Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-12-30

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

All times shown according to UTC.

Time Nick Message
00:00 arescorpio joined #git
00:02 ochorocho__ joined #git
00:05 dsdeiz joined #git
00:08 hahuang61 joined #git
00:11 hanneso joined #git
00:11 howitdo joined #git
00:11 mizu_no_oto joined #git
00:15 PHPanos joined #git
00:15 WaReZ joined #git
00:16 howitdo joined #git
00:17 m0viefreak joined #git
00:21 Tobbi joined #git
00:28 xaviergmail joined #git
00:31 mentazoom joined #git
00:31 Achylles joined #git
00:34 cdg joined #git
00:42 jeffreylevesque joined #git
00:51 Derperperd joined #git
00:53 Derperperd joined #git
00:54 bronson joined #git
00:55 clemf_ joined #git
00:55 jagob joined #git
00:57 thebope joined #git
00:58 Schrostfutz joined #git
00:59 Garen joined #git
01:07 ToBeCloud joined #git
01:08 Garen joined #git
01:12 doener joined #git
01:14 weckstrom joined #git
01:23 thebope joined #git
01:23 Mutter joined #git
01:29 clemf joined #git
01:31 Random832 joined #git
01:35 cdg joined #git
01:42 snowkidind joined #git
01:43 grekkos joined #git
01:46 grekkos I'm writing a post-commit script to sync the commits of my smaller project into a subdirectory of a larger project for the benefit of being able to have the smaller project repo separate and cleaner for dev, but to have it all together with a bunch of other projects in a larger repository. I'm having some strange issues running a specific git command to get the current branch name in the larger repo, it's failing the git command with a fatal error: not a g
01:46 grekkos I can't figure out why it thinks that the directory is not a git repo. The command works fine on the command line in my interactive environment
01:46 SpeakerToMeat joined #git
01:49 mostlybadfly joined #git
01:51 bipolar joined #git
01:53 mizu_no_oto joined #git
01:55 bronson joined #git
01:56 grekkos oh I figured it out
01:56 llamapixel grekkos:  change directory mismatch?
01:58 Nilesh_ joined #git
01:59 morenoh149 joined #git
02:02 dling joined #git
02:03 ochorocho__ joined #git
02:04 Limes_ joined #git
02:06 dave0x6d joined #git
02:08 NeverDie joined #git
02:09 Salander27_ joined #git
02:10 hahuang61 joined #git
02:13 malide joined #git
02:14 aw1 joined #git
02:15 finalbeta joined #git
02:16 Tobarja joined #git
02:19 pks joined #git
02:19 cyphase joined #git
02:20 andrew710 joined #git
02:21 unreal joined #git
02:27 dfas joined #git
02:27 d^sh joined #git
02:32 cagedwisdom joined #git
02:33 gtristan joined #git
02:34 xaviergmail joined #git
02:35 cdg joined #git
02:38 Vampire0 joined #git
02:40 e14 joined #git
02:46 cyphase joined #git
02:49 ilbot3 joined #git
02:49 Topic for #git is now Welcome to #git, the place for git help and serious business | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.11.0 | Getting "cannot send to channel"? /msg gitinfo .voice | The git-jokes project has been suspended for lack of contribution. Help revive it! https://madeitwor.se/git-jokes
02:49 cyphase joined #git
02:49 cyphase joined #git
02:50 cyphase joined #git
02:53 GavinMagnus left #git
02:54 looper joined #git
02:56 bronson joined #git
02:57 buffal0 joined #git
02:57 nowhere_man joined #git
02:59 PaulCapestany joined #git
03:03 nckpz joined #git
03:06 aavrug joined #git
03:07 aavrug joined #git
03:07 cyphase joined #git
03:10 Lyqyd joined #git
03:12 Goplat joined #git
03:14 [log] joined #git
03:14 rivarun joined #git
03:15 Darren_ joined #git
03:16 muuse joined #git
03:16 salamanderrake joined #git
03:17 yanome joined #git
03:18 LuckyStewie_ joined #git
03:18 Vortex35 joined #git
03:19 toothe joined #git
03:19 toothe How would I read the man page for the 'rm' feature?
03:19 preaction git help rm
03:19 bmahe joined #git
03:20 Hathadar joined #git
03:22 safe joined #git
03:23 LuckyStewie_ left #git
03:26 LuckyStewie joined #git
03:26 LuckyStewie Hello All
03:26 gitinfo LuckyStewie: 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.
03:26 LuckyStewie !backup
03:26 gitinfo Worried about your data while trying stuff out in your repo? The repository in its entirety lives inside the .git directory in the root of your work tree so to backup everything `cp -a path/to/workdir path/to/backup` or equivalent will suffice as long as the repo is not modified during backup. See also http://sethrobertson.github.com/GitBestPractices/#backups
03:30 LuckyStewie left #git
03:30 [log] joined #git
03:36 grekkos toothe: or you can use man git-rm to get the actual man page, works on my system at least to use the hyphenated command name
03:36 gitinfo toothe: the git-rm manpage is available at http://jk.gs/git-rm.html
03:37 cyphase joined #git
03:38 hexagoxel joined #git
03:48 morenoh149 joined #git
03:49 sunri5e joined #git
03:52 menip joined #git
03:54 b1tchcak_ joined #git
03:55 raspado joined #git
03:55 pijiu joined #git
03:56 ResidentBiscuit joined #git
03:57 bronson joined #git
04:02 chachasmooth joined #git
04:03 hahuang61 joined #git
04:10 theskillwithin joined #git
04:11 a_thakur joined #git
04:11 a_thakur joined #git
04:11 svm_invictvs joined #git
04:14 fstd_ joined #git
04:28 dirtyroshi joined #git
04:36 Darren_ joined #git
04:43 subhojit777 joined #git
04:53 Darren_ joined #git
04:55 tojo joined #git
04:57 Derperperd joined #git
04:58 sundev70 joined #git
04:58 bronson joined #git
04:59 Emperor_Earth joined #git
05:01 leb joined #git
05:07 jnewt joined #git
05:09 rchavik joined #git
05:10 rchavik joined #git
05:11 rchavik joined #git
05:12 Cabanossi joined #git
05:13 ayogi joined #git
05:13 rchavik joined #git
05:15 lektrik joined #git
05:15 Derperpe1 joined #git
05:20 the_cybersphinx joined #git
05:24 dsdeiz joined #git
05:26 the_cybersphinx joined #git
05:28 bocaneri joined #git
05:29 NeverDie joined #git
05:30 giraffe joined #git
05:34 watabou joined #git
05:35 kaldoran joined #git
05:36 howitdo joined #git
05:37 shubhnik joined #git
05:37 morenoh149 joined #git
05:39 umayr joined #git
05:47 menip joined #git
05:53 cagedwisdom joined #git
05:55 d0nn1e joined #git
05:59 bronson joined #git
06:00 kexmex joined #git
06:05 ochorocho__ joined #git
06:07 [0xAA] joined #git
06:08 [0xAA] joined #git
06:10 duderonomy joined #git
06:11 ogny joined #git
06:16 SpeakerToMeat joined #git
06:17 vuoto joined #git
06:18 dsdeiz joined #git
06:19 Derperperd joined #git
06:21 overlord_tm joined #git
06:33 nedzadhr1jica joined #git
06:44 pks joined #git
06:49 netj joined #git
06:50 ThomasLocke joined #git
06:50 ThomasLocke joined #git
06:51 cyan__ joined #git
06:52 cagedwisdom joined #git
06:56 Raging_Hog joined #git
06:57 jorj joined #git
07:00 bronson joined #git
07:00 morenoh149 joined #git
07:01 cagedwisdom joined #git
07:03 cagedwisdom joined #git
07:06 dreiss joined #git
07:07 cagedwisdom joined #git
07:07 GT4066 joined #git
07:07 cagedwisdom joined #git
07:18 dsdeiz joined #git
07:24 stick-shift joined #git
07:31 zeroed joined #git
07:32 bannakaffalatta joined #git
07:35 marlinc joined #git
07:35 cagedwisdom joined #git
07:36 cagedwisdom joined #git
07:43 lektrik joined #git
07:45 Sceorem joined #git
07:47 Werring joined #git
07:47 [log] joined #git
07:47 iron_houzi joined #git
07:49 netj joined #git
07:55 zzz joined #git
07:56 Singmyr joined #git
07:58 dave0x6d joined #git
07:58 [log] joined #git
08:00 bronson joined #git
08:02 navidr joined #git
08:02 afuentes joined #git
08:03 zeroed joined #git
08:06 ochorocho__ joined #git
08:09 pks joined #git
08:15 xaa__ joined #git
08:16 danslo joined #git
08:17 howitdo joined #git
08:17 cagedwisdom joined #git
08:20 pks joined #git
08:25 wogi joined #git
08:26 wogi_ joined #git
08:37 hahuang61 joined #git
08:39 YuGiOhJCJ joined #git
08:41 anli joined #git
08:41 anli I was changing the url of a git submodule, when my colleague did gid submodule sync, he did not get a new remote, why not?
08:44 Cabanossi joined #git
08:45 Balliad joined #git
08:47 a_thakur joined #git
08:48 a_thakur joined #git
08:49 morenoh149 joined #git
08:50 leftriver95 joined #git
08:53 jikz joined #git
08:55 jozwior_ joined #git
08:56 Mowgli joined #git
08:57 FabTG joined #git
08:57 PCatinean joined #git
09:01 bronson joined #git
09:10 kurkale6ka joined #git
09:12 qpdb joined #git
09:13 cagedwisdom joined #git
09:14 SteffanW joined #git
09:15 Mowgli joined #git
09:23 robotroll joined #git
09:30 Seveas because sync doesn't update urls in the local config. I believe only init does that, but am not a submodule user so don't know it for sure.
09:33 clmsy joined #git
09:37 nitesh` joined #git
09:38 nitesh`` joined #git
09:42 cagedwisdom joined #git
09:42 vrederv joined #git
09:43 vrederv Hey folks can I use gitweb to serve a directory containing multiple repositories?
09:47 vrederv Apparently it works when you initialize a repo in the parent directory.
09:48 marcogmonteiro joined #git
09:49 askb joined #git
09:50 uva I used gitweb a long time ago, and I think it might have been possible (even without that hack)
09:51 rnsanchez joined #git
09:53 vrederv uva, well without the .git repo in the parent dir I get fatal: Not a git repository (or any of the parent directories): .git
09:53 wgrant joined #git
09:53 mgedmin gitweb works for me
09:54 vrederv I didn't say that it doesn't work.
09:54 mgedmin I use Ubuntu's gitweb package, which sets $projectroot = "/var/cache/git";
09:54 mgedmin I don't have a /var/cache/git/.git
09:54 mgedmin I have a bunch of git repos underneath (/var/cache/git/foo.git etc.)
09:56 uva yeah maybe it uses .git as an export trigger; but it probably does not need a "git init"-ed directory
09:56 celebre joined #git
09:56 vrederv mgedmin, setting $projectroot doesn't seem to work for me
09:56 mgedmin (tbh /var/cache seems like a daft place for git repos; I should move them somewhere else ...)
09:57 ochorocho__ joined #git
09:57 chll_ joined #git
10:00 vrederv How do you change gitweb project descriptions?
10:02 bronson joined #git
10:03 King_Hual joined #git
10:07 bl0m1_ joined #git
10:10 howitdo joined #git
10:11 Silox| joined #git
10:12 pajpax joined #git
10:13 nitesh joined #git
10:14 nitesh` joined #git
10:14 aw1 joined #git
10:15 subhojit777 joined #git
10:20 cagedwisdom joined #git
10:21 prawnzy joined #git
10:23 demize The manpage has almost everything you could need.
10:25 pluszak left #git
10:25 mgedmin edit the 'description' file inside reach repo.git/
10:32 [log] joined #git
10:36 tor8 joined #git
10:37 netj joined #git
10:38 morenoh149 joined #git
10:38 hahuang61 joined #git
10:38 tor8 Hi. I have a small feature request, that git submodule init on shallow clones use the --no-single-branch argument to clone. It makes shallow submodules a lot more useful since then you can point to tags, and non-master branches.
10:40 Mowgli joined #git
10:40 Zarthus you probably want feature requests to go to the git issue repository, wherever that may be
10:42 rahtgaz joined #git
10:42 tor8 I haven't found a bug tracker anywhere. Is the mailing list the way to go?
10:43 leeN joined #git
10:43 uva Yes.
10:48 duderonomy joined #git
10:50 nitesh joined #git
10:51 nitesh`` joined #git
10:51 nowhere_man joined #git
10:52 denisMone joined #git
10:53 * mgedmin is constantly getting confused about MSYS vs MSYS2 -- which one is Git for Windows based on?
10:56 shiriru joined #git
10:58 robattila256 joined #git
10:59 shinnya joined #git
11:01 Darcidride joined #git
11:02 demize MSYS2.
11:02 demize Old MSYS is essentially dead.
11:03 bronson joined #git
11:05 zzz joined #git
11:05 PHPanos joined #git
11:06 nowhere_man joined #git
11:09 davimore_ joined #git
11:11 Derperperd joined #git
11:20 nowhere_man joined #git
11:20 byte512 joined #git
11:21 yakaar joined #git
11:22 g_zen joined #git
11:25 rahtgaz joined #git
11:26 nowhere_man joined #git
11:32 rory left #git
11:32 Derperperd joined #git
11:33 cdown joined #git
11:33 oskarkv joined #git
11:37 theoceaniscool joined #git
11:37 Mowgli joined #git
11:45 iamtakingiteasy joined #git
11:45 [log] joined #git
11:45 askb joined #git
11:46 nowhere_man joined #git
11:47 iamtakingiteasy joined #git
11:48 publio joined #git
11:48 thiago joined #git
11:50 ertesx joined #git
11:51 nowhere_man joined #git
11:54 benedikt left #git
11:56 krillr joined #git
11:57 WaReZ joined #git
11:58 zzz joined #git
11:58 Schrostfutz joined #git
11:59 mentazoom joined #git
12:00 dordonez joined #git
12:04 bronson joined #git
12:05 cdown_ joined #git
12:06 CheckDavid joined #git
12:07 rivarun joined #git
12:08 Vortex34 joined #git
12:14 mablae joined #git
12:15 almy joined #git
12:18 andrew joined #git
12:18 batrick joined #git
12:20 subhojit_ joined #git
12:21 cbreak joined #git
12:21 prawnzy joined #git
12:25 manuelschneid3r joined #git
12:26 InfoTest joined #git
12:27 morenoh149 joined #git
12:33 zero85 joined #git
12:34 jagob joined #git
12:35 arunpyasi joined #git
12:36 Oebele joined #git
12:36 qpdb joined #git
12:38 uva left #git
12:39 hahuang61 joined #git
12:39 Seveas mgedmin: git for windows 1.x was based on msys, git for windows 2.x is based on msys2
12:40 * thiago likes msys2 because it comes with pacman
12:41 Seveas I like not having to deal with any of that because I run linux :)
12:41 thiago I like the ability to add stuff to it, as if I were on Linux
12:41 jeffreylevesque joined #git
12:42 buffal0 joined #git
12:43 cdown_ joined #git
12:45 cdown_ joined #git
12:46 buffal0 joined #git
12:47 BlessJah pacman like in archlinux package manager?
12:47 kadoban joined #git
12:47 Seveas it sure as hell isn't a yellow figure eating dots and ghosts...
12:49 * BlessJah wonders how did pacman end up in msys2
12:50 BlessJah oh, because they "wanted a package management system to provide easy installation of packages"
12:51 buffffal00 joined #git
12:52 mentazoom joined #git
12:53 Puffball joined #git
12:55 ochorocho__ joined #git
12:55 SpeakerToMeat joined #git
12:57 glebihan joined #git
13:03 Shock- joined #git
13:04 masuberu joined #git
13:05 bronson joined #git
13:07 blackjid joined #git
13:08 wogi joined #git
13:11 thiago blackjid: yes
13:15 ikelso joined #git
13:21 denisMone joined #git
13:24 ajyg joined #git
13:26 nostart joined #git
13:27 nostart hi to everybody
13:27 gitinfo nostart: 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.
13:28 tor8 left #git
13:32 porsche944 joined #git
13:32 Derperperd joined #git
13:35 WaReZ joined #git
13:37 Phil-Work joined #git
13:38 [log] joined #git
13:40 siddart joined #git
13:41 rwb joined #git
13:44 celebre joined #git
13:46 nettoweb joined #git
13:48 Tee_Pee joined #git
13:57 floatingman joined #git
13:58 siddart joined #git
14:02 jagob joined #git
14:03 Jackneill joined #git
14:05 m0viefreak joined #git
14:06 bronson joined #git
14:09 kexmex joined #git
14:13 ikelso joined #git
14:16 morenoh149 joined #git
14:16 shmup thiago: msys2/git is garbage i discovered with large git repositories. did you have that experience?
14:16 shmup like i had gitbash open and mysys2 and the difference was staggering
14:19 cmn joined #git
14:20 jnavila joined #git
14:21 sdothum joined #git
14:22 prawnzy joined #git
14:25 delboy1978uk joined #git
14:25 delboy1978uk how do i clean all the crap out of my working branch? tried git clean -f but untracked nonsense is still there
14:25 ShekharReddy joined #git
14:27 elsevero joined #git
14:27 thiago shmup: it's slow
14:27 ikelso joined #git
14:27 thiago delboy1978uk: git clean deletes all untracked (but not ignored) files
14:28 thiago delboy1978uk: you may want to add the -d and -x switches
14:31 mrxevious0x joined #git
14:31 todd_dsm joined #git
14:32 almy joined #git
14:32 Khisanth joined #git
14:34 kexmex joined #git
14:40 hahuang61 joined #git
14:44 Tobbi joined #git
14:45 Es0teric joined #git
14:50 JanC_ joined #git
14:50 siddart joined #git
14:51 Balliad joined #git
14:53 jason237 joined #git
14:53 guest joined #git
14:54 danslo joined #git
14:54 xaa__ joined #git
14:55 PaulCapestany joined #git
14:55 ochorocho__ joined #git
14:55 mrxevious0x left #git
14:56 guest_ joined #git
14:57 zzz joined #git
14:57 Es0teric joined #git
14:59 kexmex joined #git
14:59 ayogi joined #git
15:02 thebope joined #git
15:05 diamvw joined #git
15:05 diamvw left #git
15:05 diamvw joined #git
15:06 porsche944 joined #git
15:07 bronson joined #git
15:08 Derperperd joined #git
15:12 WaReZ joined #git
15:14 bu5hm4n joined #git
15:14 shinnya joined #git
15:14 bu5hm4n hello, can i make the submodules init automatically on clone ?
15:14 delboy1978uk left #git
15:15 cbreak bu5hm4n: --recursive
15:15 cdg joined #git
15:15 bu5hm4n cbreak, thank you! :)
15:18 ShapeShifter499 joined #git
15:20 almy_ joined #git
15:21 cdown joined #git
15:22 almy_ joined #git
15:22 almy_ left #git
15:23 frobnic joined #git
15:23 almy joined #git
15:26 jagob joined #git
15:28 Panacea joined #git
15:29 morenoh149 joined #git
15:32 ertes say i have two commits (C <- B <- A) and (D <- A) where "<-" means "child of"…  if i cherry-pick B and C at D i get (C' <- B' <- D <- A)…  if i now rebase -i such that i arrive at (D' <- C'' <- B'' <- A), can i expect that B'' = B and C'' = C?
15:33 cdown_ joined #git
15:33 ertes i would expect so, because i would consider that equivalent to a regular rebase
15:35 doener joined #git
15:35 svm_invictvs joined #git
15:39 subhojit_ joined #git
15:41 Darcidride joined #git
15:42 dviola joined #git
15:42 PaulCapestany joined #git
15:42 jnavila joined #git
15:45 cbreak no
15:49 denisMone joined #git
15:51 Panacea joined #git
15:52 NeverDie joined #git
15:53 WaReZ joined #git
15:54 shubhnik joined #git
15:55 ertes hmm…  so reordering via rebase -i, and then restoring the original order via rebase -i does not necessarily yield the same thing?  (assume no conflicts)
15:55 Tobbi joined #git
15:56 e14 joined #git
15:57 davimore__ joined #git
15:57 cbreak ertes: indeed
15:57 cbreak ertes: don't forget that commits also contain meta data such as the committer and author, and associated dates
15:58 ertes cbreak: and those change?
15:58 cbreak when ever you commit, the current date is used for committer date
15:58 cbreak and cherry pick / rebase creates commits
15:59 ertes cbreak: i see…  and those are used for hash calculation?
15:59 cbreak if the previous committer was not you, then the commmitter also changes, to you
15:59 cbreak of course they are
15:59 ertes ok, thank you!
15:59 gldn joined #git
15:59 cbreak if they weren't, then the whole cryptographic verifiability would be even more worthless than it already is :)
15:59 gldn left #git
15:59 ertes yeah, i suppose =)
16:00 nettoweb joined #git
16:04 martinjlowm joined #git
16:04 eihabarabia joined #git
16:06 nettoweb1 joined #git
16:08 bronson joined #git
16:11 ogny joined #git
16:12 danslo joined #git
16:12 [log] joined #git
16:13 leftriver51 joined #git
16:17 bu5hm4n joined #git
16:17 leftriver51 Hi folks! Looking for a little assistance in understanding the interaction between git commit --only and pre-commit hooks.
16:18 leftriver51 I have a pre-commit hook that reformats any staged files and re-adds them to the index prior to the commit. Everything works fine when using git commit without the --only flag, but using --only yields unexpected results.
16:20 cdg joined #git
16:20 al-damiri joined #git
16:20 joki joined #git
16:22 leftriver51 When using --only, after the commit, the reformatted code is committed (which is expected), but the poorly-formatted code is staged and the reformatted code is listed as modified in the working tree.
16:23 prawnzy joined #git
16:23 quizzi joined #git
16:24 cbreak leftriver51: sounds normal
16:24 cbreak leftriver51: your hook probably fucks up the working dir
16:24 Akki2200 joined #git
16:25 rivarun joined #git
16:25 leftriver51 cbreak: thanks. is there a way to use a pre-commit hook modify content to be included in a commit in the context of git commit --only?
16:26 cbreak it sounds bad to even mess with anything in a hook
16:26 cbreak why don't you do it properly?
16:26 nicksloan joined #git
16:26 cbreak leftriver51: man gitattributes
16:26 gitinfo leftriver51: the gitattributes manpage is available at http://jk.gs/gitattributes.html
16:26 cbreak look for smudge / clean filters
16:27 snowkidind joined #git
16:27 guest_ joined #git
16:28 leftriver51 cbreak: thanks, will take a look. would you recommend just running tests / linters in pre-commits and exiting nonzero if they fail, and leaving formatting concerns aone?
16:28 leftriver51 *alone
16:28 cbreak yes
16:29 Tobbi joined #git
16:29 jimi_ joined #git
16:33 leftriver51 cbreak: can you recommend any documentation on how git commit --only interacts with the index? the official documentation is light on details
16:34 cfoch joined #git
16:34 cbreak no idea, sorry
16:34 cfoch Hi, I am applying a patch, but it doesn't apply
16:34 cfoch https://paste.fedoraproject.org/515118/11565214/
16:34 cbreak but it seems like it doesn't touch the index at all
16:34 cbreak or if it touches it, it restores it to the original state
16:34 cfoch How can I force it, so I resolve conflicts?
16:36 guest_ joined #git
16:38 e14 joined #git
16:39 Cassiopaya joined #git
16:40 hahuang61 joined #git
16:41 miguelfernandes joined #git
16:41 ptmiguel joined #git
16:42 IlNominePatre joined #git
16:43 leftriver51 cbreak: Yeah, I don't understand it either. I'd expect that files staged during a pre-commit hook for a --only commit would be rolled into that commit and would be clean once the commit completes.
16:43 cbreak leftriver51: I wouldn't
16:43 narendraj9 joined #git
16:43 leftriver51 Why not?
16:43 cbreak I would expect the working dir state to be committed, as described in the man page
16:43 cbreak not the index state
16:44 leftriver51 Ok, sure. I also tried just changing the working dir state and *not* staging the affected file(s), and they end up committed in their prior-to-pre-commit state with the changes left as modifications to the working tree.
16:45 ogny joined #git
16:45 leftriver51 If --only used working dir state, I'd think that any pre-commit modifications would make it in.
16:48 Lyqyd joined #git
16:50 cbreak leftriver51: it uses working dir state
16:50 frobnic joined #git
16:50 cbreak leftriver51: your hook just runs too late to capture this :)
16:52 leftriver51 That doesn't make sense to me, though, because the changes imposed by the hook do end up in the commit. If they were too late, I wouldn't expect that to happen.
16:53 Darren_ joined #git
16:54 leftriver51 (FWIW, I agree that smudge/clean looks like a better way of doing this, but at this point I'm just trying to understand the observed behavior.)
16:56 ochorocho__ joined #git
17:00 imachuchu joined #git
17:02 cfoch how do I apply a very old patch?
17:02 cbreak leftriver51: no
17:02 cbreak what happens is that git prepares the index for you
17:02 cbreak then, when it is done, it restores the original index
17:02 cfoch (it has conflicts due to a lot of new code introduced)
17:02 cbreak probably :)
17:03 gtristan joined #git
17:03 cbreak so if  you mess with the working dir, it won't affect the commit, but if you mess with the index before the commit, it won't end up in the commit either
17:03 theoceaniscool joined #git
17:03 cbreak cfoch: apply it to the original base
17:03 cbreak after that you can cherry-pick it
17:04 jccn joined #git
17:05 madewokherd joined #git
17:05 leftriver51 cbreak: what you describe is the behavior I would expect, but not the behavior I see. The modifications from the pre-commit hook are part of the --only commit.
17:06 dave0x6d joined #git
17:07 cbreak leftriver51: only if you modify the index, right?
17:07 leftriver51 as part of the hook, yes
17:07 cbreak so it is exactly how I expect
17:08 cbreak git backs up the index
17:08 Ranhir joined #git
17:08 cbreak git stages the file contents
17:08 cbreak git calls your hook
17:08 cbreak git creates the commit
17:08 cbreak git restores the original index
17:08 _ADN_ joined #git
17:08 cbreak if you mess with the file contents, then your changes are now still there
17:08 cbreak if you mess with the index, your changes are no longer in the index
17:09 bronson joined #git
17:09 cbreak everything sounds reasonable
17:09 cbreak -> don't change things in commit hooks :)
17:10 howitdo joined #git
17:11 leftriver51 That mostly makes sense. It sounds, though, like the call to "git add" during the hook is affecting both the temporary index and the backed-up index, adding the modified file to the temporary index and the unmodified file to the backed-up index
17:11 cbreak leftriver51: why do  you think that?
17:12 leftriver51 Well, I know that it's adding the modified file to the temporary index because the modifications make it into the commit, but after the commit is complete, the unmodified file is staged and the modified version is in the working tree
17:12 cbreak leftriver51: so?
17:12 cbreak the unmodified file was already staged before the whole thing started
17:12 leftriver51 no, it wasn't
17:13 cbreak of course it was
17:13 cbreak does it show up in git status?
17:13 leftriver51 as modified but unstaged
17:13 cbreak then the unmodified version is obviously staged.
17:14 cbreak if it was not staged, it'd show up as new file
17:14 cbreak (untracked)
17:15 Noldorin joined #git
17:15 cbreak leftriver51: if a file exists in the most recent commit, but is not staged, it shows up in git status as removed
17:16 cbreak leftriver51: if a file exists in the most recent commit, and is staged in that version, and the working dir file is also the same, then it does not show up in git status because it is unmodified
17:16 leftriver51 In other words, all unchanged files that are part of the most recent commit are implicitly staged?
17:18 cbreak explicitly
17:18 cbreak the staging area / index contains the next commit (unless you use things like --only)
17:18 shubhnik joined #git
17:18 cbreak after a normal commit, it contains the most recent commit (the one just created)
17:18 leftriver51 right
17:19 cfoch How do I checkout 3 commits after <commit_number>?
17:19 cbreak if the index is empty, every file will show up as removed
17:19 cbreak cfoch: manually find the commit you want to check out
17:19 cfoch cbreak: it is a very old commit!
17:19 cbreak git history is a DAG, the D stands for Directed
17:19 cbreak you can only go into the direction of parents
17:20 Atm0spher1c joined #git
17:20 kadoban cfoch: If you know a ref that contains in, do a long --one-line and look N lines up from the commit hash you know.
17:20 kadoban s/long/log/
17:20 cdg joined #git
17:21 kadoban If you don't ... do whatever that thing is that tells you refs that contain a commit first.
17:21 * kadoban runs away
17:22 Schrostfutz joined #git
17:22 leftriver51 cbreak: Okay, let me rephrase. "It sounds, though, like the call to "git add" during the hook is affecting both the temporary index and the backed-up index, adding the modified, hook-formatted changes to the temporary index and staging the modified, non-hook-formatted changes to the backed-up index."
17:22 joki joined #git
17:22 b1tchcakes joined #git
17:23 cbreak leftriver51: no
17:23 cbreak it sounds like it doesn't affect the backed up index at all
17:23 cbreak leftriver51: according to your description, the index before the commit and after contains the same thing
17:23 mario__ joined #git
17:23 leftriver51 I apologize for not using precise language.
17:24 hahuang61 joined #git
17:25 leftriver51 After the commit (which contains the hook-enforced modifications), git status shows changes to be committed, as well as changes not staged for commit, for the same file.
17:25 leftriver51 The changes to be committed are incorrectly formatted, and the changes not staged for commit are correctly formatted.
17:26 leftriver51 That's why I said that it seems that "git add" during the commit is affecting both the temporary and the backed-up indices.
17:27 leftriver51 s/commit/pre-commit hook/
17:27 ikelso joined #git
17:28 cbreak leftriver51: but git add didn't affect the backed up index
17:28 leftriver51 cbreak: it seems to me that it did, because after the commit, there are staged changes that weren't staged prior to the commit.
17:28 cbreak the changes to be committed (the index) still contain the incorrectly formated) version
17:28 cbreak just as before
17:29 cbreak so nothing changed
17:29 ToBeCloud joined #git
17:29 leftriver51 But prior to the commit, those changes weren't staged, and after the commit, they were.
17:30 cbreak leftriver51: git status before the commit showed changes ?
17:30 cbreak (as changes that are not to be committed)?
17:30 leftriver51 git status before the commit showed changes not staged for commit
17:30 cdg joined #git
17:31 mizu_no_oto joined #git
17:31 leftriver51 git status after the commit (using --only) showed those same changes staged for commit
17:31 cbreak maybe git commit with --only staged them before the backup
17:33 leftriver51 Maybe? This is why I'm confused.
17:33 leftriver51 Brb
17:34 dreiss joined #git
17:35 NeverDie joined #git
17:37 leftriver25 joined #git
17:39 cagmz joined #git
17:39 ogny joined #git
17:40 e14 joined #git
17:41 cdg joined #git
17:43 prawnzy joined #git
17:44 mvensky joined #git
17:48 hamster joined #git
17:49 tang^ joined #git
17:49 hanthings joined #git
17:50 mvensky joined #git
17:53 jagob joined #git
17:53 Noldorin joined #git
17:57 mvensky joined #git
17:57 cdown joined #git
17:59 mvensky joined #git
17:59 ochorocho__ joined #git
18:01 mvensky joined #git
18:02 King_Hual joined #git
18:04 duderonomy joined #git
18:06 Vangelis joined #git
18:06 ace_33 joined #git
18:07 kexmex joined #git
18:09 bronson joined #git
18:11 netj joined #git
18:15 Atm0spher1c joined #git
18:15 netj joined #git
18:16 leftriver51 cbreak: back.
18:16 aw1 joined #git
18:16 theoceaniscool joined #git
18:16 mvensky joined #git
18:17 Synthead joined #git
18:17 theoceaniscool joined #git
18:18 mvensky joined #git
18:19 Balliad joined #git
18:20 theoceaniscool joined #git
18:20 prawnzy joined #git
18:21 ayogi joined #git
18:22 jimi_ joined #git
18:25 Es0teric joined #git
18:26 mvensky joined #git
18:30 cdown_ joined #git
18:32 xaviergmail joined #git
18:32 mvensky joined #git
18:34 NeverDie joined #git
18:35 boombatower joined #git
18:38 mvensky joined #git
18:39 Balliad joined #git
18:39 satifant joined #git
18:40 fahadash joined #git
18:40 Balliad joined #git
18:44 cyan__ joined #git
18:45 mvensky joined #git
18:47 Phrohdoh_ joined #git
18:48 Phrohdoh_ Is there any mechanism to force git-hooks on a project? For example setting the hook directory to <root>/git-hooks/ which would, for example, contain `pre-push`.
18:49 porsche944 joined #git
18:52 rewt Phrohdoh, not in git itself; that's more of a policy thing
18:52 guest joined #git
18:52 phroa you can change the core.hooksPath config variable, but that too has to be done by the user at some point
18:53 Phrohdoh Yeah, that's a shame.
18:53 phroa if you control the system image you could theoretically edit /usr/share/git/hooks/<x>
18:53 celebre joined #git
18:53 phroa er, /git-core/templates/hooks rather
18:53 cagedwisdom joined #git
18:56 thiago Phrohdoh: you can modify your git init or git clone builtin to install your hooks for you
18:56 thiago you cannot force other people to do the same
18:57 thiago see /usr/share/git-core/templates
18:57 ochorocho__ joined #git
18:59 mvensky joined #git
19:00 Schrostfutz joined #git
19:00 Phrohdoh Ok, thanks everyone.
19:00 Phrohdoh left #git
19:01 Vampire0_ joined #git
19:02 NwS joined #git
19:02 Es0teric joined #git
19:07 BSaboia joined #git
19:10 bronson joined #git
19:14 jjasinski joined #git
19:15 eblake left #git
19:22 BSaboia joined #git
19:26 duderonomy joined #git
19:28 pajpax left #git
19:33 duderonomy joined #git
19:34 cagedwisdom joined #git
19:35 PHPanos joined #git
19:35 ikelso joined #git
19:40 tyreld joined #git
19:43 ixjlyons joined #git
19:44 ixjlyons left #git
19:44 theoceaniscool joined #git
19:47 sparr joined #git
19:49 Mobutils joined #git
19:51 mehola joined #git
19:54 Zkqp joined #git
19:55 Spec-Chum joined #git
19:56 gmacisme joined #git
19:56 jimi_ joined #git
19:59 glebihan joined #git
20:01 aidalgol joined #git
20:03 cyphase joined #git
20:04 Vinnie_win joined #git
20:05 NeverDie joined #git
20:05 dendazen joined #git
20:08 robotroll joined #git
20:08 ignarps joined #git
20:09 dlam joined #git
20:10 dlam is there a thing to prevent a checkin when you `git push`?  Like i want it to error if i ever checkin a file "settings.py" and it has  "DEBUG = False"
20:11 dlam my goal is trying to prevent something accidently pushed
20:11 bronson joined #git
20:11 rewt dlam, have a look at pre-commit hooks
20:12 rewt dlam, there's also server-side hooks you can have that refuse pushes, but then the pusher has to rewrite their history to be able to push
20:12 looper joined #git
20:14 cagedwisdom joined #git
20:18 jccn joined #git
20:18 svm_invictvs joined #git
20:18 Derperperd joined #git
20:19 zzz joined #git
20:20 vdv joined #git
20:23 PHPanos joined #git
20:23 rockfruit hi.  I'm doing a big merge (like three hundred commits divergence).  (yes, thats my own shitty fault, shruggie). I use meld, and an editor, and I re-run meld after I've 'ported' some functionality.  How do you guys handle these kind of shambolic week-long merges?
20:24 rockfruit It's a lot of fun, to be sure.. but I always wonder if there are "nicer" tools.
20:24 AlineGomes joined #git
20:30 e14 joined #git
20:31 iale joined #git
20:35 Kwinp joined #git
20:40 khagler joined #git
20:41 Pseudocrat joined #git
20:44 ochorocho__ joined #git
20:46 imMute joined #git
20:53 d0nn1e joined #git
20:54 Tobbi joined #git
20:54 Tobarja joined #git
20:59 dlam rewt:  ooo thanks thanks
21:01 jimi_ joined #git
21:03 mentazoom joined #git
21:04 romerocesar joined #git
21:08 rossome joined #git
21:09 slappymcfry joined #git
21:11 imachuchu is it possible to add a submodule with a relative path to a remote? If my cloned origin is "git+ssh://myname@repository.company.com/gitprojects/main_project" is there something like "git+ssh://../other_project" I can use to create a submodule?
21:11 celebre joined #git
21:11 Sasazuka joined #git
21:11 howitdo joined #git
21:11 imachuchu since myname is in the url I can't just embed it into the submodule, else it won't work for anyone else
21:12 bronson joined #git
21:13 imachuchu yes, yes I can. It's just "git add ../other_project" (since main_project is my origin)
21:16 [cust] joined #git
21:20 svm_invictvs joined #git
21:22 roxlu_ joined #git
21:22 roxlu_ hey when I use `git clone --depth 1` is there still a way to get a specific remote branch?
21:26 mentazoom joined #git
21:28 matsaman joined #git
21:31 xaviergmail joined #git
21:32 Balliad joined #git
21:32 [cust] joined #git
21:32 xaa joined #git
21:35 WaReZ joined #git
21:36 rossome joined #git
21:37 pikajude joined #git
21:38 elsevero joined #git
21:39 buzzLightBeer joined #git
21:40 inovas joined #git
21:44 ajf- joined #git
21:44 cdg joined #git
21:46 tang^_ joined #git
21:48 btree joined #git
21:52 gmacisme joined #git
21:57 slappymcfry joined #git
22:01 Tobbi joined #git
22:03 [log] joined #git
22:03 hanthings joined #git
22:04 Akki2200 joined #git
22:04 nicksloan joined #git
22:05 btree left #git
22:07 jagob joined #git
22:10 hahuang61 joined #git
22:12 dreiss joined #git
22:12 [cust] joined #git
22:13 diogenese joined #git
22:13 bronson joined #git
22:16 jnavila joined #git
22:17 kulelu88 joined #git
22:17 sarbs_ joined #git
22:20 dave0x6d joined #git
22:21 romerocesar joined #git
22:21 mentazoom joined #git
22:23 porsche944 joined #git
22:28 silnp joined #git
22:33 prg3 joined #git
22:33 matsaman joined #git
22:34 vdv joined #git
22:34 prg3_ joined #git
22:34 tterrag joined #git
22:41 [log] joined #git
22:41 xaviergmail joined #git
22:47 ochorocho__ joined #git
22:52 newbie1 joined #git
22:52 cyan__ joined #git
22:54 mentazoom joined #git
23:01 dsdeiz joined #git
23:01 fmeerkoetter joined #git
23:03 rwb joined #git
23:03 jstein_ joined #git
23:07 MaximusColourum joined #git
23:11 Eryn_1983_FL joined #git
23:11 Derperperd joined #git
23:14 bronson joined #git
23:17 Balliad joined #git
23:20 jottr joined #git
23:23 YuGiOhJCJ joined #git
23:24 xaviergmail joined #git
23:26 LaptopBrain joined #git
23:26 kyan joined #git
23:28 WayToDoor joined #git
23:32 Balliad joined #git
23:36 danielmahon joined #git
23:36 diogenese joined #git
23:38 LaptopBrain joined #git
23:38 DieguezZ joined #git
23:47 prg3 joined #git
23:52 LaptopBrain joined #git
23:54 SuperCat left #git
23:55 coasterz joined #git

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