[00:17:41] --- jblaine has left [00:51:08] --- Simon Wilkinson has become available [01:37:54] --- Jeffrey Altman has left: Replaced by new connection [01:37:55] --- Jeffrey Altman has become available [03:36:22] --- meffie has become available [04:26:40] --- steven.jenkins has left [04:28:46] --- steven.jenkins has become available [04:59:35] --- lars.malinowsky has left [06:11:55] --- lars.malinowsky has become available [06:15:41] --- jaltman/FrogsLeap has become available [06:28:59] --- jaltman/FrogsLeap has left: Replaced by new connection [06:29:05] --- jaltman/FrogsLeap has become available [06:29:17] --- jblaine has become available [06:39:53] --- jaltman/FrogsLeap has left: Disconnected [06:44:09] --- jaltman/FrogsLeap has become available [07:08:19] --- pod has left [07:18:16] --- jaltman/FrogsLeap has left: Replaced by new connection [07:18:17] --- jaltman/FrogsLeap has become available [07:18:32] --- Jeffrey Altman has left: Replaced by new connection [07:40:10] --- jaltman/FrogsLeap has left: Replaced by new connection [07:40:11] --- jaltman/FrogsLeap has become available [07:46:42] --- jblaine has left [07:49:01] --- jblaine has become available [07:49:55] --- deason/gmail has become available [07:53:46] ! [remote rejected] HEAD -> refs/changes/4897 (squash commits first) [07:53:58] I did (???) [07:54:09] "git commit --squash 2324911fa57e70f0345c2a9da6fc47eb30c1fa12" [07:54:46] maybe I squashed the wrong direction? [07:55:36] commit --squash does not squash, it just makes a commit that's ready tt be squashed with git rebase --autosquash. [08:00:21] * jblaine once again makes a total mess of his local git repo [08:28:35] --- lars.malinowsky has left [08:29:18] --- jaltman/FrogsLeap has left: Replaced by new connection [08:29:19] --- jaltman/FrogsLeap has become available [09:34:03] --- jaltman/FrogsLeap has left: Disconnected [09:34:10] --- jaltman/FrogsLeap has become available [10:11:08] --- jaltman/FrogsLeap has left: Replaced by new connection [10:11:09] --- jaltman/FrogsLeap has become available [10:19:57] Derrick: Thanks - I'll see if I can implement that (not familiar with the concept of a "rock", but I follow you) [10:21:38] --- jaltman/FrogsLeap has left: Replaced by new connection [10:21:39] --- jaltman/FrogsLeap has become available [10:22:44] Derrick: Should that struct go in xstat_common.c or in a new xstat_common.h [10:29:23] the "rock" is just a pointer to anything, so the called-back function can have some state information that the "framework" doesn't need to know anything about [10:30:05] they're common in languages that don't have closures, but they get called different things all the time I think; "context" may be a more familiar name [10:30:34] ah [10:32:55] turns out the extracted poller also has an xstat_fs_Init() flag arg that it builds up, specific to FS :| [10:33:26] initFlags |= XSTAT_FS_INITFLAG_DEBUGGING, etc [10:34:33] As the sands of time flow through the hourglass, so does the signature grow [10:35:32] ah, actually, a new int initFlags arg kills off int debug and int oneshot [10:42:46] * jblaine sighs at his C depth [10:51:04] --- jaltman/FrogsLeap has left: Disconnected [11:03:46] --- jaltman/FrogsLeap has become available [11:06:06] Derrick: Any idea where I might find a similar struct of func pointers, showing how to fill and interact with it? [11:20:32] --- jaltman/FrogsLeap has left: Replaced by new connection [11:20:37] --- jaltman/FrogsLeap has become available [11:21:12] ah, storeOps and fetchOps [11:24:20] the unix cm has fetchops and storeops for disk cache and memcache... I'm not sure how easy that is to follow, though; the bnode_ops in src/bozo may be simpler [11:30:33] --- jaltman/FrogsLeap has left: Disconnected [11:30:49] --- jaltman/FrogsLeap has become available [11:38:07] git: I've made changes on branch 'xstat'. When I switch to branch master, I see "M src/xstat/Makefile.in" and then looking at src/xstat/Makefile.in shows my changes from branch 'xstat' [11:38:11] what am I missing? [11:38:18] I would expect those changes to not be in master [11:45:00] you hadn't committed the changes to xstat/Makefile.in [11:45:22] yeah, I kinda didn't want to, but okay [11:45:31] and if you switch to a branch that's exactly the same as the one you're currently on (or it's an ancestor maybe?) it'll keep your working tree the same [11:45:38] got it [11:45:40] thanks [11:45:48] if you switch to completely different branch it won't let you switch at all [11:46:03] if you don't want to commit your changes even temporarily, you can "stash" them with "git stash" [11:46:11] ahhhh [11:55:43] so there's no quick way to say "show me the HEAD version of this file" that I can find, am I not seeing something? [11:56:16] I don't see a way to tell checkout to dump to a new filename, for instance [11:56:30] and can't see how to use git show, assuming that would do what I am saying [11:56:46] I'd rather not see a diff [11:58:03] I guess I could cp foo foo.save; git checkout HEAD -- foo; look at it, then mv foo.save foo [12:00:07] which is basically git stash; git checkout; retrieve stash... heh [12:01:02] I'm looking for git showmethecontents HEAD foo.c [12:18:12] the closest I can think of is 'git blame foo.c HEAD' [12:18:25] that will say where each line came from, but I guess it also includes the source [12:22:56] thanks [12:25:55] Man, I don't get git. I did a 'git checkout master; git branch -D xstat' and the uncommitted xstat changes still show in master [12:26:04] I suppose that's related to what you said earlier, but geez [12:29:11] I think you need to re-read http://wiki.openafs.org/AFSLore/GitDevelopers/ [12:29:28] and many other full tutorials [12:29:34] no disagreement here [12:30:09] if you want to create a new branch that tracks the 'master' in the upstream repository git checkout -b origin/master [12:30:55] 'master' is just a branch in your local repo. If you committed junk onto, checking it out again is not going to change that [12:31:09] so I most likely want to do that for each of my project branches [12:31:15] I hadn't committed junk onto it [12:31:21] git reset --hard got me sane [12:31:33] but I follow what you're saying [12:32:30] I had made a nontracking 'xstat' branch, done development there, and just wanted to see a version of the file from before my mods [12:32:32] and failed [12:39:51] is it bad form to init a var at declaration? numX = CountListItems(foolist_passed_in); [12:39:56] err int numX... [12:41:22] initializing from a constant is acceptable and in many cases required. initializing from a function is not ok [12:41:30] ok [12:41:51] In C++ its fine. In C, not so much [12:43:48] I wonder if anyone would cry if I ripped the ~30 "if (debug) { fprintf... }" blocks in xstat*test.c [12:44:09] I suspect so. [12:44:09] you mean the ones that were just put in? [12:45:27] I don [12:45:31] "if (debugging_on) " isn't new [12:45:33] I don't know about "just" [12:46:50] git log doesn't show anything substantial about debugging for ~9 years [12:47:13] meffie added something in 2010 for timeval stuff [12:47:59] I don't know if they need to stay as they are, but I don't think they can just.... go away [12:48:25] I have enough fish to fry really, just tossing it out there as they're annoying me [12:48:31] --- jaltman/FrogsLeap has left: Replaced by new connection [12:48:32] --- jaltman/FrogsLeap has become available [12:50:15] --- Russ has become available [12:50:41] --- jaltman/FrogsLeap has left: Disconnected [12:50:58] --- jaltman/FrogsLeap has become available [12:53:12] --- jaltman/FrogsLeap has left: Replaced by new connection [12:53:13] --- jaltman/FrogsLeap has become available [13:10:12] --- jaltman/FrogsLeap has left: Disconnected [13:10:29] --- jaltman/FrogsLeap has become available [13:19:27] Derrick: turns out storeOps stuff is easier translateable to this case -- the bnode stuff is funkier [13:22:39] To show a particular revision of a file: [13:23:22] git show : for example, git show openafs-stable-1_4_0:src/rx/rx.c [13:34:44] ahhh, excellent [13:35:40] --- jakllsch has become available [13:55:46] --- jaltman/FrogsLeap has left: Disconnected [13:58:19] --- jaltman/FrogsLeap has become available [14:02:07] --- jaltman/FrogsLeap has left: Disconnected [14:06:51] --- jaltman/FrogsLeap has become available [14:15:40] hah, it builds! [14:15:47] surely it will not run [14:16:32] nope [14:55:04] > I guess I could cp foo foo.save; git checkout HEAD -- [14:55:26] btw, you can git show a blob, to see old revisions of files. [14:58:58] you need the commit number and the file path; for example [14:59:03] git show 13aeb1dc:src/xstat/xstat_fs_test.c [15:01:08] oh, didnt read all the way to the bottom. what simon said. [15:05:33] jblaine: have you looked at collectd yet? seems like it would be useful for you. [15:51:28] --- deason/gmail has left [18:07:45] --- deason/gmail has become available [19:06:22] --- jaltman/FrogsLeap has left: Disconnected [20:13:33] --- jaltman has become available [20:36:48] --- jaltman has left: Disconnected [20:41:50] --- jaltman has become available [20:58:06] --- jaltman has left: Disconnected [21:06:35] --- jaltman has become available [22:14:45] --- jaltman has left: Disconnected [22:15:26] --- jaltman has become available [23:26:52] --- deason/gmail has left [23:47:03] --- jaltman has left: Replaced by new connection [23:47:04] --- jaltman has become available [23:49:17] --- jaltman has left: Replaced by new connection [23:49:18] --- jaltman has become available