | *** join/#brlcad anuragmurty (~anurag@14.139.128.14) |
| 14:37.04 | *** join/#brlcad npcdoom (~npcdoom@190.39.142.150) |
| 14:37.05 | *** join/#brlcad npcdoom (~npcdoom@gugve/developer/npcdoom) |
| 14:49.01 | *** join/#brlcad Neil___ (~chatzilla@117.228.217.22) |
| 16:07.40 | *** join/#brlcad anuragmurty1 (~anurag@14.139.128.14) |
| 16:33.14 | CIA-128 | BRL-CAD: 03starseeker * r50354 10/brlcad/branches/STABLE/ (4 files in 3 dirs): Merge rtwizard fix to STABLE |
| 16:53.00 | cristina | starseeker: those are the steps that I made (the ones that you described). i used cmake . -DCMAKE_BUILD_TYPE=Debug without -DBRLCAD_BUNDLED_LIBS=Bundled because of someone's suggestion as an alternative |
| 16:53.08 | cristina | I'll use make VERBOSE=1 |
| 16:56.23 | *** join/#brlcad npcdoom (~npcdoom@190.39.142.150) |
| 16:56.23 | *** join/#brlcad npcdoom (~npcdoom@gugve/developer/npcdoom) |
| 17:19.48 | starseeker | cristina: try turning on BRLCAD_BUNDLED_LIBS |
| 17:20.04 | starseeker | that should work |
| 17:22.15 | starseeker | or rather, if both fail I'd like to see pastebins of both failures |
| 17:23.37 | cristina | starseeker: ok, this is the failure when BRLCAD_BUNDLED_LIBS is on: pastebin.mozilla.org/1604080 |
| 17:24.39 | cristina | I'll recompile it right now without BRLCAD_BUNDLED_LIBS and paste the errors as well |
| 17:31.15 | cristina | starseeker: this is withot BRLCAD_BUNDLED_LIBS flag: http://pastebin.mozilla.org/1604094 |
| 17:31.19 | cristina | *without |
| 17:38.25 | starseeker | cristina: what version of freetype do you have installed? |
| 17:39.14 | cristina | 2.4.8-1ubuntu2 |
| 17:39.20 | starseeker | and fontconfig? |
| 17:40.12 | cristina | fontconfig: 2.8.0-3ubuntu9 |
| 17:40.22 | starseeker | and the dev package for fontconfig is installed? |
| 17:41.12 | cristina | yes |
| 17:41.57 | cristina | i've got installed fontconfig, fontconfig-config, libfontconfig1-dev, libfontconfig1 |
| 17:42.03 | starseeker | ok... |
| 17:42.11 | starseeker | let me try something... |
| 17:42.20 | *** join/#brlcad brlcad1 (~anurag@14.139.128.12) |
| 17:51.04 | *** join/#brlcad brlcad2 (~anurag@14.139.128.12) |
| 17:51.12 | CIA-128 | BRL-CAD: 03starseeker * r50355 10/brlcad/trunk/ (3 files in 3 dirs): Try explicitly calling out fontconfig in addition to freetype. |
| 17:51.20 | starseeker | cristina: see if that helps |
| 18:01.29 | cristina | starseeker: I rerun cmake -DCMAKE_BUILD_TYPE=Debug -DBRLCAD_BUNDLED_LIBS=Bundled . after updating the repository. I am getting the same errors: http://pastebin.mozilla.org/1604080 |
| 18:04.25 | starseeker | one second... |
| 18:04.34 | starseeker | (sorry, hard to test this when I'm not seeing the failure...) |
| 18:05.03 | CIA-128 | BRL-CAD: 03starseeker * r50356 10/brlcad/trunk/src/other/tk/CMakeLists.txt: Use the right FONTCONFIG variable... |
| 18:05.39 | starseeker | that may work better |
| 18:08.32 | cristina | ok :), just a few minutes please |
| 18:08.40 | starseeker | no problem |
| 18:17.35 | cristina | starseeker: ok, so I assume it got pass that error because it says "Built target wish" but now it got into one that I've encountered before: http://pastebin.mozilla.org/1604116 |
| 18:17.48 | cristina | the one referring to the acos symbol |
| 18:19.56 | *** join/#brlcad mohitdhingras (~root@14.139.128.12) |
| 18:21.19 | *** part/#brlcad mohitdhingras (~root@14.139.128.12) |
| 18:21.42 | *** join/#brlcad mohitdhingras (~root@14.139.128.12) |
| 18:21.57 | starseeker | hmm. interesting |
| 18:22.07 | starseeker | let me try something... |
| 18:23.37 | *** join/#brlcad anuragmurty (~anurag@14.139.128.12) |
| 18:26.51 | CIA-128 | BRL-CAD: 03starseeker * r50357 10/brlcad/trunk/src/other/tcl/ (CMake/tcl.cmake CMakeLists.txt): Hmm - ieee lib link needs more study. Explicitly add the math library to the list of libs for tcl. |
| 18:27.26 | cristina | I'll make an update |
| 18:28.13 | starseeker | I'll be back in a bit |
| 18:28.39 | cristina | ok, thank you for your help |
| 18:29.01 | anuragmurty | hi all! My proposal for voxelize was selected for GSoC this year.. wanted to get some initial references to start with.. |
| 19:07.41 | cristina | starseeker: so the last update definitely got over the 'undefined symbol: acos' error. The compilation got at 83% but there's another build error referring to xslt: http://pastebin.mozilla.org/1604145 |
| 19:15.49 | starseeker | cristina: did any docbook files get compiled? |
| 19:16.03 | starseeker | anuragmurty: what were you looking for? |
| 19:16.18 | starseeker | anuragmurty: have you completed the initial steps outlined in our procedure? |
| 19:19.22 | cristina | starseeker: I'd say some got compiled: "Generating resources/other/docbook-schema/schema.sentinel" |
| 19:19.41 | starseeker | that's something else - I was thinking in particular about about.html |
| 19:20.51 | starseeker | also - if you look in the file build_region_article_html.cmake, what is the path to xsltproc? |
| 19:21.11 | anuragmurty | starseeker: yes. I did build and compile the source code without tcl bindings, but am recompiling it with them now. |
| 19:21.38 | starseeker | crdueck: you *might* be able to complete the doc building if you run make without any parallel flags |
| 19:21.49 | starseeker | sorry cristina not crdueck |
| 19:22.03 | starseeker | I thought we had that fixed, but maybe not... |
| 19:24.24 | starseeker | cristina: it's almost like it's not using our local xsltproc |
| 19:24.49 | starseeker | is this a build with or without the -DBRLCAD_BUNDLED_LIBS=Bundled setting? |
| 19:25.03 | starseeker | if the latter, try it again with the bundled libs |
| 19:25.15 | cristina | starseeker: it's a build with -DBRLCAD_BUNDLED_LIBS=Bundled |
| 19:25.31 | cristina | I am not using parallel flags, the command was just make VERBOSE=1 |
| 19:25.46 | starseeker | can you pastebin the contents of /home/cristina/work/gsoc2012/brlcad/doc/docbook/articles/en/build_region_article_html.cmake |
| 19:27.12 | cristina | starseeker: this is the path to xsltproc: //home/cristina/work/gsoc2012/brlcad/bin/xsltproc |
| 19:27.16 | cristina | and here is the content: |
| 19:27.50 | cristina | http://pastebin.mozilla.org/1604184 |
| 19:28.07 | starseeker | no, that's the .in file |
| 19:28.11 | starseeker | I need the generated cmake file |
| 19:28.24 | starseeker | build_region_article_html.cmake |
| 19:28.35 | starseeker | ah, wait |
| 19:28.43 | starseeker | sorry, nevermind - forgot I added the header |
| 19:28.57 | starseeker | wonder why we're getting the //home prefix |
| 19:29.33 | cristina | shouldn't it be that? |
| 19:30.00 | starseeker | I would have expected ""//home/cristina/work" to look like ""/home/cristina/work" |
| 19:30.05 | cristina | ah |
| 19:30.05 | cristina | yes |
| 19:34.00 | cristina | I've tested accessing the path in the terminal with 2/ and it makes no difference |
| 19:34.18 | starseeker | yeah, xsltproc might be more sensitive to that path structure |
| 19:34.46 | starseeker | I can't figure out why your path is getting reported like that though - why the two leading slashes? |
| 19:34.54 | starseeker | what version of CMake do you have? |
| 19:35.45 | starseeker | cristina: what happens if you do the following: |
| 19:36.01 | starseeker | cmake -P /home/cristina/work/gsoc2012/brlcad/doc/docbook/articles/en/build_region_article_html.cmake |
| 19:36.08 | starseeker | (all one line) |
| 19:36.25 | cristina | cmake 2.8.7 |
| 19:37.02 | cristina | I get this: runtime error |
| 19:37.02 | cristina | xsltApplyStylesheet: saving to //home/cristina/work/gsoc2012/brlcad/share/brlcad/7.21.0/doc/html/articles/en/build_region.html may not be possible |
| 19:37.02 | cristina | CMake Error at doc/docbook/articles/en/build_region_article_html.cmake:55 (message): |
| 19:37.02 | cristina | <PROTECTED> |
| 19:37.28 | starseeker | now, what happens if you remove the //home instances in that file and replace them with /home ? |
| 19:39.11 | cristina | ah |
| 19:39.15 | cristina | now it works |
| 19:39.24 | cristina | nothing was printed, no error |
| 19:39.38 | starseeker | so - why are your paths adding a leading slash? |
| 19:46.16 | cristina | maybe something appends to that command a slash |
| 19:48.09 | cristina | waw |
| 19:48.10 | cristina | ok |
| 19:48.30 | CIA-128 | BRL-CAD: 03starseeker * r50358 10/brlcad/trunk/misc/CMake/DocBook.cmake: Try to make input paths more 'DocBook friendly' |
| 19:48.36 | starseeker | cristina: that may work around the issue |
| 19:48.38 | cristina | so by looking at all my environment variables I can see OLDPWD= OLDPWD=//home/cristina/work/gsoc2012/brlcad/src/other |
| 19:48.49 | cristina | and PWD=//home/cristina/work/gsoc2012/brlcad |
| 19:49.16 | cristina | sorry, there was a typo, it's just OLDPWD=//home/cristina/work/gsoc2012/brlcad/src/other |
| 19:49.25 | starseeker | hmm. |
| 19:49.28 | starseeker | weird. |
| 19:49.44 | starseeker | it may be moot - see if r50358 gets by it |
| 19:49.49 | cristina | ok |
| 19:57.32 | CIA-128 | BRL-CAD: 03starseeker * r50359 10/brlcad/trunk/src/other/ (4 files in 4 dirs): keep all the tcl.cmake copies the same to avoid surprises |
| 20:20.50 | cristina | starseeker: thanks a lot. Everything builds fine now. I'm sorry for bothering you |
| 20:40.29 | starseeker | cristina: not a all - those are exactly the kind of problems we want to find and fix |
| 20:41.09 | starseeker | it's quite interesting you were seeing issues no one else has reported |
| 20:41.30 | starseeker | the "//home" issue I'm not surprised was new, but the tcl/tk building issues were a surprise |
| 21:20.29 | CIA-128 | BRL-CAD: 03starseeker * r50360 10/brlcad/trunk/src/libbu/ (bitv.c endian.c fnmatch.c): These uses of inline cause a strict failure with clang 3.0 in c89 mode (note that they do work with c99) |
| 21:23.01 | CIA-128 | BRL-CAD: 03starseeker * r50361 10/brlcad/trunk/src/libpkg/example/server.c: clang doesn't let us get away with this style of quellage - add the UNUSED wrapper. |
| 21:28.20 | starseeker | brlcad: getting a bunch of strict failures in librt with clang related to the BU_LIST_MAIN_PTR macro: |
| 21:28.24 | starseeker | http://pastebin.mozilla.org/1604369 |
| 21:36.27 | CIA-128 | BRL-CAD: 03starseeker * r50362 10/brlcad/trunk/src/fbed/empty.c: Another quellage we can't get away with in clang. |
| 21:57.00 | CIA-128 | BRL-CAD: 03starseeker * r50363 10/brlcad/trunk/src/ (libpkg/example/client.c librt/primitives/metaball/metaball.c): Couple more clang strict c89 issues |
| 22:03.28 | starseeker | ``Erik: this looks like it might be up your alley: http://pastebin.mozilla.org/1604412 |
| 22:18.40 | CIA-128 | BRL-CAD: 03starseeker * r50364 10/brlcad/trunk/src/lgt/ (lgt.c reflect.c): More quellage fun |
| 22:24.14 | CIA-128 | BRL-CAD: 03starseeker * r50365 10/brlcad/trunk/src/tab/tabinterp.c: hopefully the last of the quellage errors - looks like we may be down to offsetof and other offset-type issues with gnu89 |
| 22:54.48 | *** join/#brlcad KimK (~Kim__@209.248.147.2.nw.nuvox.net) |
| 23:04.06 | brlcad | starseeker: act, the removal of inline is nfg ... that will mess a few things up |
| 23:04.20 | brlcad | better to test for it during cmake and #define to nothing if not supported |
| 23:04.28 | brlcad | but there's undoubtedly a way to support it |
| 23:10.51 | *** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com) |