| 00:21.47 | CIA-85 | BRL-CAD: 03starseeker * r37969 10/brlcad/trunk/src/libgcv/Makefile.am: Add the distcleanfiles list to libgcv. |
| 00:24.48 | CIA-85 | BRL-CAD: 03starseeker * r37970 10/brlcad/trunk/configure.ac: Old version is more verbose, get both words outta there. |
| 00:30.58 | CIA-85 | BRL-CAD: 03starseeker * r37971 10/brlcad/trunk/src/libgcv/Makefile.am: obj_rules.h depends on obj_rules.cc. |
| 00:39.25 | starseeker | GRRRR |
| 00:39.40 | starseeker | why is distcheck ignoring the if flags and running flex??? |
| 00:54.21 | CIA-85 | BRL-CAD: 03starseeker * r37972 10/brlcad/trunk/src/libgcv/obj/ (5 files): Pre-generated files are now prefixed with _static |
| 01:05.59 | starseeker | confound it autotools |
| 01:07.54 | starseeker | thinks he may see it now - autotools is pre-running all possible lex/yacc based rules in order to stash the pre-generated sources in the tarball. Problem is, that's exactly what I DON'T want it to do. |
| 01:17.27 | *** join/#brlcad PrezKennedy (Matthew@whitecalf.net) | |
| 01:33.58 | starseeker | brlcad: OK, I give. It looks like the ll and yy file could be "dumbed down" to the level of the older tools by defining the old style global variables and such (and forefiting any advantages of reentrant behavior) |
| 01:35.27 | starseeker | the above changes successfully compile on both setups, but distcheck will always fail without the newer flex etc. because apparently autotools insists in "pre-processing" the lex and yacc code for a distcheck |
| 01:37.09 | starseeker | should I start dumbing down the ll and yy code to force it to work with flex 2.5.4 and friends? |
| 01:39.10 | starseeker | or are we ok with needing the newer tools for a distcheck? |
| 03:20.30 | *** join/#brlcad ``Erik_ (~erik@c-69-140-109-104.hsd1.md.comcast.net) | |
| 03:20.42 | *** join/#brlcad starseek1r (~starseeke@BZ.BZFLAG.BZ) | |
| 03:28.00 | *** join/#brlcad starseeker (~starseeke@BZ.BZFLAG.BZ) | |
| 03:28.00 | *** join/#brlcad ``Erik (~erik@c-69-140-109-104.hsd1.md.comcast.net) | |
| 04:25.25 | CIA-85 | BRL-CAD: 03starseeker * r37973 10/brlcad/trunk/configure.ac: Oooops, typo. |
| 04:36.47 | starseeker | blinks - obj code built on my gentoo box this time |
| 05:38.37 | brlcad | starseeker: shouldn't need to have the tools for a distcheck, but could make configure require a min if it is to compile them at all (similar to src/mged/Makefile.am keying off of WITH_PARSERS to decide whether to traverse src/mged/points) |
| 05:38.55 | brlcad | probably would prefer to backport, though |
| 05:39.58 | brlcad | being reentrant doesn't buy us anything useful that cannot be provided via other mechanisms |
| 05:41.26 | brlcad | also, fyi -- if you list a file as a BUILT_SOURCES, that means it goes in the dist |
| 05:41.45 | brlcad | those files could be made a separate rule in the meantime and the lexer/parser files just extra disted until it's all sorted out |
| 05:42.28 | brlcad | it's release time, so more concerned about getting everything locked down solid, tested, and tagged this week |
| 06:37.54 | brlcad | pretty awesome: http://www.methods.co.nz/asciidoc/ |
| 06:38.32 | brlcad | lets us take something similar to our HACKING or README files and generate Docbook from them |
| 06:41.14 | brlcad | technically could do all docs as simple text files, providing simple ease of editing (via simplified tagless text markup) |
| 06:42.20 | brlcad | the show-stopper is probably being able to include files within files, but then it's still useful for the few docs that need to stay in ascii format (e.g., the CAPS files) |
| 06:46.46 | brlcad | hm, looks like it does have an inclusion mechanism.. something to think about |
| 07:22.07 | *** join/#brlcad 15SAAIU7J (~stevegt@c-67-164-110-226.hsd1.ca.comcast.net) | |
| 08:37.36 | *** join/#brlcad roberthl (~robert@2001:ba8:1f1:f03d::2) | |
| 08:37.36 | *** join/#brlcad roberthl (~robert@silentflame/member/roberthl) | |
| 08:51.09 | louipc | I've used asciidoc before and it's horribly implemented |
| 08:51.18 | louipc | performance is in the toilet |
| 08:51.52 | louipc | probably doesn't scale very well, but your milage may vary :P |
| 12:06.09 | *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ) | |
| 12:30.43 | ``Erik | 'tagless', or a mortal friendly tagging system? (a la wiki) |
| 12:34.27 | CIA-85 | BRL-CAD: 03d_rossberg * r37974 10/brlcad/trunk/src/libgcv/ (CMakeLists.txt obj/obj_rules_static.cc): made it compile with MS Visual Studio (CMake build with *_static.* files) |
| 13:30.51 | d-lo | ``Erik: yeah baby! Crank it up! lol |
| 13:32.04 | ``Erik | richard turned my volume all the way down yesterday, had to re-level it :/ |
| 13:32.27 | ``Erik | <-- musta been in a very quiet track when he left, thought it was stopped *shrug* |
| 13:32.37 | d-lo | lol |
| 13:32.42 | d-lo | suuuure. :) |
| 13:33.14 | ``Erik | heh, ask him when he gets in O.o |
| 13:33.15 | d-lo | What's rich's problem anyways? A little metal gets the productivity (and heartrate) up! |
| 13:33.45 | ``Erik | anything involving: noise, smell, light, etc renders him unable to be productive... *cough* |
| 13:34.58 | ``Erik | keith and I both like the noisy cave coder environment, so *shrug* that's why I keep suggesting we send him upstairs O:-) |
| 13:35.54 | d-lo | So we need to make two big rooms and put a "Warning: Hearing Damage" sign on one :) |
| 13:36.46 | ``Erik | well, downstairs, I think richard is the odd man out for environment (this side of the restrooms) *shrug* |
| 13:41.17 | CIA-85 | BRL-CAD: 03starseeker * r37975 10/brlcad/trunk/src/libgcv/Makefile.am: Take a stab at doing the obj building without using BUILT_SOURCES |
| 13:41.26 | *** join/#brlcad parigaudi (~quassel@pd95b7f5e.dip0.t-ipconnect.de) | |
| 13:48.35 | starseeker | dunno if that fixes distcheck or not, can't test here |
| 13:49.39 | brlcad | yawns |
| 13:49.58 | brlcad | ``Erik: mortal friendly -- it'd parse up our hacking file without hardly any changes |
| 13:50.16 | brlcad | recognizing the sections, paragraphs, separators, etc |
| 13:51.00 | brlcad | free of "tags", just not necessarily free of markup |
| 13:53.24 | CIA-85 | BRL-CAD: 03starseeker * r37976 10/brlcad/trunk/src/libgcv/Makefile.am: Don't need OBJ_BUILT if taking this approach... |
| 13:55.06 | starseeker | grrrrr. it's still running the lex and yacc commands here... |
| 13:59.29 | starseeker | fine, I'll try to backport it |
| 13:59.34 | starseeker | this sucks |
| 13:59.58 | starseeker | whole day figuring out how to do version checking down the drain... |
| 14:00.11 | starseeker | plus the old, crudy global variable approach to things |
| 14:00.15 | starseeker | mutter, mutter... |
| 14:01.21 | starseeker | would prefer to insist on updated tools but knows it's impractical... |
| 14:09.22 | ``Erik | tradeoffs *shrug* I'd rather say "thou shalt have automake 1.9+" for per target CPP flags, but *shrug* :) |
| 14:20.40 | ``Erik | sweet, our incrTcl build is all busted to hell due to pstdint.h being included before stdint.h O.o |
| 14:21.03 | ``Erik | (on fbsd, that is) |
| 14:37.12 | brlcad | still hitting the rounds to make things portable with that new header |
| 14:46.38 | CIA-85 | BRL-CAD: 03brlcad * r37977 10/brlcad/trunk/include/common.h: only include stdint.h or pstdint.h if one doesn't seem to be included already. define the __STDC_CONSTANT_MACROS and __STDC_LIMIT_MACROS so that we get consistent behavior for C++ compilation as well. |
| 14:47.18 | brlcad | that should prevent pstdint.h from getting included |
| 14:47.28 | brlcad | doesn't need to get in the way if it's a proper c99 system |
| 14:47.36 | d_rossberg | :))) it looks like rt^3 has a problem with BRL-CAD's version number |
| 14:47.52 | brlcad | heh |
| 14:48.15 | brlcad | d_rossberg: how so? |
| 14:49.23 | d_rossberg | autogen fails because of "configure.ac:106: error: AC_SUBST: `0' is not a valid shell variable name" |
| 14:50.01 | d_rossberg | there should be the MAJOR_VERSION number |
| 14:51.28 | d_rossberg | then i looked at the scripts an found a "cat include/conf/MAJOR" there |
| 14:51.32 | d_rossberg | :))) |
| 14:52.38 | brlcad | that's for rt^3's "version", not the brlcad module's |
| 14:52.57 | brlcad | rt^3's got the same version files in include/conf/ .. presently just set to 0.1.0 |
| 14:54.24 | brlcad | sounds like it didn't do the m4 correctly or something |
| 14:56.21 | brlcad | there is a define() macro that should set a variable called "MAJOR_VERSION" with a value equal to the contents of include/conf/MAJOR (via cat) .. then AC_SUBST macro call on that variable |
| 14:56.50 | d_rossberg | should the cmake build work anyway? i had to make some adjustments to get it working (e.g. tcl85 => tcl8.5) |
| 14:56.52 | brlcad | sounds like it's not setting the shell variable somehow and only creating an m4 var |
| 14:57.44 | brlcad | they should both work, but haven't been made robust |
| 14:58.07 | brlcad | it's a bit of a mess at the moment, last I looked |
| 14:58.24 | brlcad | what version of autoconf and m4 do you have there? |
| 14:59.27 | d_rossberg | autoconf 2.65, m4 1.4.13 |
| 15:00.33 | brlcad | ah, I see the difference .. it shouldn't be subst'ing those variables |
| 15:02.20 | CIA-85 | BRL-CAD: 03brlcad * r37978 10/rt^3/trunk/configure.ac: should not be AC_SUBSTituting the m4 variables. would have to set them to a shell variable, and we don't need them separated like this anyways. |
| 15:04.47 | brlcad | how that ever survived, I do not know |
| 15:05.18 | brlcad | I don't think the autotools build path has been kept up to date |
| 15:09.17 | d_rossberg | probable, now i got "configure.ac:733: required file `src/libNetwork/Makefile.in' not found" |
| 15:10.34 | d_rossberg | (will change it to */GS/*) |
| 15:10.35 | ``Erik | means automake needs to be re-run |
| 15:11.01 | ``Erik | (or the configure.ac and Makefile.am files are out of sync) |
| 15:11.12 | d_rossberg | that's it |
| 15:23.21 | brlcad | d_rossberg: my inclination is to remove the autotools build files from rt^3 and only have cmake there |
| 15:23.26 | brlcad | since it's new and smaller |
| 15:23.45 | d_rossberg | sorry, i was in the wrong directory; this configure.ac etc. belongs to the non-cmake build :-[ |
| 15:23.46 | brlcad | but do want to make sure that cmake is doing most of the things that autotools is performing now |
| 16:18.07 | CIA-85 | BRL-CAD: 03erikgreenwald * r37979 10/brlcad/trunk/src/librt/primitives/nmg/nmg_tri_mc.c: pewpewpew! |
| 16:19.04 | CIA-85 | BRL-CAD: 03erikgreenwald * r37980 10/brlcad/trunk/src/conv/g-egg.c: add ability to use marching cubes algorithm for tesselation |
| 16:19.38 | d-lo | 'pewpewpew' lol |
| 16:19.48 | d-lo | s/right/write/g |
| 16:20.15 | d-lo | oopsie, wrong window |
| 16:25.18 | ``Erik | dang, no egg-g converter, so I can't see the results yet heh |
| 16:25.34 | d-lo | get coding! |
| 16:25.49 | ``Erik | "dear heldpesk: pleased to be installing game engine on my computer, kthxbai!" |
| 16:26.14 | d-lo | well, you do have the power..... |
| 16:26.24 | ``Erik | by the power of greyskull! |
| 16:32.16 | d_rossberg | it's a little bit pity that it wasn't BRL-CAD's version number, so i have to do some more work ... |
| 16:35.00 | CIA-85 | BRL-CAD: 03erikgreenwald * r37981 10/brlcad/trunk/src/conv/g-egg.c: remove some debugging statements |
| 16:41.09 | CIA-85 | BRL-CAD: 03davidloman * r37982 10/rt^3/trunk/src/GS/netMsg/: Drop empty dir left over from previous refactor. |
| 16:47.34 | CIA-85 | BRL-CAD: 03davidloman * r37983 10/rt^3/trunk/ (2 files in 2 dirs): Added a convenience method to NetMsg for quickly sending opcode only message to the remotehost. |
| 16:57.08 | CIA-85 | BRL-CAD: 03davidloman * r37984 10/rt^3/trunk/ (2 files in 2 dirs): Added the ability to peek at the top of a NetMsgFactory's Message queue. Needed for checking the opcode of a NetMsg prior to a Portal offering it up to the 'user'. |
| 17:08.43 | d-lo | question for the pros: Which is better to use for a list of, say, error codes? Enums or MACROs? |
| 17:14.36 | CIA-85 | BRL-CAD: 03davidloman * r37985 10/rt^3/trunk/ (3 files in 3 dirs): Wire in hooks to NetPortal for remote disconnection. |
| 17:34.32 | brlcad | depends and specifically for error codes the difference is nominal, but if they logically all group together, enums naturally group them |
| 17:39.59 | brlcad | they have the added benefit of being typedefable so the compiler can test if you're using valid values |
| 17:41.42 | d-lo | awesome. Spanks! |
| 17:44.39 | brlcad | the problem is when the grouping is not well thought out and you end up with error-prone hackeries like having a "last" element so you can blindly iterate through a range of potential values, or making some values be bit-maskable and others not, having enums and relying on specific values elsewhere, etc |
| 18:29.56 | ``Erik | macros are nifty if you want to set bits as error codes (to compact many into a word), enums are good for linear sequences (y'know... enumerations) :D |
| 18:30.46 | ``Erik | "these 3 errors happened" is easy in macro land, requires explicit permutations in enum land :D |
| 18:32.08 | ``Erik | pouts cuz green turtle was out of the beer he usually gets, but impressed with the (overpriced) selection of drafts... had a newscastle O.o :D |
| 18:38.13 | CIA-85 | BRL-CAD: 03erikgreenwald * r37986 10/brlcad/trunk/src/libgcv/ (Makefile.am region_end_mc.c region_start_mc.c): move start back to end... |
| 18:42.17 | CIA-85 | BRL-CAD: 03erikgreenwald * r37987 10/brlcad/trunk/src/libgcv/region_end_mc.c: type fixes |
| 18:43.49 | brlcad | ../../../misc/ylwrap ../../../src/libgcv/obj/obj_grammar.yy y.tab.c obj_grammar.cc y.tab.h obj_grammar.h y.output obj_grammar.output -- bison -y -d -p obj_parser_ |
| 18:43.52 | brlcad | make[2]: ../../../misc/ylwrap: Command not found |
| 18:45.39 | brlcad | the custom parser/lexer options being used in gcv are going to be rather error-prone or specific to a particular implementation if they're not left to libtool |
| 18:47.02 | brlcad | and WITH_MODERN_PARSERS seems wholly unnecessary (and will be eventually misleading) .. parsers should be on or off until we get to the point that they're required for some critical piece of functionality |
| 18:53.21 | starseeker | brlcad: I'm working on back-porting the ll and yy code |
| 19:03.43 | CIA-85 | BRL-CAD: 03erikgreenwald * r37988 10/brlcad/trunk/src/libgcv/region_end_mc.c: for great pewpewpew! |
| 19:04.13 | CIA-85 | BRL-CAD: 03erikgreenwald * r37989 10/brlcad/trunk/src/conv/g-egg.c: use region end func from libgcv instead of a local one |
| 19:17.23 | d-lo | that libgcv ``Erik is working sure sounds exciting! :P |
| 19:25.44 | starseeker | space invaders - the library! |
| 19:28.04 | ``Erik | well now, here's a single sph: http://brlcad.org/~erik/mc/sph2.png |
| 19:28.08 | ``Erik | I think that's enough to call it done! :D |
| 19:29.19 | louipc | whoa what's that |
| 19:29.37 | ``Erik | a sphere! dur :D |
| 19:29.52 | starseeker | hehehe |
| 19:29.55 | louipc | is it like the half finished death star? |
| 19:30.26 | brlcad | well done! |
| 19:36.22 | starseeker | apparently, no one does run time testing of lex abilities |
| 19:56.11 | ``Erik | starts thinking he made the same mistake he did when doing the metaball conversion stuff O.o |
| 20:04.55 | *** join/#brlcad louipc (~louipc@archlinux/trusteduser/louipc) | |
| 20:51.43 | *** join/#brlcad stevegt_ (~stevegt@cislunar.TerraLuna.Org) | |
| 21:03.48 | *** join/#brlcad cosurgi (~cosurgi@atak.bl.pg.gda.pl) | |
| 21:05.42 | CIA-85 | BRL-CAD: 03bob1961 * r37990 10/brlcad/trunk/ (include/ged.h src/libged/grid.c): Expose the ged_snap_to_grid function. |
| 21:08.38 | brlcad | gah, starseeker ... did you actually inject the edit string as argv[0]?? |
| 21:08.51 | brlcad | in ged_red() .. maybe elsewhere too |
| 21:10.57 | brlcad | that's really awful.. api-wise they're no longer consistent (or usable without custom hacking) and break the pattern |
| 21:11.10 | CIA-85 | BRL-CAD: 03bob1961 * r37991 10/brlcad/trunk/src/tclscripts/lib/Ged.tcl: Added the snap_view, pane_snap_view and pane_screen2view methods. |
| 21:12.24 | starseeker | tries to recall... |
| 21:12.44 | brlcad | even as a temp measure.. that's no good |
| 21:13.09 | brlcad | it should be an option |
| 21:13.20 | brlcad | -e editstring or whatever |
| 21:14.27 | brlcad | I get a memory violation on 64bit linux if I run red with EDITOR set, there's a bug somehwere |
| 21:15.13 | starseeker | ok. let me disable the obj files so that part of the build isn't busted and I'll dig into it |
| 21:17.11 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 21:17.44 | *** join/#brlcad cosurgi (~cosurgi@atak.bl.pg.gda.pl) | |
| 21:20.40 | CIA-85 | BRL-CAD: 03starseeker * r37992 10/brlcad/trunk/ (7 files in 3 dirs): Disable libgcv obj build - need to come at it differently, with functionality testing of lex and yacc tools in configure.ac. |
| 21:29.11 | CIA-85 | BRL-CAD: 03starseeker * r37993 10/brlcad/trunk/m4/ (Makefile.am ax_compare_version.m4): If doing functionality based testing, won't need version number comparison. |
| 21:33.20 | ``Erik | ah ha, progress... sorta |
| 21:34.17 | *** join/#brlcad Ralith (~ralith@69.90.48.97) | |
| 21:39.29 | ``Erik | a /cl |
| 22:19.05 | CIA-85 | BRL-CAD: 03brlcad * r37994 10/brlcad/trunk/src/libgcv/CMakeLists.txt: no more 'static' files |
| 22:52.03 | CIA-85 | BRL-CAD: 03bob1961 * r37995 10/brlcad/trunk/src/libtclcad/ged_obj.c: Added the go_snap_view function. This function takes view x,y and snaps it to the grid if snapping is turned on. It then returns the possibly altered values. |
| 22:54.51 | CIA-85 | BRL-CAD: 03bob1961 * r37996 10/brlcad/trunk/src/tclscripts/archer/Archer.tcl: Updated the writePreferencesBody method to write out settings for the grid. Updated the handleObjCenter and endObjTranslate methods to snap the view points to the grid. |
| 23:26.06 | CIA-85 | BRL-CAD: 03starseeker * r37997 10/brlcad/trunk/src/mged/utility1.c: Might want to allocate enough memory for the new argv rays for the editor commands... |
| 23:37.31 | CIA-85 | BRL-CAD: 03brlcad * r37998 10/brlcad/trunk/src/libged/ (ged_private.h put_comb.c red.c): remove the unnecessary _ged_tmpcomb_init global... there are way too many it-will-get-fixed-later globals in libged! |
