Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-05-14

| Channels | #git index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 VladGh joined #git
00:08 fstd_ joined #git
00:08 dirtyroshi joined #git
00:12 bronson joined #git
00:14 m0viefreak joined #git
00:15 dreiss joined #git
00:17 nkouevda joined #git
00:20 Cabanossi joined #git
00:23 tarkus joined #git
00:31 rivarun joined #git
00:31 causasui joined #git
00:31 causasui joined #git
00:37 miha_S7_ joined #git
00:38 Rodya_ joined #git
00:41 Masber joined #git
00:41 tourdownunder joined #git
00:43 juvenal joined #git
00:43 jameser joined #git
00:44 hexa- joined #git
00:48 jameser joined #git
00:51 m0viefreak joined #git
00:51 m0viefreak joined #git
00:52 m0viefreak joined #git
00:52 thebope joined #git
00:53 Rodya_ joined #git
00:54 Geminizer joined #git
00:55 m0viefreak joined #git
00:55 om_henners joined #git
00:57 patarr joined #git
00:58 monokrome joined #git
01:02 Zialus joined #git
01:05 Cabanossi joined #git
01:06 kdfjadflkj joined #git
01:06 kdfjadflkj left #git
01:06 IanLiu joined #git
01:08 SwiftMatt joined #git
01:11 blackwind_123 joined #git
01:14 finalbeta joined #git
01:14 rivarun joined #git
01:16 jon-mac joined #git
01:16 cebula joined #git
01:18 red82 joined #git
01:23 MarioBranco joined #git
01:26 MarioBranco joined #git
01:26 Gsham joined #git
01:27 MarioBranco joined #git
01:31 sunri5e joined #git
01:33 MarioBranco joined #git
01:35 Cabanossi joined #git
01:36 MarioBranco joined #git
01:38 rivarun joined #git
01:39 dsdeiz joined #git
01:39 dsdeiz joined #git
01:40 Geminizer joined #git
01:41 rivarun joined #git
01:47 Es0teric joined #git
01:50 ilbot3 joined #git
01:50 Topic for #git is now Welcome to #git, the place for git help. Just git help, apparently. | Public logs at https://gitirc.eu/log | First visit? Read: https://gitirc.eu | Current stable version: 2.13.0 | Getting "cannot send to channel"? /msg gitinfo .voice | This is serious business. https://madeitwor.se/git-jokes
01:50 transhuman can someone give me an explanation of git-checkout-index -a in laymen terms? I am not sure I understand this command
01:50 transhuman specifically what is meant by index?
01:51 transhuman thanks in advance
01:51 rewt index is the staging area; before you commit files, you add them to the index, and when you commit, your commit will only have the changes that are in the index
01:53 Gurkenglas joined #git
01:56 m0viefreak joined #git
01:58 d^sh joined #git
02:00 aw1 joined #git
02:00 ertesx joined #git
02:07 miha_S7_ joined #git
02:09 lagothri1 joined #git
02:11 durham joined #git
02:13 bronson joined #git
02:17 meLon joined #git
02:20 Vortex35 joined #git
02:21 transhuman thanks rewt, in that case I dont think this script does what I want it to ...I would like to export at particular commit level ....https://github.com/dasch/git-export/blob/master/git-export
02:23 transhuman sorry I mean at a particular tag
02:25 ojdo joined #git
02:28 AlexKvazos joined #git
02:32 hexagoxel joined #git
02:32 durham joined #git
02:32 ayogi joined #git
02:33 valhala_22 joined #git
02:34 m0viefreak joined #git
02:41 hfp joined #git
02:41 hfp_work joined #git
02:42 transhuman is this the best way to deal with exporting the differences between two tags (probably multiple commits) http://stackoverflow.com/questions/9078820/how-to-create-patch-between-two-tags-with-multiple-commits-between-them
02:43 sz0 joined #git
02:44 m0viefreak joined #git
02:50 diogenese joined #git
02:57 aw1 joined #git
02:58 chachasmooth joined #git
02:58 SwiftMatt joined #git
02:58 patarr joined #git
02:59 jimi_ joined #git
03:10 FuzzyWuz_ joined #git
03:12 efco joined #git
03:21 dasidiom joined #git
03:25 theiostream joined #git
03:33 Cabanossi joined #git
03:34 tourdownunder joined #git
03:36 menip joined #git
03:38 miha_S7_ joined #git
03:41 raijin joined #git
03:54 Emperor_Earth joined #git
03:54 somiaj joined #git
03:57 somiaj I'm looking into using git hooks to help deploy. I have a lot of .tex files that get built into .pdf, and I'd like to include this in the deploy. Is there a way to get the files that were updated in a git --work-tree= --git-dir= checkout -f?
03:58 somiaj goal, when the repo is updated, I only want to build with latexpdf the .tex files that have changes.
03:59 red82 joined #git
04:01 AfroThundr joined #git
04:04 Cabanossi joined #git
04:06 SwiftMatt joined #git
04:09 kfoonamalik joined #git
04:11 shinnya joined #git
04:11 Introoter joined #git
04:14 bronson joined #git
04:14 nihal joined #git
04:15 sea-gull joined #git
04:25 kfoonamalik joined #git
04:40 austin987 joined #git
04:42 crypt0kraken joined #git
04:50 rivarun joined #git
04:52 govg joined #git
04:54 treaki joined #git
04:56 patarr joined #git
04:56 atomicb0mb joined #git
04:59 robattila256 joined #git
05:03 clickjack joined #git
05:05 freimatz joined #git
05:07 a_thakur joined #git
05:07 a_thakur joined #git
05:07 miha_S7_ joined #git
05:12 schleppel joined #git
05:13 dsdeiz joined #git
05:14 salamanderrake joined #git
05:16 bronson joined #git
05:16 valhala_22 joined #git
05:20 Anja_ joined #git
05:21 YuGiOhJCJ joined #git
05:23 diogenese joined #git
05:26 duderonomy joined #git
05:27 sea-gull joined #git
05:32 jrahmy_ joined #git
05:35 ych joined #git
05:36 valhala_22 joined #git
05:37 cam27 joined #git
05:41 zerow joined #git
05:41 diogenese joined #git
05:42 nihal joined #git
05:46 Hunter1_ joined #git
05:49 grift joined #git
05:53 MineCoins joined #git
05:54 dasidiom joined #git
05:54 MACscr_ joined #git
05:56 zulutango joined #git
06:04 Murr joined #git
06:07 MACscr joined #git
06:10 jstimm joined #git
06:11 cr34ton joined #git
06:19 Cabanossi joined #git
06:21 ZexaronS joined #git
06:26 valhala_22 joined #git
06:29 jaggz left #git
06:31 oskarkv joined #git
06:32 AfroThundr joined #git
06:33 dreiss joined #git
06:34 jaggz joined #git
06:34 AfroThundr joined #git
06:34 jaggz what happens when I did a git push instead of git push -u origin master?
06:34 jaggz github wasn't updated and the later command just says everything is up to date
06:35 jaggz oh wait, no, it worked.. nevermind :)
06:35 jaggz :)
06:35 jaggz left #git
06:38 miha_S7_ joined #git
06:38 valhala__ joined #git
06:39 Hunter1_ joined #git
06:39 jagob joined #git
06:44 MineCoins joined #git
06:45 watabou joined #git
06:52 zefferno joined #git
06:52 PityDaFool joined #git
06:57 roelmonnens joined #git
06:58 valhala_22 joined #git
07:00 Murii joined #git
07:00 valhala__ joined #git
07:01 warthog9 joined #git
07:06 lindenle joined #git
07:08 visual hey how do i remove certain files from the index
07:08 visual added them to gitignore after git spotted them
07:09 sea-gull joined #git
07:12 ayogi joined #git
07:14 VGfreak4 joined #git
07:18 wit4r7 joined #git
07:19 visual git --rm cached <file> for anybody wondering the smae
07:19 Cabanossi joined #git
07:29 robotroll joined #git
07:34 warthog9 joined #git
07:36 miczac joined #git
07:47 j416 visual: git status will tell you
07:47 TomyLobo joined #git
07:47 j416 visual: "git reset" untracks files, it is the opposite of "git add" in a sense
07:48 j416 visual: "git rm (--cached)" deletes a file
07:48 j416 correction; "git reset" undos "git add", untrack is the wrong word. sorry.
07:49 _ng joined #git
07:50 j416 visual: tl;dr sounds like you're spot on for removing a file.
07:51 acetakwas joined #git
08:01 Milos joined #git
08:01 _main_ joined #git
08:02 _main_ joined #git
08:04 _main_ joined #git
08:05 jrahmy_ joined #git
08:07 miha_S7_ joined #git
08:08 sea-gull joined #git
08:12 valhala_22 joined #git
08:18 Marumarsu joined #git
08:18 borkr joined #git
08:18 aw1 joined #git
08:20 miha_S7_ joined #git
08:20 Cabanossi joined #git
08:26 tourdownunder joined #git
08:28 stepho joined #git
08:28 Masber joined #git
08:33 garyserj joined #git
08:36 ter0 joined #git
08:43 _ng joined #git
08:43 om_henners joined #git
08:45 roelmonnens joined #git
08:49 flavius joined #git
08:50 D630 joined #git
09:00 valhala_22 joined #git
09:00 j08nY joined #git
09:01 manuelschneid3r joined #git
09:01 dhollin3 joined #git
09:03 valhala_22 joined #git
09:06 cdown joined #git
09:07 ExpiredDev joined #git
09:14 star_prone joined #git
09:14 tymczenko joined #git
09:15 dersand joined #git
09:19 misosoup joined #git
09:24 crypt0kraken joined #git
09:24 zerow joined #git
09:29 ShapeShifter499 joined #git
09:30 Masber joined #git
09:34 catsup joined #git
09:35 Cabanossi joined #git
09:35 catsup joined #git
09:36 freimatz joined #git
09:37 shinnya joined #git
09:39 Murii joined #git
09:39 rchavik joined #git
09:41 ShapeShifter499 joined #git
09:41 tkore joined #git
09:42 Acerific joined #git
09:42 Acerific joined #git
09:43 Tobbi joined #git
09:43 vac joined #git
09:45 strk git svn fetch # [svn-remote "svn"] unknown
09:45 strk what does that mean ?
09:45 strk it only happens if `git svn fetch` is issued by a script, not from the shell
09:47 aw1 joined #git
09:47 strk oops, never mind, my fault
09:48 Spydar007 joined #git
09:48 Spydar007 joined #git
09:48 lowercaseman joined #git
09:51 plos joined #git
09:51 mikecmpbll joined #git
09:52 pycer joined #git
09:55 leeN joined #git
09:58 cdown joined #git
10:01 strk how can I skip the long (from r1) update of svn refs on `svn fetch` ?
10:01 roelmonnens joined #git
10:04 njbair joined #git
10:04 vac left #git
10:04 steve joined #git
10:06 sea-gull joined #git
10:09 PioneerAxon_ joined #git
10:10 cdown_ joined #git
10:10 pycer left #git
10:12 cebula joined #git
10:16 Lunatrius` joined #git
10:16 MattMaker joined #git
10:20 Cabanossi joined #git
10:20 aw1 joined #git
10:22 aw1 joined #git
10:24 anuxivm joined #git
10:29 m4sk1n joined #git
10:34 cdown_ joined #git
10:36 jstein_ joined #git
10:37 valhala_22 joined #git
10:43 sea-gull joined #git
10:43 phinxy joined #git
10:45 navidr joined #git
10:46 jozwior joined #git
10:47 jozwior_ joined #git
10:49 pbandark joined #git
10:55 ExpiredDev joined #git
10:55 Levex joined #git
10:55 dasidiom joined #git
10:58 theoceaniscool joined #git
10:59 miha_S7_ joined #git
11:01 _ADN_ joined #git
11:02 m0viefreak joined #git
11:06 zeroed joined #git
11:09 sarri joined #git
11:09 sarri joined #git
11:09 Aljone joined #git
11:10 xanadu_ joined #git
11:16 ZyX-I joined #git
11:18 earnestly joined #git
11:20 GT4066 joined #git
11:23 jackrandom joined #git
11:27 sunri5e joined #git
11:28 deltab joined #git
11:29 SkyPatrol_ joined #git
11:33 deltab joined #git
11:34 mikecmpb_ joined #git
11:35 dasidiom joined #git
11:39 deltab joined #git
11:41 thebope joined #git
11:42 Spydar007 joined #git
11:42 Spydar007 joined #git
11:43 bket joined #git
11:43 valhala_22 joined #git
11:47 Sadeukko joined #git
11:48 Sadeukko joined #git
11:48 grift joined #git
11:49 Introoter joined #git
11:50 raijin joined #git
11:51 grift joined #git
11:56 MineCoins joined #git
11:57 xanadu_ joined #git
11:59 romancer joined #git
12:01 romancer I'm trying to merge branches which results in some conflicts. I'm starting `mergetool` which is configured and works.
12:01 romancer However, I'm receiving 'submodule merge conflict' and 3 options: User local "submodule commit", remote file or abort.
12:02 romancer How can I force git to compare remote/local with mergetool?
12:06 jozwior joined #git
12:10 romancer seems to be sorted with `rm --cached`
12:10 _ikke_ submodules are not actually merged
12:11 JanC_ joined #git
12:11 dendazen joined #git
12:11 _ikke_ The conflict is in what commit should be checked out
12:12 User458764 joined #git
12:12 Sadeukko joined #git
12:13 blackwind_123 joined #git
12:15 livingstn joined #git
12:16 MattMaker joined #git
12:19 ilithium joined #git
12:19 ilithium Hey all :)
12:19 ilithium Anyone around today?
12:19 _ikke_ Most of the time, there is someone
12:19 davimore joined #git
12:19 xanadu_ joined #git
12:19 QwertyDragon joined #git
12:20 ilithium Fab, thanks. I'm trying to figure out how to do some stuff with branches. I've a project that's quite old and, for one reason or another, most of the work has been done in a branch other than master. What I'd like to do is effectively change 'master' to 'old-master' and then make the other branch 'master'. I found an answer here: http://stackoverflow.com/questions/2763006/make-the-current-git-branch-a-master-branch but I'm not sure if this is the right way t
12:21 ilithium So I was hoping someone could explain this to me so I'd understand
12:21 Raging_Hog joined #git
12:21 grumble joined #git
12:23 _ikke_ ilithium: What the accepted answer does is merge master into the other branch, but make sure that all changes from the other branch are preserved
12:23 _ikke_ ilithium: that way, you are not rewriting history
12:23 ilithium Ah, OK. So I could create branch 'archive', do that and 'archive' would be identical to master, including it's history?
12:24 NeverDie joined #git
12:24 SkyPatrol joined #git
12:25 _ikke_ You still have to do the other merge (which can fast-forward), then they should be equal
12:27 Sadeukko joined #git
12:29 livingstn joined #git
12:33 nowhere_man joined #git
12:34 ilithium _ikki_ thanks - so that would make 'archive' equal to 'master'. So how would I then make the other branch 'master'? Would I do a rebase on that?
12:35 thebope joined #git
12:39 sea-gull joined #git
12:41 ShapeShifter499 joined #git
12:41 tourdownunder joined #git
12:42 jnavila joined #git
12:49 TSS_ joined #git
12:52 shinnya joined #git
12:52 phinxy did copy a folder to another naming it backup
12:52 phinxy now the backup seems to experience the same issue as the first one which i git pulld
12:53 phinxy if i only knew git better i would roll back commits
12:54 dasidiom joined #git
12:56 jameser joined #git
12:58 MineCoins joined #git
12:59 valhala_22 joined #git
12:59 red82 joined #git
13:01 meLon joined #git
13:04 valhala_22 joined #git
13:05 angrybacon joined #git
13:07 finalbeta joined #git
13:07 angrybacon Hello, not sure what I did but Github show an extra branch "HEAD" while I can't see it with `git branch`
13:07 angrybacon https://github.com/angrybacon/mdi
13:08 adac joined #git
13:09 raijin joined #git
13:10 Levex joined #git
13:10 strk I'm not finding a way to setup git-svn so it matches the hashes used by a remote mirroring system
13:10 strk one of the problems is that the mapping between SVN author and email changed overtime
13:11 livingstn joined #git
13:11 strk so now I cannot re-fetch all revisions and end up using the same hashes unless I use an author-script to change full-name based on revision (which I don't even know what it is)
13:11 strk what I'd like to do is convincing git-svn that the mappings are to be built from metadata, which are present in git logs
13:11 strk how to do tha t?
13:12 gnixev joined #git
13:13 NeverDie joined #git
13:17 NeXTSUN joined #git
13:20 Cabanossi joined #git
13:23 Noldorin joined #git
13:23 livingstn joined #git
13:24 valhala_22 joined #git
13:25 red82 joined #git
13:27 Levex joined #git
13:29 red82 joined #git
13:29 thebope joined #git
13:29 valhala_22 joined #git
13:32 astrofog joined #git
13:32 robattila256 joined #git
13:34 valhala_22 joined #git
13:36 livingstn joined #git
13:37 valhala_22 joined #git
13:37 Pulp joined #git
13:41 bouncy joined #git
13:46 MineCoins joined #git
13:47 red82 joined #git
13:47 User458764 joined #git
13:47 valhala_22 joined #git
13:49 misosoup joined #git
13:51 kexmex joined #git
13:53 kyan joined #git
13:53 valhala_22 joined #git
13:56 clickjack joined #git
13:57 miha_S7_ joined #git
14:00 dsdeiz joined #git
14:00 dsdeiz joined #git
14:02 User458764 joined #git
14:05 MineCoins joined #git
14:10 cdown joined #git
14:10 cam27 joined #git
14:14 Eryn_1983_FL joined #git
14:16 valhala__ joined #git
14:20 angrybacon joined #git
14:23 valhala_22 joined #git
14:24 theoceaniscool joined #git
14:24 sea-gull joined #git
14:24 jstimm joined #git
14:30 acetakwas joined #git
14:30 robattila256 joined #git
14:30 Gsham joined #git
14:31 acetakwas joined #git
14:32 MarioBranco joined #git
14:32 acetakwas joined #git
14:33 aielima joined #git
14:35 Cabanossi joined #git
14:36 MarioBranco joined #git
14:36 ayogi joined #git
14:37 livingstn joined #git
14:38 transhuman hi I am looking at the following command git diff tag1 tag2 -- > the-patch.diff does this put all the differences found in git between tags (with multiple commits) into a file (like a normal patch file where you have your initial copy of git and it layers the changes in all the files over that git level? so in this case the initial commit level is at tag1 and changes can be layered over those files??
14:41 fmcgeough joined #git
14:41 aw1 joined #git
14:49 dsdeiz joined #git
14:49 dsdeiz joined #git
14:50 kiichi joined #git
14:50 durham joined #git
14:56 Nilesh____ joined #git
14:58 jozwior joined #git
14:59 miha_S7_ joined #git
14:59 madewokherd joined #git
15:00 Hunter1 joined #git
15:00 theiostream joined #git
15:01 SteffanW joined #git
15:02 dasidiom joined #git
15:05 Cabanossi joined #git
15:05 durham joined #git
15:05 BSaboia joined #git
15:06 valhala_22 joined #git
15:10 valhala_22 joined #git
15:10 valhala_22 joined #git
15:11 a_thakur joined #git
15:11 pbandark1 joined #git
15:14 Goplat joined #git
15:14 theiostream joined #git
15:18 ghostdog joined #git
15:18 cdown joined #git
15:19 CussBot joined #git
15:21 clickjack joined #git
15:21 fmcgeough joined #git
15:22 tmsmith joined #git
15:25 MarioBranco joined #git
15:28 tango_ joined #git
15:29 ski7777 joined #git
15:31 maroloccio joined #git
15:34 overlord_tm joined #git
15:34 _ng joined #git
15:34 h12o joined #git
15:35 babilen joined #git
15:35 tonyj joined #git
15:36 borkr joined #git
15:36 dave0x6d joined #git
15:37 Lyqyd joined #git
15:38 Gsham joined #git
15:38 p1k is it possible to force git to try and apply a patch and produce a conflict (such as on a merge) rather than leave an error
15:39 bremner p1k: in what context? what git command?
15:39 bremner see e.g. git am -3
15:40 chubek_ joined #git
15:40 p1k bremner: git apply ?
15:40 hseg joined #git
15:41 j416 p1k: man git-apply search for -3
15:41 gitinfo p1k: the git-apply manpage is available at https://gitirc.eu/git-apply.html
15:41 j416 same thing
15:41 p1k bremner: particularly git simply seems to complain about the context lines not being at the same line number as in the patch
15:42 hseg Having trouble getting git-annex to work under cygwin on windows. Any suggestions for alternatives?
15:42 p1k I cannot do -3
15:42 p1k error: while searching for: <displaced loc> error: repository lacks the necessary blob to fall back on 3-way merge.
15:42 LeBlaaanc joined #git
15:43 aw1 joined #git
15:44 bremner hseg: plenty of suggestions for alternatives to windows, but I guess that's not what you meant.
15:44 p1k I tried the --ignore-whitespace flag
15:45 hseg No, unfortunately
15:45 transhuman what about using docker or something hseg?
15:45 crypt0kraken joined #git
15:45 bremner hseg: git-lfs is the obvious idea, although I don't know much about it other than the big hosters seem to be into it
15:45 transhuman docker on windows
15:46 transhuman just my two  cents worth..ask the other guys though for sure
15:46 transhuman anyone have an answer to my question ?
15:47 hseg transhuman: Docker? Seems like a bit of a sledgehammer
15:47 QwertyDragon joined #git
15:47 fmcgeough joined #git
15:47 transhuman well they some times work!
15:47 hseg bremner: Does git-lfs support putting the large files in some local directory? And tracking potential sources?
15:47 transhuman some times its easier to avoid a problem
15:47 transhuman anyways I will shut up about it now
15:48 jt joined #git
15:49 transhuman hi I am looking at the following command git diff tag1 tag2 -- > the-patch.diff does this put all the differences found in git between tags (with multiple commits) into a file (like a normal patch file where you have your initial copy of git and it layers the changes in all the files over that git level? so in this case the initial commit level is at tag1 and changes can be layered over those files??
15:50 transhuman ie does it capture all the changes between those two tags?
15:52 p1k thanks for terrible defaults git
15:52 transhuman or do I need to do a git rebase and then a git format-patch or something as I was just reading?
15:53 transhuman which of course would loose some possibly good information to have in the future
15:54 Hello71 if you checkout tag1, copy out the working tree, then checkout tag2, then diff the two with "diff -ru..." it will be the same
15:56 transhuman so let me make sure I understand this...tag1 + diff = tag2?
15:56 transhuman (which is what I want)
15:57 fmcgeough joined #git
15:57 theiostr_ joined #git
15:58 dasidiom joined #git
15:58 transhuman I am trying to get a history of changes in a series of files to a particular git between version releases of that project so tag1 equals version 2.2.1 and tag two would be 2.2.2 and tag 3 would be 2.2.3...etc etc and my series of patch files would contain those differences
15:59 miha_S7_ joined #git
16:00 transhuman Hello71, is that what you are saying?
16:01 p1k has no equivalent of `patch` --fuzz ?!
16:03 theiostream joined #git
16:04 transhuman ok then what about this 1) git checkout tag2 2) git format-patch tag1 --stdout > patch1to2.patch ? would that include the changes in all the commits between two tags?
16:04 transhuman thanks guys, by the way for the responses
16:04 transhuman its appreciated
16:06 jnavila joined #git
16:10 fmcgeough joined #git
16:10 beerdrop joined #git
16:12 User458764 joined #git
16:13 Hello71 transhuman: yes, that is right
16:13 Hello71 just do git diff
16:13 Hello71 with format-patch you get the commit messages
16:13 Hello71 so you can do "git am" and have exactly the same git log, not only the same tree
16:14 blackwind_123 joined #git
16:14 Hello71 well, assuming you also have the foreign objects.
16:14 manuelschneid3r joined #git
16:17 Hunter1 joined #git
16:17 pbandark1 joined #git
16:18 xanadu_ joined #git
16:18 transhuman Hello71, what do you mean by foreign objects ..sounds like it might be a problem...thanks in advance
16:18 MattMaker joined #git
16:19 transhuman Hello71, got a doc on foreign objects and git?
16:19 Hello71 if you have merge, then you also need the other side of the merge
16:19 Hello71 (or sides, in case of octopus merge)
16:21 theiostr_ joined #git
16:24 Abbott I currently have a project managed by git that has three forks, master, new_code, and repair_functionality. I want to delete repair_functionality, fork master to dist or oldstable or some name that means it's the old version and doesn't work anymore, then merge new_code to master unconditionally keeping everything from new_code and deleting everything that new_code isn't using that's in master. The end result
16:24 jnavila joined #git
16:24 Abbott will be two forks, master and old (or whatever). Is this good style? Is there a way to "merge" that is really just replacing everything with a branch?
16:25 cbreak Abbott: sure
16:25 jast Abbott: seems like a sensible approach, and yes, there is a way to do it :)
16:25 cbreak go to that branch, and git merge -s ours
16:25 cbreak that will keep that branch
16:25 jast but that's the opposite of what's intended here...
16:26 cbreak then ff merge into master
16:26 jast oh, yeah, I guess if you do it back and forth...
16:26 jast that would put the merge in new_code, too, though
16:26 dfas joined #git
16:26 cbreak the idea behind that is to keep the first parent as mainline
16:26 cbreak the old mainline is second parent
16:26 crypt0kraken joined #git
16:27 cbreak not sure if git finally has -s theirs either :)
16:27 cbreak (anyway, I think it's better to use ours, because of parent order
16:27 cbreak )
16:28 jast I guess that makes sense, I've never thought about that
16:28 menip joined #git
16:30 NeverDie joined #git
16:31 xanadu_ joined #git
16:33 Abbott so git merge -s ours uses merge strategy "ours" which sounds like it just copies the branch to be merged instead of trying to do changes, right? or would ours mean to keep master?
16:34 rpdce joined #git
16:35 theiostream joined #git
16:35 jast 'ours' is a dummy merge: it marks a branch as merged but doesn't actually apply any changes from it
16:35 jast you want the opposite, hence the idea to do an 'ours' merge inside new_code, and then merge back in the other direction
16:35 Cabanossi joined #git
16:37 transhuman Hello71, I will have to look that up, sounds like what I want to do isn't as simple as a couple commands..sigh
16:37 ptepetey joined #git
16:38 pks joined #git
16:38 kexmex joined #git
16:38 Abbott okay, so merge master into new_code with ours, then fast forward master to where new_code will point to? then I can delete new_code, as master will be at the same place?
16:39 jast transhuman: actually a simple 'diff' is enough to have the differences between two commits/refs/tags. you might want to include the --binary flag so you can apply changes to binary files, too
16:39 jast everything else is just bells and whistles
16:39 jast Abbott: that's the idea, yeah
16:39 thebope joined #git
16:40 h12o joined #git
16:40 Abbott okay, I'll give that a try
16:40 Abbott thanks jast
16:41 Vampire0 joined #git
16:43 hrehf joined #git
16:44 hrehf how can I "forget" a file without touching the file? (I think in gitspeak this would be "remove a file from the index but not from working copy")
16:44 cbreak git rm --cached
16:44 cbreak this will NOT remove history
16:44 cbreak it'll just stage the file for removal
16:44 jast one thing to be aware of is that if you make a commit in which the file is no longer tracked, and someone else updates a clone, applying that commit will remove the file in question in that clone, though of course it can be manually restored from the previous commit afterwards
16:45 zhuyue123 joined #git
16:45 hrehf cbreak: but I don't want it removed or touched. Just leave it in the filesystem, in all clones etc
16:46 cbreak hrehf: it will be left in the filesystem in YOUR clone
16:46 cbreak hrehf: it will NOT be in the next commit (since you removed it)
16:46 cbreak everyone that merges your new commit will merge all your changes
16:46 cbreak including your removal
16:46 cbreak that means your removal is applied too
16:46 cbreak -> the fille will be removed
16:47 hrehf so is there a way to just remove it from index, not from fs?
16:47 justan0theruser joined #git
16:47 jast locally, yes
16:47 hrehf hm, weird, this sounds like a really useful thing to do
16:47 jast but the only way to let other repos know that it should vanish from the index is by making a commit without it. and those clones will remove it from the working tree when updating.
16:47 sea-gull joined #git
16:48 cbreak hrehf: as I said
16:48 hrehf Okay, thanks. (because it's interesting: why is "just dont track this anymore" not a valid action?)
16:48 cbreak git rm --cached will do that
16:48 Abbott jast: when I do git merge -s ours, I should be checked out on new_code, right? should the exact command be git merge -s ours master? then checkout master, and git merge and it will ask me to ff?
16:48 cbreak hrehf: that's what it does
16:48 cbreak it stops tracking the file
16:48 hrehf so why would it ever touch an untracked file in fs?
16:48 cbreak stopping to track the file means it stops existing in git
16:48 hrehf (on the other copies)
16:49 cbreak hrehf: because someone other than you, merging your changes, wants your changes
16:49 cbreak and removing the file is a change
16:49 cbreak the new state is "that file is not there"
16:49 cbreak so it is removed
16:49 cbreak git will not touch untracked files, but files that become untracked it will remove
16:50 hrehf well, to me it seems that's not necessary or logical at all :/
16:50 cbreak hrehf: why?
16:50 hrehf the action should not be "remove it from index and filesystem [sometime in the future]"
16:50 cbreak it seems totally logical to me :)
16:51 cbreak hrehf: why not?
16:51 hrehf I see no logical connection between tracking a file and removing it
16:51 cbreak why not?
16:51 cbreak it's tracked
16:51 cbreak that means git maintains it
16:51 hrehf as lon as you track it, you might remove it. but when I tell you to stop tracking it, why would you remove it?
16:51 cbreak because removing is an action
16:52 jast there's no such thing as "stop tracking"
16:52 cbreak like changing files
16:52 jast you're *removing* it (from the index)
16:52 theoceaniscool left #git
16:52 jast in the commit, that is recorded as the absence of the file. in terms of a diff, the file is removed in that commit.
16:52 theoceaniscool joined #git
16:52 jast and since *normally* you want a removed file to be removed when a commit is applied, that's what happens
16:52 freddy_ joined #git
16:52 transhuman thanks jast
16:53 jast this situation is sufficiently more common that it's pretty much acceptable to have some extra manual effort when you want other behaviour, at least in my opinion
16:53 raijin joined #git
16:53 hrehf jast: I think i'm just confused that there's no "remove it from index only" action :-)
16:53 M6HZ joined #git
16:54 jast given Git's structure it's pretty much impossible to make that distinction
16:54 hrehf for me it's really common to have something like a config in the repo for faster deplooyment until the project gets "published"
16:54 jast the index is the staging area: it records what should go in the next commit
16:54 miha_S7_ joined #git
16:54 jast so if you remove something from the index and commit that, it's not in the commit
16:55 hrehf at some point I will stop tracking the file, but ofc you don't want it removed from the users systems
16:55 cbreak hrehf: as I said repeatedly
16:55 cbreak git rm --cached
16:55 jast and whether you did so intending to keep the file around vs. intending to get rid of the file altogether, Git can't tell from the index, and certainly not from the resulting commit
16:55 cbreak hrehf: the users should stop being lazy and get their own config files
16:55 cbreak or write your program to create your config file from a default file if it doesn't exist
16:56 hrehf cbreak: yes, thank you, I totally saw that and will use it. But as I understand it will remove files on the next remote pull
16:56 cbreak if your program can't handle a missing config file, you have bigger problems :)
16:57 hrehf cbreak: well, i always check in all my scripts etc, even to push cahnges between servers etc
16:57 hrehf so it really makes sense
16:57 cbreak have you tried ansible?
16:57 cbreak rsync?
16:57 freddy_ joined #git
16:59 hrehf jast: I think my point is that there would need to be  a disctinction between "stop tracking -> delete from fs " and "stop tracking file"
16:59 hrehf but I wasted enough of your time, thanks for the discussion & help!
16:59 thiago joined #git
17:00 User458764 joined #git
17:00 jast and my point is that the structures don't allow for that distinction to exist
17:00 hrehf cbreak: rsync sucks for this, because you need merging when working on multiple systems
17:01 cbreak you work on a server? :O
17:01 cbreak bad href, bad!
17:01 hrehf cbreak: what's ansible? the website doesn't want to tell me what it /actually/ does :D
17:01 hrehf haha
17:01 hrehf well, it's not "production" per-se
17:01 hrehf I don't depend on my rss filtering scripts THAT much :-)
17:02 cbreak hrehf: ansible is a configuration / deployment tool
17:02 cbreak it can be used to set up computers remotely, install stuff, manage them
17:02 hrehf ah
17:02 cbreak can be used to manage many of them at once
17:03 hrehf cool, will check it out
17:03 hrehf thanks guys!
17:04 zeroed- joined #git
17:05 om_henners joined #git
17:05 lindenle joined #git
17:05 hrehf oh, another question: why does rebasing need a clean wc?
17:06 kadoban joined #git
17:06 hrehf bah, maybe that's a dumb question if rebase always uses the currently checked out commit
17:09 kexmex joined #git
17:09 scarabx joined #git
17:09 DaGoaty joined #git
17:14 ptdleos joined #git
17:14 m4sk1n joined #git
17:19 dendazen joined #git
17:20 LeBlaaanc joined #git
17:21 diogenese joined #git
17:26 BSaboia joined #git
17:27 Douko joined #git
17:30 Hunter1 joined #git
17:32 TbobbyZ joined #git
17:32 jnavila joined #git
17:32 justanotheruser joined #git
17:34 TbobbyZ joined #git
17:35 MarioBranco joined #git
17:35 kfoonamalik joined #git
17:37 zeroed joined #git
17:38 synthmeat joined #git
17:38 synthmeat joined #git
17:39 Lyqyd_ joined #git
17:39 Atm0spher1c joined #git
17:39 PioneerAxon joined #git
17:42 h12o joined #git
17:45 zeroed joined #git
17:46 mentazoom joined #git
17:46 synthmeat joined #git
17:47 grayjoc joined #git
17:49 paws___ joined #git
17:49 thebope joined #git
17:49 roby10 joined #git
17:49 raijin joined #git
17:50 njbair joined #git
17:53 Naan joined #git
17:54 sunri5e_ joined #git
18:00 dasidiom joined #git
18:01 Vgr_ joined #git
18:01 Vgr_ joined #git
18:01 lindenle_ joined #git
18:01 owlbynight-away joined #git
18:01 apotry joined #git
18:02 codedmart_ joined #git
18:02 grayjoc joined #git
18:02 Whisket_ joined #git
18:02 l2y_ joined #git
18:02 aboccia__ joined #git
18:03 devhost_ joined #git
18:03 barq_ joined #git
18:03 raatiniemi_ joined #git
18:04 frobnic_ joined #git
18:04 Kuukunen- joined #git
18:04 Ruler-_ joined #git
18:05 Jan-_ joined #git
18:05 moritz_ joined #git
18:05 Cabanoss- joined #git
18:06 Lyqyd joined #git
18:07 valhala_22 joined #git
18:07 Hunter1 joined #git
18:07 ePirat- joined #git
18:07 binary10101 joined #git
18:07 theskillwithin_ joined #git
18:08 cebor_ joined #git
18:08 c0d3d joined #git
18:08 zifnab| joined #git
18:08 Gunni^ joined #git
18:08 Arkaniad_ joined #git
18:08 zero1985 joined #git
18:08 cydrobolt_ joined #git
18:08 Matsv- joined #git
18:08 raijin joined #git
18:08 cwre_ joined #git
18:08 yaiyan joined #git
18:08 Cthalupa joined #git
18:09 saidelike_ joined #git
18:09 theskillwithin_ left #git
18:09 adac joined #git
18:10 reality404 joined #git
18:10 devster31 joined #git
18:10 doppo joined #git
18:10 bittyx joined #git
18:10 robinsmidsrod joined #git
18:11 polymorphisme joined #git
18:11 safe joined #git
18:11 theskillwithin joined #git
18:11 theskillwithin wow thats fun, if i get netsplit, i heave to leave this channel just to change my name back to my registered name
18:11 spb no you don't
18:11 Someguy123 joined #git
18:11 spb you just have to identify before changing nick
18:12 theskillwithin u cant speak in the chan unless ur registered, so it wont let you change ur name because that requires speaking into the chan of the name change
18:12 linuxmodder joined #git
18:12 theskillwithin lol nope i just tried it
18:12 pratch joined #git
18:12 oeopt joined #git
18:12 spb you tried to change nicks before identifying
18:12 spb try doing it the other way round
18:12 theskillwithin no i was identified first
18:12 theskillwithin Message(435): theskillwithin #git Cannot change nickname while banned on channel
18:13 spb if it didn't let you change nicks before, then clearly you weren't identified
18:13 theskillwithin its exact reason was the channel
18:13 theskillwithin i had to leave this channel, then it let me change the nick
18:13 theskillwithin i was in 20 other chans too btw that i didn't have to leave
18:14 tgummerer noob question: how do you identify before changing nicks?
18:14 theskillwithin idk my znc does it somehow
18:14 thekingofbandit joined #git
18:14 spb theskillwithin: i just checked. you were using this nickname when you identified today
18:14 tazle_ joined #git
18:14 oprypin tgummerer  provide IRC server password as   username:password
18:14 theskillwithin ?
18:15 theskillwithin on the netsplit my nick changed to theskillwithin_
18:15 tgummerer makes sense, thanks oprypin
18:15 theskillwithin so when i tried changing it back, that message occured
18:15 transhuman joined #git
18:15 SpeakerToMeat joined #git
18:15 spb no, on the netsplit you got disconnected and made a new connection with that nick
18:15 bremner anyway, this doesn't have much to do with git.
18:15 theskillwithin correct
18:15 spb and the new connection wasn't identified
18:15 jnavila joined #git
18:15 Mission-Critical joined #git
18:15 theskillwithin my point is to change my nick BACK
18:15 theskillwithin i had to leave the chan
18:15 ankit01ojha joined #git
18:15 oprypin btw take note of this, theskillwithin,  it's nice to not use nickserv authentication, and instead use   password = username:password  or SASL
18:15 spb no you didn't, you just had to identify to nickserv before changing nicks
18:15 theskillwithin i mean the error msg specifically points out this chan, it couldnt be more clear
18:15 _ADN_ joined #git
18:16 theoceaniscool joined #git
18:16 oprypin i use znc with this and didnt have such a problem
18:16 theskillwithin nope
18:16 theskillwithin all i did to get it to work was leave this chan
18:16 theskillwithin then i could change nicks and i was already identified
18:16 spb which you didn't have to do
18:16 theskillwithin then explain this" Message(435): theskillwithin #git Cannot change nickname while banned on channel"
18:16 swalladge joined #git
18:16 nikow joined #git
18:16 spb you weren't identified when you tried to change nicks
18:16 jt joined #git
18:16 NeXTSUN joined #git
18:16 theskillwithin omfg
18:17 mmlb joined #git
18:17 theskillwithin its because the chan quites people who aren't registered, and but it also prevents you from changing names.
18:17 bremner yes, it's a know issue with freenode
18:17 spb no, it only prevents you from changing nicknames *while you're not identified*
18:17 theskillwithin NO
18:17 kadoban You seem to be violently agreeing. Can you stop it?
18:18 Geminizer joined #git
18:18 theskillwithin see everyone else gets it that i talk to about it but u spd so /shrug
18:18 theskillwithin cuz this happens in like 4 other chans
18:18 spb theskillwithin: i'm pretty sure i know what the server does, considering that i wrote it
18:18 theskillwithin and there like yep, thats just how it is
18:18 GT4066 joined #git
18:19 pks joined #git
18:19 theskillwithin i would suggest testing it to really see because I am quite sure
18:19 spb i have done. extensively.
18:19 patrickr[m] joined #git
18:19 gnixev joined #git
18:19 timlyo[m] joined #git
18:19 spb as have many other people
18:19 dikiaap joined #git
18:20 theskillwithin then why does it only give that msg from chans that require you to be authed to speak?
18:20 MattMaker joined #git
18:20 dyce[m] joined #git
18:20 theskillwithin and why does it instantly work when i part from those chans
18:20 bremner ffs. would you two let it go?
18:20 saintaquinas[m] joined #git
18:20 pee joined #git
18:20 spb theskillwithin: because those are the only channels in which you're quieted until you identify
18:21 Seveas bremner: do you want to build a snowman?
18:21 bremner yes!
18:21 Seveas Come on let's go and play!
18:21 kadoban I wouldn't mind building a snowman, but it's like 90F :-/
18:22 Seveas kadoban: how much is that in first world measurements?
18:22 kadoban 35C possibly? Too lazy to check
18:22 spb moderately warm
18:22 spb 32
18:23 Seveas friggen hot
18:23 Seveas moderately warm is 18°C
18:23 spb depends on your frame of reference really
18:26 Gustavo6046 joined #git
18:28 BSAlb joined #git
18:28 raijin joined #git
18:28 Phylock joined #git
18:29 dreiss joined #git
18:29 NeXTSUN joined #git
18:29 Altiare joined #git
18:30 BSab joined #git
18:31 pks joined #git
18:32 infernix joined #git
18:32 Shakeel_ joined #git
18:35 Cabanossi joined #git
18:36 rivarun joined #git
18:37 somiaj left #git
18:37 bl0m1 joined #git
18:37 MineCoins joined #git
18:41 LeBlaaanc joined #git
18:41 de_fragment joined #git
18:42 pks joined #git
18:43 h12o joined #git
18:45 tmsmith joined #git
18:45 aielima joined #git
18:46 bl0m1 joined #git
18:46 Eryn_1983_FL hi peeps
18:46 Eryn_1983_FL anybody ever use git-annex?
18:46 duderonomy joined #git
18:46 Eryn_1983_FL so uh i kind of made a booboo atleast ihope it isn't
18:47 SkyPatrol_ joined #git
18:47 Eryn_1983_FL i got my big repo of files using it and its taking hours to add files and i am worried it will be unfixable if it fills up myu drive
18:47 yqt joined #git
18:48 _ikke_ Eryn_1983_FL: Sorry, haven't used it myself
18:49 jimmyrcom joined #git
18:49 Eryn_1983_FL ok i found their channl
18:54 aw1 joined #git
18:56 ghostdog joined #git
19:01 rewt joined #git
19:02 mellernoia joined #git
19:03 satifant joined #git
19:03 bitmod joined #git
19:08 scarcry joined #git
19:09 grayjoc joined #git
19:12 tymczenko joined #git
19:12 User458764 joined #git
19:15 yaewa joined #git
19:16 sea-gull joined #git
19:17 star_prone joined #git
19:19 Darcidride joined #git
19:19 zivester joined #git
19:19 ki0 joined #git
19:21 Sadeukko joined #git
19:23 vamiry joined #git
19:25 durham joined #git
19:25 groverdill joined #git
19:26 cads joined #git
19:27 mostlybadfly joined #git
19:30 Lyqyd_ joined #git
19:36 Cabanossi joined #git
19:40 tanuki joined #git
19:40 Hunter1 joined #git
19:43 PioneerAxon joined #git
19:44 zerow joined #git
19:45 manuelschneid3r joined #git
19:47 Gsham joined #git
19:47 Milos joined #git
19:49 Schrostfutz joined #git
19:49 livingstn joined #git
19:50 Schrostfutz hi, I have a small question: I have a feature branch that diverged off of the master. I accidentaly forked another feature branch from there instead of master. I want to reintegrate the second, but not the first, into the master. I rebased onto master which redid all of branch 1s commits. When later merging branch 2 also into master, will it detect that all of the commits are already there?
19:51 _ikke_ What did you rebase on master?
19:52 theiostream joined #git
19:52 thiago you forgot the --onto option when rebasing onto master
19:58 davimore joined #git
19:58 cebula joined #git
20:02 Schrostfutz _ikke_, the second branch. I want to merge it now but am afraid I'll run into issues when merging the first branch
20:03 thiago then you did the rebase wrong. Use the --onto argument.
20:03 Schrostfutz thiago: Okay, I'll try that
20:03 transhuman I  have been thinking of my objective and the 1) git checkout tag2 and the 2) git diff --binary flag tag1 tag2 -- > the-patch.diff this will put all the changes to the tree in one file. I have been thinking about the usability of this data in this format for my application. Now I am considering is it possible to get a complete trace of the changes of code contained in a file over the different commits/versions/tags . So
20:03 transhuman lets say the file is iodata.h this may have changed 30 times throughout the history of the git tree.  It would be nice to have this in one file and changes to the next file in another file over time...Is that possible?
20:04 transhuman a diff for each file over the whole history
20:04 thiago transhuman: git format-patch can do that
20:04 transhuman this would be much better training data for my neural net
20:05 Cabanossi joined #git
20:05 thiago git rev-list --reverse -- iodata.h | xargs -n1 -I@ git format-patch -n1 @ -- iodata.h
20:05 thiago or something like that
20:06 transhuman thiago, ok I will look into this command for this application
20:06 transhuman thanks
20:08 AfroThundr joined #git
20:10 CDanU joined #git
20:11 CDanU How do I make 'git stash' return a non zero if there is nothing to stash?
20:12 thiago change the source code
20:13 kexmex joined #git
20:13 pbrewczynski_c joined #git
20:13 kadoban CDanU: If this is for a script, what I've done is check  'git rev-parse --verify -q refs/stash' before and after the attempted stash and see if they match.
20:13 Eugene CDanU - you can wrap `git` with a hilarious crappy bash wrapper liek I do
20:14 clickjack joined #git
20:14 Eugene https://github.com/EugeneKay/scripts/blob/master/HOMEDIR/.bashrc#L127
20:15 CDanU Ok, thanks. So it seems that I am not the only one with this problem.
20:15 dendazen joined #git
20:15 Eugene I didn't say this was a good idea ;-0
20:15 moei joined #git
20:17 CDanU Eugene: wow :P, you know that you can define simple aliases in the gitconfig right?
20:18 thiago you can also verify whether the worktree or index differ from the commit before stashing
20:18 durham joined #git
20:18 Eugene You can do a lot of things with git aliases, but importantly you cannot redefine builtins
20:18 Eugene I know exactly what I am doing, and it's a naughty thing.
20:19 CDanU ok thx for the input I can work with that
20:19 Anja_ joined #git
20:19 Eugene The sane thing to do is add your own subcommand; put git-foobar somewhere in $PATH and `git` will pick it up as `git foobar`
20:20 Eugene Eg, `git stuff` instead of stash. Add your own behaviour, bada-boom
20:20 Eugene Or for a simple thing, use alias
20:20 MattMaker joined #git
20:21 CDanU Jeah I just taught that I am to stupid to find an 'obviously had to be implemented feature
20:22 TheSimonator joined #git
20:22 thiago why were you teaching that?
20:22 CDanU typo :)
20:23 thiago oh
20:25 sbasso joined #git
20:26 leeN joined #git
20:26 MineCoins joined #git
20:26 sonicmagister joined #git
20:28 Hunter1 joined #git
20:32 ewitz joined #git
20:33 gnixev joined #git
20:35 MarioBranco joined #git
20:39 bigjazzsound joined #git
20:40 sonicmagister quit
20:47 chrisM_1 joined #git
20:52 fstd joined #git
20:52 yinye joined #git
20:53 tmsmith joined #git
20:54 Rotaerk joined #git
20:58 hseg joined #git
20:59 ertesx joined #git
21:01 hseg Hi. I have a bunch of repos hosted at the same place (atm locally, but will be moved to a vps soonish). Is there a way of combining them under one alias so that when I later change to connection via ssh I will just have to change one url?
21:03 hseg left #git
21:03 hseg joined #git
21:03 hseg Oops
21:05 hseg I mean, I know that if I have an SSH alias for a server that it also works for git.
21:06 hseg So the question is whether something similar exists for local repos.
21:06 rudi_s hseg: AFAIK no, but you could also use SSH for local repos.
21:07 hseg Oh, wait, it seems that url.<base>.insteadOf could work.
21:08 hseg rudi_s: i.e. install SSH locally and have the URL be localhost?
21:08 rudi_s hseg: Yes. I use that for a few tasks. But if insteadOf helps, then there should be no need.
21:10 gesh joined #git
21:10 gesh Right. It's just that I don't want to go through the pain of setting up SSH on windows.
21:13 hseg Just checking: .gitconfig doesn't support sub-subsections, correct?
21:17 tmsmith joined #git
21:19 theiostream joined #git
21:19 rmbeer_ joined #git
21:19 rmbeer_ hello
21:19 gitinfo rmbeer_: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
21:20 hseg I'm unsure what invocation of git-config would insert the lines [url "/path/to/srv"] insteadOf = srv:// into .gitconfi
21:20 hseg I'm unsure what invocation of git-config would insert the lines [url "/path/to/srv"] insteadOf = srv:// into .gitconfig
21:20 rmbeer i have problem for use .git/ for get a copy to work directory
21:21 rmbeer i use 'git checkout -- *' but can't get a copy from the last copy
21:21 hseg I thought git config --global 'url."/path/to/srv"' insteadOf 'srv://' would do the trick, but I get error: invalid key: url."/path/to/srv"'
21:21 rmbeer i have a accident and remove several files from the work directory
21:21 hseg left #git
21:21 hseg joined #git
21:22 rmbeer anyone helpme?
21:22 hseg joined #git
21:23 justan0theruser joined #git
21:24 hseg OK, nm, turns out I needed git config --global 'url./path/to/srv/.insteadOf' 'srv://'
21:24 dsdeiz joined #git
21:24 dsdeiz joined #git
21:25 rmbeer i can't fix it with nothing, 'git pull', 'git checkout [other branch]'
21:25 dermoth joined #git
21:26 hseg This works well, except for one detail. Checking with git remote -v, I see that clones made with the url changed have the alias expanded.
21:26 fission6 joined #git
21:26 hseg This means that if I ever change the url to use, git won't notice.
21:28 acresearch joined #git
21:28 acresearch people, i am new to git, i cannot commit (there is something wrong with doing a git commit -a) what can i so?
21:29 boombatower joined #git
21:30 tgummerer acresearch: what do you mean by "i cannot commit"? Is there an error message?
21:30 acresearch tgummerer: i get (*** Please tell me who you are.)
21:30 acresearch do i have to sign in?
21:31 oprypin https://help.github.com/articles/setting-your-username-in-git/
21:31 rmbeer error: RPC failed; curl 52 Empty reply from server
21:31 tgummerer ah.  Before committing you have to set up your name and email address
21:31 rmbeer what is this??
21:31 acresearch people, i am new to git, i cannot commit (there is something wrong with doing a git commit -a) what can i so?oh
21:31 acresearch oh
21:31 Acerific joined #git
21:31 hseg acresearch: Wait, you're getting a message along the lines of "give me a username and email"?
21:31 rmbeer i try remove all repository from my disk and 'get clone' to new directory
21:32 acresearch hseg: i think
21:32 acresearch hseg: why?
21:32 rmbeer and get this error
21:32 transhuman thiago on the command git rev-list --reverse -- iodata.h | xargs -n1 -I@ git format-patch -n1 @ -- iodata.h I see one problems duplicate files. So I have this command git log --pretty=format: --name-only --diff-filter=A | sort -u | tee files-list.txt it includes the path to the file : Question can the first command you gave me be changed to include the path to the file (this avoids duplicates) but once again has a small
21:32 transhuman problem of what happens when a file changes path in the tree over the history (so thats TWO questions) thanks in advance
21:33 hseg Because git maintains attribution information for each commit. It will note who committed and what their email address is.
21:33 hseg This is useful in collaboration, for example.
21:34 acresearch hseg: won't it do that when it pushed?
21:34 thebope joined #git
21:34 oprypin acresearch, pushing cannot change commits, they must already have that information in them
21:34 hseg Yes, but how do you think git will know e.g. your email address?
21:35 oprypin well the answer would be "github magic" but no, it doesnt do that
21:35 acresearch hseg: isn't my email address my username?
21:35 oprypin acresearch, it's not an actual username used anywhere, it's usually your real name
21:35 acresearch hseg: sorry for the simple questions, i have been using git from the web broswr and just started learning the command line
21:35 zeroed joined #git
21:35 oprypin O_o
21:36 hseg oprypin: Consequence of Github making web committing easy.
21:36 acresearch hseg: it is
21:36 hseg acresearch: Look at the link oprypin posted: https://help.github.com/articles/setting-your-username-in-git/
21:36 oxenfree joined #git
21:37 oxenfree hey guys, one of my repos was marked as a "Roff" project, does anyone know what files to exclude in gitattributes?
21:37 tymczenko joined #git
21:37 oxenfree having trouble googling around
21:37 hseg BTW, I notice that the urls given by git remote show origin and git config --get remote.origin.url differ. Any flags I can pass to git remote to get the result from git config?
21:38 hseg Specifically, git remote expands insteadOf directives.
21:38 acresearch hseg: was git updated recently? all youtibe videos do not highlight the username thing
21:38 Gustavo6046_ joined #git
21:39 hseg acresearch: No, AFAIK this was in the original design for git.
21:39 UTAN_dev_ joined #git
21:39 ki0_ joined #git
21:39 jrm2 joined #git
21:39 hseg For example, one of the git commands "git-blame" lists who wrote each line in a file.
21:39 acresearch hmmm
21:39 Nugget_ joined #git
21:39 hseg This is useful for attributing the sources of bugs.
21:39 acresearch ok
21:40 mage__ joined #git
21:40 psprint_ joined #git
21:41 _moogs_ joined #git
21:41 mmattice_ joined #git
21:41 theiostr_ joined #git
21:41 kenansulayman joined #git
21:41 miczac_ joined #git
21:42 rebrec_ joined #git
21:42 Loomaanaatii_ joined #git
21:42 Mir4g3_ joined #git
21:42 Calinou_ joined #git
21:42 Celelibi_ joined #git
21:42 Daviey_ joined #git
21:42 troyt_ joined #git
21:42 Jguy| joined #git
21:42 leonarth_ joined #git
21:42 skalpin1 joined #git
21:42 glebihan_ joined #git
21:42 pepijndevos joined #git
21:42 vishal_ joined #git
21:42 FrozenFire[alt] joined #git
21:42 jon-mac joined #git
21:42 Peng joined #git
21:42 y_sekino joined #git
21:42 Peng joined #git
21:42 markovh joined #git
21:43 markmarkmark joined #git
21:43 ewitz joined #git
21:43 danmackay_ joined #git
21:43 ski4x7 joined #git
21:43 holdsworth joined #git
21:43 polymorphisme joined #git
21:43 markovh joined #git
21:43 polyzen joined #git
21:44 evil-god joined #git
21:44 Learath2 joined #git
21:44 strugee joined #git
21:44 nanga joined #git
21:44 nanga joined #git
21:45 blackwind_123 joined #git
21:45 Kamilion joined #git
21:45 db joined #git
21:46 alexandre9099 joined #git
21:46 incredible joined #git
21:48 angrybac_ joined #git
21:50 jaeckel joined #git
21:50 Marumarsu joined #git
21:54 askb joined #git
21:54 di0_ joined #git
21:57 arooni is there a shortcut like git add -u for newly created files ?
22:01 re1 joined #git
22:02 tourdownunder joined #git
22:04 miha_S7_ joined #git
22:05 menip joined #git
22:05 hseg left #git
22:06 Cabanossi joined #git
22:07 transhuman whats wrong with adding paths to the file like so with this command git rev-list --reverse -- net/netfilter/xt_bpf.c | xargs -n1 -I@ git format-patch -n1 @ -- /net/netfilter/xt_bpf.c
22:07 transhuman trying to avoid duplicate named files from being picked up
22:08 transhuman at the same time this works git rev-list --reverse -- xt_bpf.c | xargs -n1 -I@ git format-patch -n1 @ -- xt_bpf.c
22:08 transhuman but I think it has a problem with duplicates
22:09 Cabanossi joined #git
22:10 diriger joined #git
22:21 tourdownunder joined #git
22:21 MattMaker joined #git
22:25 Schrostfutz thiago: Thanks, --onto did exactly what I wanted
22:27 romank joined #git
22:28 rmbeer joined #git
22:29 tmsmith_ joined #git
22:31 nutron joined #git
22:32 MarioBranco joined #git
22:32 angrybacon joined #git
22:35 h12o joined #git
22:35 thebope joined #git
22:37 anuxivm joined #git
22:40 pharma_joe joined #git
22:43 Gsham joined #git
22:45 watabou joined #git
22:50 clickjack joined #git
22:51 fission6 joined #git
22:52 ahrs joined #git
22:52 rxc joined #git
22:52 rxc joined #git
22:56 shgysk8zer0 joined #git
23:00 SkyPatrol joined #git
23:02 h12o joined #git
23:04 theoceaniscool joined #git
23:04 miha_S7_ joined #git
23:07 SkyPatrol joined #git
23:07 TheSimonator joined #git
23:11 Levex joined #git
23:11 kexmex joined #git
23:11 aidalgol joined #git
23:11 h12o joined #git
23:15 SkyPatrol joined #git
23:17 CDanU I know that I can use 'git stash -k -u' to stash unstaged and untracked changes. Is it somehow possible to only stash unstaged and untracked files with a specific file extension and/or in a specific directory?
23:19 sonOfRa Not sure if git stash takes a pathspec, but if it does, "git stash -k -u -- *.ext" should do the trick
23:20 h12o joined #git
23:21 tmsmith joined #git
23:22 CDanU thx, I am going to do a test case, looks good so far, no complains from git for the command itself.
23:24 tourdownunder joined #git
23:25 jaziz joined #git
23:26 theskillwithin im taking this nodejs course, and the course files are on git,  i want to be able to pull if they change something but also to be able to push to my own github, is this possible?
23:29 theskillwithin ahh
23:29 theskillwithin git remote add NAME url
23:29 theskillwithin git push -u NAME branch
23:30 h12o joined #git
23:30 CDanU sonOfRa: yes it works flawlessly :), tig seems to have problems showing the untracked files
23:30 transhuman whats wrong with adding paths to the file like so with this command git rev-list --reverse -- net/netfilter/xt_bpf.c | xargs -n1 -I@ git format-patch -n1 @ -- /net/netfilter/xt_bpf.c
23:31 transhuman if I dont include the paths it works...sorry for the double post but noone was around
23:34 CDanU transhuman: 'git rev-list [OPTION] <commit-id>' I don't see the <commit-id>
23:34 theskillwithin tig <3
23:35 theskillwithin i like that its git backwards
23:36 stepho joined #git
23:36 M6HZ joined #git
23:36 TheSimonator joined #git
23:39 transhuman CDanU I thought it tracked by the file name not by the commit-id...I have a list of all the files in the history that ever were git log --pretty=format: --name-only --diff-filter=A | sort -u | tee list-of-files.txt I would like to find the complete history of changes to the files ..is that possible? I thought that command could specify a path to  a particular file to track
23:39 skered joined #git
23:40 transhuman the other problem is what happens when they change location or are renamed
23:40 stepho_ joined #git
23:41 transhuman and what if they have duplicate names in the tree
23:41 Ardethian joined #git
23:41 publicarray joined #git
23:41 CDanU transhuman: you could try HEAD as commit id, git can track simple renames, but yes this will be a problem. How do you get duplicate names?
23:43 transhuman I assume in the torvals/linus tree that duplicate file names exist ... I might be wrong but see this as a potential problem
23:43 transhuman I am using all this data to train a neural net
23:44 transhuman or at least thats my intention
23:45 CDanU transhuman: I personally can't really help with your problem, but I am interested to hear what you want to achieve
23:47 transhuman I am creating a coding sentiment with the history (developed by the changes over time (arrow points toward enhanced development and better code (but not always) and combining this sentiment with code synthesis
23:47 transhuman also I think I am going to add in sentiments on pieces of code developed by looking at lkml.org data
23:48 transhuman their by hopefully making a better trained neural net that actually can make sensible code
23:49 transhuman s/their by/there by
23:50 CDanU as in making existing code better or making good code out of thin air based on the training
23:50 transhuman existing code better
23:50 bgerber joined #git
23:51 transhuman I dont think without adding in a mathematical representation /picture of the code that you could actually synthesis new code
23:54 kyan joined #git
23:54 CDanU ok I am not sure how you could handle the renames, maybe something that is used in image processing (sum of squares?) mapped onto you code? ascii could be a nice mono img :). I am not sure on this but there should not be duplicate files at a time. You could detect when a file is being deleted and than added again.
23:54 transhuman perhaps I should rephrase that as sudo code
23:55 transhuman pseudo
23:55 transhuman how would I detect that CDanU
23:57 jstimm joined #git
23:57 tunekey joined #git
23:57 transhuman git log --full-history -- your_file
23:59 pharma_joe joined #git

| Channels | #git index | Today | | Search | Google Search | Plain-Text | summary