March 23, 2015 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
March 23, 2015 [00:02:56] <finkVC> jwhowarth: experimental/jwhowarth/main/finkinfo gcc5.patch[00:02:58] <finkVC> add gcc5.patch[00:03:25] <finkVC> jwhowarth: experimental/jwhowarth/main/finkinfo gcc5.info[00:03:27] <finkVC> add gcc5.info[00:08:33] *** amacks has joined #fink[00:29:02] *** asmeurer has joined #fink[00:29:59] *** asmeurer has quit IRC[00:37:16] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/languages gcc48.info[00:37:18] <finkVC> prune C/R of gcc4x's not in 10.7 tree and switch to new package naming for gcc5/gcc6 releases[00:37:20] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/languages gcc49.info[00:37:22] <finkVC> prune C/R of gcc4x's not in 10.7 tree and switch to new package naming for gcc5/gcc6 releases[00:37:24] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/languages gcc47.info[00:37:26] <finkVC> prune C/R of gcc4x's not in 10.7 tree and switch to new package naming for gcc5/gcc6 releases[00:37:44] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/languages gcc46.info[00:37:45] <finkVC> prune C/R of gcc4x's not in 10.7 tree and switch to new package naming for gcc5/gcc6 releases[00:40:08] *** howarth has quit IRC[00:47:51] <Murr> crazy idea of the day: If I added a facility to build packages whose source is specified by a git repository + commit ID, would that have a chance of being accepted into the core?[00:48:22] <Murr> I'm working on the port of a project that rarely gets official releases[00:57:37] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/devel cloog-org.info[00:57:39] <finkVC> add isl2 to C/R[00:57:51] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/devel cloog-org2.info[00:57:53] <finkVC> add isl2 to C/R[01:10:44] *** Yakeson_chihiro has joined #fink[01:35:58] *** amacks has quit IRC[01:37:03] *** mischi has quit IRC[02:00:13] *** amacks has joined #fink[02:25:57] *** Craigwd2000 has joined #fink[03:22:14] *** asmeurer has joined #fink[03:31:24] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/languages gcc48.info[03:31:26] <finkVC> add missing #[04:02:40] *** asmeurer has quit IRC[04:05:18] <akh> Murr: We’ve talked about that. As long as whatever versioning exists is deterministic, _I’d_ accept it. :-)[04:05:33] <Murr> cool[04:06:08] <akh> later all[04:06:11] *** akh has quit IRC[04:55:28] <dmacks_away> Murr: deterministic *and monotonic* (maybe prefix the commit-ID with a datestamp?) Sorting MD5s is not gonna work:)[04:57:32] <dak180> dmacks_away / Murr: seems like something autorevision would be good at[05:56:26] *** amacks has quit IRC[05:58:42] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/devel cloog-org2.info[05:58:44] <finkVC> revert addition of isl2 to C/R[06:58:12] *** dak180 has quit IRC[07:02:21] *** dak180 has joined #fink[07:15:02] <finkVC> jwhowarth: experimental/jwhowarth/main/finkinfo gcc5.info[07:15:04] <finkVC> lower revision[09:12:17] *** mischi has joined #fink[09:17:12] *** mischi has quit IRC[10:15:06] *** finkVC has quit IRC[10:55:29] <Murr> Oh, interesting idea with the autorevision, dak180. You're the original author of this, aren't you?[10:58:45] <Murr> but I think a timestamp will be preferable. Relying on VCS_NUM for monotonicity seems a bit dangerous to me[11:00:07] <Murr> e.g. in a project, I make a version for rev A on trunk. For a while, there's experimental mac support on a branch, so I make a version for rev B on that branch. That support gets squash merged into trunk, so it might well be that trunk has FEWER commits than the branch used to have[11:53:20] *** RangerRick has quit IRC[11:53:38] *** RangerRick has joined #fink[12:25:24] *** mischi has joined #fink[12:54:59] *** amacks has joined #fink[13:32:40] *** amacks has quit IRC[14:30:40] *** TheSin has quit IRC[14:48:12] *** Yakeson_chihiro has quit IRC[15:15:24] *** TheSin has joined #fink[15:15:24] <Chanbot> All hail my mighty and rightful ruler![15:25:19] *** Yakeson_chihiro has joined #fink[15:47:16] *** howarth has joined #fink[15:47:30] <howarth> has anyone ever seen this[15:47:56] <howarth> I reproducibly have issues with Xquartz not autolaunching from the Terminal window[15:48:05] <howarth> the trigger seems to be as follows[15:48:42] <howarth> if I switch my boot volume from 10.10 to 10.8, and reboot and then switch back to 10.10[15:49:19] <howarth> the only thing that remotely makes sense is the fact that Spotlight tries to index the 10.10 volume[15:49:37] <howarth> and when I reboot back I see a bunch of mdflagwriter processes in 'top'[15:50:23] <howarth> unfortunately it is pretty much impossible to cleanly keep the indexing off of the other drive without totally disabling it[15:50:46] <howarth> the Privacy settings for Spotlight appear to be tagged to the drive in question[15:51:11] <howarth> so if you disable indexing of the 10.10 volume under 10.8, it will also be disabled for itself under 10.10[15:51:43] <howarth> likewise if you disable the indexing of the 10.8 volume under 10.10, it will be disabled for itself under 10.8[15:52:44] <howarth> I suspect that they had to store it locally on the volume to deal with the possibility of it being renamed[16:32:24] *** howarth has quit IRC[17:14:40] *** finkVC has joined #fink[17:14:40] <finkVC> dmacks: dists/10.7/stable/main/finkinfo/sci geant4.9.info[17:14:41] <finkVC> dist-restrict to match gcc47 dep, per maintainer[17:41:20] *** howarth has joined #fink[17:42:19] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/x11 pmw2-py.patch[17:42:21] <finkVC> add upstream demos and tests fixes[17:42:27] <finkVC> jwhowarth: dists/10.7/stable/main/finkinfo/x11 pmw2-py.info[17:42:29] <finkVC> add upstream demos and tests fixes[17:55:49] *** mischi has quit IRC[18:26:13] *** akh has joined #fink[18:42:52] *** howarth has quit IRC[18:52:05] *** jack has quit IRC[18:53:21] *** jack has joined #fink[18:53:29] *** jack has joined #fink[19:00:25] *** asmeurer has joined #fink[19:10:29] *** asmeurer has quit IRC[19:21:57] *** jack has quit IRC[19:22:49] *** jack has joined #fink[19:23:01] *** jack is now known as Guest12419[19:24:05] *** howarth has joined #fink[19:25:03] *** Guest12419 is now known as jack[19:25:03] *** jack has joined #fink[19:41:35] *** Yakeson_chihiro has quit IRC[19:52:00] <howarth> akh[20:21:31] <akh> ?[20:25:19] <howarth> never mind, you answered the user on list[20:25:27] <howarth> about the gmp in local crap[20:26:25] <akh> Ok[20:36:56] <howarth> You only under your older OS's from within VMware or such, rigjht?[20:37:14] <howarth> This Xquartz autolaunching glitch is a major pain in the ass[20:37:52] <akh> Agreed, yeah.[20:38:11] <howarth> have you seen it on your machine as well?[20:38:16] <akh> Yup[20:38:29] <howarth> what is your analysis?[20:38:44] <akh> Nothing to add. :-)[20:38:50] <howarth> the best I can puzzle out is that it must be something related to Spotlight indexing[20:39:27] <howarth> reinstalling Xquartz doesn't immediately solve it[20:39:41] <howarth> and it just starts working again after awhile[20:40:01] <howarth> do you never see it until you switch OS releases, right?[20:40:19] <howarth> i.e. as long as you keep rebooting under 10.10, it doesn't pop up[20:40:35] <howarth> I am trying to convince Jeremy to try to look at it[20:40:54] <akh> My recollection is that it’s 10.10-only, but I can take a look at that.[20:41:13] <howarth> but the trigger is aways switching OS releases no?[20:41:33] <howarth> or have you seen it come and go while you stayed completely on 10.10[20:42:00] <akh> I believe I’ve seen it come and go even staying on 10.10.[20:42:06] <howarth> ah[20:42:31] *** finkVC has quit IRC[20:42:33] <howarth> do you save application states across logouts?[20:43:03] <akh> I try not to, but Software Update sometimes forces the issue.[20:44:00] <howarth> so you can't really say that is tied to saved application states[20:44:17] <howarth> hopefully Apple will treat 10.11 like iOS 9[20:44:31] <howarth> and focus on performance and bugs again rather than features[20:44:57] <akh> Yeah; I’ll keep an eye out for that.[20:45:22] <howarth> the problem is that the issue doesn't seem to have a handle to grab onto[20:45:47] <howarth> other than the hang when you execute an X11 binary in the terminal window[20:46:22] <howarth> too many Finder tricks involved with Xquartz imho[20:47:00] <howarth> like the fact that the application in /Applications is really named Xquartz.app at the filesystem level but shows up in the Finder as X11[20:48:06] <howarth> Jeremy says[20:48:08] <howarth> We don't use Spotlight (at least not directly). The only place we have any interaction with Spotlight is when /opt/X11/bin/Xquartz launches /Applications/Utilities/XQuartz.app. It does so via LaunchServices and then falls back on direct execution if that fails. LaunchServices uses Spotlight to find the bundle in case it was relocated.[20:48:08] <howarth> Please take a sysdiagnose 'sudo sysdiagnose -q' and also get me the launchctl output when in this blocked state.[20:49:10] <howarth> any idea how I am supposed to get launchctl output[20:51:05] <akh> No clue there.[20:52:27] <howarth> http://erikslab.com/2011/02/04/logging-with-launchd/[21:16:54] <howarth> akh, you are seeing these hands on straight 10.10.2 right (not a 10.10.3 seed)?[21:22:05] <howarth> hmm[21:22:05] <howarth> http://obscuredclarity.blogspot.com/2011/02/debugging-launchd-configuration-on-mac.html[21:59:41] <akh> It was on a release OS, yeah.[22:08:39] *** howarth has quit IRC[22:14:50] *** amacks has joined #fink[22:25:21] *** howarth has joined #fink[22:25:37] <howarth> I think I may have found the problem[22:26:15] <howarth> I had installed the MacPorts xinit package while working up an xvfb-run package for MacPorts[22:26:27] <howarth> it appears this installs a plist out of tree[22:58:01] *** azstefano has quit IRC[23:11:29] *** jack has quit IRC[23:12:29] *** jack has joined #fink[23:12:52] *** jack is now known as Guest35350[23:32:15] *** TheSin has quit IRC[23:53:29] *** azstefano has joined #fink