| 00:57.33 | *** join/#brlcad Aeamus (~Enigma@unaffiliated/r0b0t1) |
| 02:15.32 | *** join/#brlcad crazy_imp (~mj@a89-182-209-220.net-htp.de) |
| 02:41.01 | ``Erik | candy mountain, charlie |
| 03:24.15 | brlcad | starseeker: hehe.. "man 2 wait" |
| 03:26.36 | starseeker | ding nabbit - ok, what's the Windows version? |
| 03:28.07 | starseeker | google isn't very helpful on this |
| 03:29.33 | CIA-48 | BRL-CAD: 03starseeker * r41787 10/brlcad/branches/cmake/src/vdeck/vdeck.c: OK, that's not the wait functionality being assumed on Unix - what's the equalivent call on Windows? |
| 03:30.56 | brlcad | wait() is related to process management |
| 03:31.26 | brlcad | there isn't a 1-1 mapping for most of the process management calls, just like there's no fork() |
| 03:31.48 | starseeker | growl |
| 03:32.13 | starseeker | vdeck looks ancient - is it still used? |
| 03:32.14 | brlcad | but like with fork(), there's a way to create a new process (CreateProcess()), so there should be something similar like WaitForProcess() but undoubtedly not just a drop-in replacement |
| 03:32.22 | brlcad | yeah, it's used |
| 03:32.54 | brlcad | I was going to deprecate it a few months back and that month some conversation came up with someone using the vdeck tools |
| 03:33.18 | starseeker | O.o com-geom card images? Wow |
| 03:34.54 | starseeker | growl... this can't be the first time this has come up - perhaps we should do a bu_wait and friends to wrap all of this? |
| 03:35.55 | starseeker | actually, it's worse - this sucker also needs /bin/sh |
| 03:38.08 | starseeker | brlcad: what about this as a compromise - it looks like vdeck needs this stuff only when it exercises it's ability to run a unix shell command |
| 03:38.43 | brlcad | that "feature" is basically a security hole, so I wouldn't shed a tear to see it removed |
| 03:38.53 | starseeker | that's obviously a no-go on Windows anyway - what about wrapping the code defining the shell stuff in ifdef HAVE_SYS_WAIT_H? |
| 03:39.24 | starseeker | ah, better yet |
| 03:39.38 | brlcad | it's possible to kick out to a shell on windows, could be made to work |
| 03:39.42 | brlcad | but it's a dumb feature |
| 03:39.42 | starseeker | can I just yank it or do we need deprecation? |
| 03:40.01 | brlcad | yank it from trunk |
| 03:40.06 | starseeker | righto |
| 03:40.12 | brlcad | security issues don't follow deprecation |
| 03:41.28 | brlcad | akin to fixing a bug |
| 03:46.12 | CIA-48 | BRL-CAD: 03starseeker * r41788 10/brlcad/trunk/ (NEWS src/vdeck/vdeck.c): Remove the 'execute shell command' misfeature from vdeck. |
| 03:47.15 | brlcad | past tense |
| 03:47.28 | starseeker | er, yeah |
| 03:48.09 | brlcad | maybe: "removed shell execution !command from vdeck" |
| 03:48.12 | CIA-48 | BRL-CAD: 03starseeker * r41789 10/brlcad/trunk/NEWS: remove -> removed |
| 03:49.10 | brlcad | the one nice thing about that vdeck implementation was that it was close to what you'd need for a bu interface for running commands |
| 03:49.52 | brlcad | just missing the windows logic, which would have actually been simpler than the unix code |
| 03:50.22 | brlcad | because you'd just CreateProcess() in the current process's context and it'd automatically block until return |
| 03:50.51 | starseeker | what does nirt do? |
| 03:51.11 | starseeker | or does our command line nirt work on Windows... |
| 03:52.40 | CIA-48 | BRL-CAD: 03starseeker * r41790 10/brlcad/trunk/NEWS: call out the command that was removed |
| 03:53.32 | starseeker | ``Erik: what do you advise on the adrt stuff in Windows? wrap it with WIN32 conditionals until the new stuff is ready for trunk merge? |
| 03:54.43 | starseeker | ah, nevermind, nirt doesn't need to fork processes |
| 03:54.45 | starseeker | silly me |
| 03:54.54 | brlcad | most of the adrt stuff in that build log looked pretty easy to fix |
| 10:47.52 | *** join/#brlcad mafm (~mafm@253.Red-83-55-205.dynamicIP.rima-tde.net) |
| 11:48.23 | DaveLo | Mernin |
| 12:17.34 | *** join/#brlcad Stattrav (~suryajith@unaffiliated/stattrav) |
| 12:18.35 | *** join/#brlcad archivist_emc (~archivist@217.34.113.62) |
| 12:20.19 | DaveLo | /* |
| 12:20.36 | DaveLo | warms up the commit machine. |
| 12:21.03 | CIA-48 | BRL-CAD: 03davidloman * r41791 10/rt^3/trunk/src/GS/: Modified SVN:IGNORE |
| 12:26.57 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) |
| 13:01.38 | CIA-48 | BRL-CAD: 03davidloman * r41792 10/rt^3/trunk/src/ (8 files in 3 dirs): Insert sqlite 3.7.4 into src/other. Going to evaluate if sqlite is a good solution for the metadata processing we need for the Repository searching/filtering functionality |
| 13:04.35 | ``Erik | what's the behavior of sqlite3 with multiple writer and reader processes? |
| 13:33.56 | DaveLo | ``Erik: the documentation *says* that multiple instances of SQLite can perform IO on the same DB file. |
| 13:34.04 | DaveLo | but I'll believe it when I see it :) |
| 13:39.27 | CIA-48 | BRL-CAD: 03starseeker * r41793 10/brlcad/branches/cmake/ (8 files in 6 dirs): Sync CMake to trunk r41972 |
| 15:47.07 | ``Erik | so yeah, candy mountain, charlie |
| 15:54.33 | CIA-48 | BRL-CAD: 03starseeker * r41794 10/brlcad/branches/cmake/src/conv/dem-g.c: Start cleaning up dem-g - first, fix the variable size array declarations. This seems to work in testing on the Mac |
| 16:10.13 | brlcad | no clean up on trunk first? |
| 16:11.14 | brlcad | sqlite is acid compliant, it all works as you'd expect |
| 16:15.53 | brlcad | sqlite's biggest limitation is that it is basically typeless |
| 16:16.26 | brlcad | so any type checking has to happen in the application on both reads and writes from the database |
| 17:44.44 | *** join/#brlcad cosurgi (~cosurgi@atak.bl.pg.gda.pl) |
| 18:51.42 | CIA-48 | BRL-CAD: 03davidloman * r41795 10/rt^3/trunk/src/other/sqlite_3_7_4/ (SQLiteAndGSNotes.txt repo.db): Check in a notes file and a sample SQLite db file. |
| 18:53.31 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 19:29.06 | CIA-48 | BRL-CAD: 03davidloman * r41796 10/rt^3/trunk/ (4 files in 2 dirs): Move sqlite3.h to public include dir |
| 19:36.29 | DaveLo | Well, happy holidays all! I'm outtie |
| 20:26.26 | starseeker | brlcad: er, yeah I guess that woudl be good wouldn't it - tend to forget about trunk when doing Windows fixes :-P |
| 21:25.10 | *** join/#brlcad mafm_ (~mafm@253.Red-83-55-205.dynamicIP.rima-tde.net) |
| 21:30.58 | CIA-48 | BRL-CAD: 03bob1961 * r41797 10/brlcad/trunk/src/tclscripts/lib/TkTable.tcl: Added more windows and emacs style bindings for cell editing and traversal. |
| 22:56.56 | *** join/#brlcad Stattrav (~suryajith@static.71.252.63.178.clients.your-server.de) |
| 23:18.07 | *** join/#brlcad Stattrav (~suryajith@static.71.252.63.178.clients.your-server.de) |
| 23:23.53 | ``Erik | http://www.daggerpress.com/2010/12/19/harford-county%E2%80%99s-food-facility-inspection-violations-for-november-2010/ |
| 23:35.23 | *** join/#brlcad Stattrav (~suryajith@static.71.252.63.178.clients.your-server.de) |
| 23:37.06 | starseeker | ``Erik: hmm. sounds like a bunch of refrigerators need fixing ;-) |
| 23:38.39 | starseeker | bit more concerned by the Basta sanitary procedures one - ew |
| 23:59.19 | starseeker | Anybody here have any experience using dynamic dns to make it possible to log into one's machine from a remote computer? |