| 00:07.55 | CIA-48 | BRL-CAD: 03starseeker * r41714 10/brlcad/branches/cmake/src/other/ (46 files in 4 dirs): |
| 00:07.55 | CIA-48 | BRL-CAD: CMake build of NetBSD's m4, the smallest m4 I can find that can bootstrap flex. |
| 00:07.56 | CIA-48 | BRL-CAD: Seems to work, will probably need a few more tweaks. Testing this combination |
| 00:07.56 | CIA-48 | BRL-CAD: of byacc and m4 triggered a few flex test failures, have emailed byacc dev to |
| 00:07.57 | CIA-48 | BRL-CAD: see if they are significant. flex will be a bit more of a challenge. |
| 01:34.39 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 02:15.51 | *** join/#brlcad crazy_imp (~mj@a89-182-207-12.net-htp.de) |
| 04:54.37 | CIA-48 | BRL-CAD: 03starseeker * r41715 10/brlcad/branches/cmake/src/other/flex/ (50 files): |
| 04:54.38 | CIA-48 | BRL-CAD: Upload the beginnings of a minimal subset of flex. This is far less ready than |
| 04:54.38 | CIA-48 | BRL-CAD: byacc and m4 - the conf.in file will require a fair number of tests be hooked |
| 04:54.39 | CIA-48 | BRL-CAD: up, the m4 binary location must be #defined in config.h (!?) and other goodies |
| 04:54.39 | CIA-48 | BRL-CAD: must be sorted out. |
| 04:57.03 | brlcad | you're really going for the motherload of branches, aren't ya? :) |
| 05:02.22 | starseeker | brlcad: heh |
| 05:02.41 | starseeker | yeah, kinda - I suppose I should be branching off of the cmake branch for this one |
| 05:05.26 | starseeker | it's not as impressive as it looks in the commit messages |
| 05:06.55 | starseeker | looks at the clock and decides he can't face another round of autoconf tests -> CMake tests right now |
| 05:08.09 | starseeker | anyway, next on the agenda is getting itcl/itk, tkhtml, and friends building on Windows |
| 05:08.39 | starseeker | this was just weekend wandering :-P |
| 05:26.42 | brlcad | more thinking that merging cmake back into trunk is going to be a royal bear to review |
| 05:27.08 | brlcad | good stuff, great stuff.. but it's already a huge branch with major changes :) |
| 05:27.31 | brlcad | the bigger it gets, the riskier and costly it is going to be to merge |
| 05:29.38 | brlcad | already estimate that it's going to take someone (and that someone shouldn't be you) at least a couple weeks to review all the changes, compare, validate, and merge |
| 05:31.13 | brlcad | adding a parser lexer probably doesn't add much time or risk to the bottom line, but it's definitely a huge entropy increase |
| 05:38.20 | brlcad | considering how much code is being modified before it's merged (and the reason to try and minimize it) .. |
| 05:38.30 | brlcad | consider (from a really rough big-picture industry average viewpoint) that you will have introduced a bug for every 100 lines of code modified or added |
| 05:41.40 | brlcad | and that's unfortunately being rather optimistic if steve mcconnell's statistics are to be believed |
| 09:05.01 | *** join/#brlcad Ralith (~ralith@216.162.199.202) |
| 09:47.04 | *** join/#brlcad merzo (~merzo@209-53-94-178.pool.ukrtel.net) |
| 10:19.39 | *** join/#brlcad mafm_ (~mafm@41.Red-83-35-148.dynamicIP.rima-tde.net) |
| 11:28.51 | DaveLo | Mernin all |
| 11:34.53 | DaveLo | brlcad: entropy as in time 'not available for useful work'? |
| 11:58.20 | *** join/#brlcad kanzure (~kanzure@131.252.130.248) |
| 12:27.58 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) |
| 13:12.58 | starseeker | brlcad: hmm. Well, I have no problems reverting the parser/lexer commits for now to keep things a bit more minimal - they're not essential |
| 13:13.30 | starseeker | or I'll just branch it off of cmake and then revert it, either way... |
| 13:31.57 | CIA-48 | BRL-CAD: 03davidloman * r41716 10/rt^3/trunk/src/libNet/Portal.cxx: Prepend '0x' on hex debug prints cause it makes it easier to diff them from base 10 numbers. |
| 13:42.19 | CIA-48 | BRL-CAD: 03davidloman * r41717 10/rt^3/trunk/src/libNet/Portal.cxx: Since pkg moves data off the Socket's recv buffer during a 'send' action without calling the callback method, add in a call and check in Portal::send(). This seems to clear up some of the 'hangup' we are experiencing. |
| 14:30.16 | *** join/#brlcad Zaebos (~irc@pd95b7f5e.dip0.t-ipconnect.de) |
| 14:39.20 | CIA-48 | BRL-CAD: 03davidloman * r41718 10/rt^3/trunk/src/libNet/PortalManager.cxx: |
| 14:39.21 | CIA-48 | BRL-CAD: Ah ha! Found the bug that was causing the restart issues |
| 14:39.21 | CIA-48 | BRL-CAD: (login->logout->login). Turns out that the FD for the incoming connection was |
| 14:39.22 | CIA-48 | BRL-CAD: ONLY added to the master FD Set *if* the new FD was set as MAXFD. Unrolled the |
| 14:39.22 | CIA-48 | BRL-CAD: logic accordingly. |
| 15:11.07 | CIA-48 | BRL-CAD: 03davidloman * r41719 10/rt^3/trunk/src/libNet/PortalManager.cxx: Debug printing disabling/cleanup. Made selector loop only try to read on returned FDs (instead of the hotwired solution of trying to read on all FDs each loop). This eliminated blocking on a call to pkg_suckin. |
| 16:14.21 | *** join/#brlcad mafm (~mafm@41.Red-83-35-148.dynamicIP.rima-tde.net) |
| 16:46.32 | brlcad | DaveLo: exactly, yes .. the more code you have, the more overhead time you have to spend just to maintain that code. that's time and effort that's not generally productively useful (but is necessary) |
| 16:53.38 | CIA-48 | BRL-CAD: 03starseeker * r41720 10/brlcad/branches/cmake/src/other/ (CMakeLists.txt flex/ m4/): Revert flex/m4/byacc commit - deal with this later, have other changes more urgently needed for CMake. |
| 16:54.50 | CIA-48 | BRL-CAD: 03starseeker * r41721 10/brlcad/branches/cmake/ (3 files in 3 dirs): Whoops, get byacc too |
| 18:09.00 | CIA-48 | BRL-CAD: 03starseeker * r41722 10/brlcad/branches/cmake/src/other/ (3 files in 3 dirs): Not terribly pretty, but get itcl/itk building with VC++ 2010 |
| 18:31.42 | CIA-48 | BRL-CAD: 03starseeker * r41723 10/brlcad/branches/cmake/src/other/ (CMakeLists.txt tkhtml/CMakeLists.txt tktable/CMakeLists.txt): Get tkhtml and tktable working (thank you Bob for the tktable logic in our msvc files.) |
| 19:04.04 | DaveLo | brlcad: So, logically speaking, there is a Maximum amount of code that a given group of individuals can reach? |
| 19:05.58 | DaveLo | I suppose that assumes that the code is being properly maintained, heh. |
| 19:07.48 | CIA-48 | BRL-CAD: 03davidloman * r41724 10/rt^3/trunk/src/GS/AccountManager.cxx: Make AccountManager print accountID as well as account uname to log upon proper authentication. |
| 19:13.22 | CIA-48 | BRL-CAD: 03davidloman * r41725 10/rt^3/trunk/ (include/AccountManager.h src/GS/AccountManager.cxx): Trying to pass the value of '-1' back to a unsigned int just doesn't work well. Made the quint32 values qint32's. |
| 19:24.34 | ``Erik | http://1.bp.blogspot.com/_a7jkcMVp5Vg/TQ2gXJIRVbI/AAAAAAAANuk/1eTuWCnPOmE/s1600/tinyhand.jpg |
| 19:28.39 | DaveLo | lol, that's nasty.... |
| 19:29.07 | ``Erik | it was sent to me, and I almost blew snot on my laptop when I opened it and read it |
| 19:32.33 | CIA-48 | BRL-CAD: 03davidloman * r41726 10/rt^3/trunk/src/GS/GSClient.cxx: Make GSClient able to handle FailureMsg's and setup proper routing for this. |
| 19:44.32 | CIA-48 | BRL-CAD: 03davidloman * r41727 10/rt^3/trunk/src/GS/AccountManager.cxx: Change logic since '0' is a valid Account ID. |
| 19:50.39 | CIA-48 | BRL-CAD: 03davidloman * r41728 10/rt^3/trunk/src/GS/GSClient.cxx: Fix the cmd line client. Was converting captured command and all args to lower case, which was messing up things like account Authentication. Made it so that only the cmd is converted to lower case. |
| 19:51.22 | DaveLo | hates hates HATES work's vpn. Its dropping out everything the screen saver turns on. lol/grrrr |
| 20:21.30 | CIA-48 | BRL-CAD: 03davidloman * r41729 10/rt^3/trunk/src/libNet/netMsg/: Add *.backup to svn:ignore list. |
| 21:00.10 | CIA-48 | BRL-CAD: 03starseeker * r41730 10/brlcad/branches/cmake/include/config_win.h: Shouldn't need pstdint.h in config_win.h now |
| 21:10.23 | CIA-48 | BRL-CAD: 03davidloman * r41731 10/rt^3/trunk/ (6 files in 3 dirs): Implement Ping and Pong Msgs. Will be used for heartbeat/keepalive and network latency measuring. |
| 21:12.45 | CIA-48 | BRL-CAD: 03starseeker * r41732 10/brlcad/branches/cmake/CMakeLists.txt: Urk - that's right, no PARALLEL for Windows |
| 21:14.23 | CIA-48 | BRL-CAD: 03starseeker * r41733 10/brlcad/branches/cmake/src/other/tk/CMakeLists.txt: Hmm, Windows seems to want both tcl and tclstub - need to look into this, what do the nmake files do? |
| 21:27.50 | *** join/#brlcad merzo (~merzo@209-53-94-178.pool.ukrtel.net) |
| 21:45.35 | *** join/#brlcad PrezKennedy (MK@whitecalf.net) |
| 22:16.39 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) |
| 23:42.18 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
