Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-09-16

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

All times shown according to UTC.

Time Nick Message
00:00 SonOfLilit I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv
00:00 SonOfLilit any idea how to debug this?
00:00 SonOfLilit (I'm trying to set up textconv or an external differ for csv files)
00:01 Ryanar joined #git
00:02 karstensrage joined #git
00:04 mingrammer joined #git
00:06 MaddHatter left #git
00:09 tesuji joined #git
00:09 kaldoran joined #git
00:11 kyan_ joined #git
00:13 syg_ joined #git
00:13 everamzah joined #git
00:14 BinaryBench joined #git
00:14 inflames joined #git
00:14 SonOf joined #git
00:16 BinaryBench Hey, I'm having an issue with git ignore, I have this in my git ignore file: Servers/, !/Servers/Game/eula.txt
00:16 BinaryBench but when I try to add that file it says it's ignored
00:16 nilg joined #git
00:17 qpdb joined #git
00:20 Rodya_ joined #git
00:22 budre joined #git
00:23 mellernoia joined #git
00:26 SonOf BinaryBench: I assume it's something to do with the slash prefix?
00:27 BinaryBench SonOf: I just noticed that, however fixing doesn't help
00:27 SonOf try simpler: ignore a*, unignore ab
00:27 SonOf does it work?
00:30 BinaryBench SonOf: I'm not sure what you mean?
00:30 mizu_no_oto joined #git
00:30 SonOf lets try to debug this by finding a minimal reproduction
00:31 SonOf in the process we will either understand the problem better or figure out what silly thing we missed
00:31 SonOf so lets try to ignore all files starting with "a" and then specifically allow "ab"
00:31 SonOf probably best to do it in a new, clean git repo
00:31 BinaryBench SonOf: Okie dokie, I'll make a new git project real quick
00:31 inflames joined #git
00:31 meLon joined #git
00:32 Mixologic_ joined #git
00:34 BinaryBench SonOf: here is my test project: http://puu.sh/rcGvn/78a6cbc8bb.zip
00:34 dsdeiz joined #git
00:34 dsdeiz joined #git
00:36 SonOf ok, I see that both are ignored
00:36 SonOf I added what I said - ignore all files with name aaa* but !aaab
00:36 SonOf then I added two files aaaa and aaab
00:37 SonOf and it worked: git status only shows aaab, not aaaa
00:37 SonOf so we've narrowed it to something to do with subdirectories
00:38 SonOf I think I figured it
00:38 SonOf I changed the ignore rule to a/*
00:38 BinaryBench SonOf: Ok, I did that as well and agree.  Any idea why the directory doesn't wrok?
00:38 SonOf and now it shows untracked a/
00:39 SonOf so that's probably the thing - you know how git doesn't look into directories until you add something within to source control?
00:39 BinaryBench I did not know this.
00:39 SonOf lets try that: lets add a file x/y, then add rules x/ and !x/y
00:40 SonOf I at least think git does that, because git status always shows untracked dirs and doesn't walk them to show files inside
00:41 SonOf mm, still doesn't work
00:42 BinaryBench SonOf: Humm, I think you're right.  When I manually add a/do.txt then commit it notices that a/dont.txt is ignored
00:42 SonOf ah, but if we ignore just x/* and not x/ it does work
00:42 e14 joined #git
00:42 prawnzy joined #git
00:43 QwertyDragon joined #git
00:43 BinaryBench SonOf: That's very odd...
00:44 SonOf no, that makes sense
00:44 SonOf if you say "ignore this dir" git doesn't even try to walk it
00:44 actinide joined #git
00:44 BinaryBench SonOf: Hum, I see
00:44 SonOf but if you say "ignore all the stuff in this dir" it does, and then it has a chance to notice that a specific thing should *not* be ignored
00:45 SonOf what git has in abundance, you will find, is internal logic
00:45 SonOf now, I wonder if one of these newcomers happens to know something about the problem *I* came here for...
00:45 SonOf I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv
00:45 SonOf any idea how to debug this?
00:45 prawnzy joined #git
00:45 SonOf (I'm trying to set up textconv or an external differ for csv files)
00:46 BinaryBench I would love to help, but have never used .gitattributes
00:46 libman joined #git
00:47 libman How do I get a list of remote github files without downloading them?
00:48 ejb joined #git
00:49 actinide joined #git
00:50 SonOf libman: I never tried, but from what I know of how git works, I would not expect this to be possible without external tools
00:51 rominronin joined #git
00:51 actinide joined #git
00:51 nettoweb joined #git
00:52 SonOf libman: git is a DAG of objects named by their hash. git network protocols ar designed to synchronize a DAG of objects named by their hash. I don't believe anyone would bother writing special code to synchronize something that isn't the repo, which is a DAG of objects named by their hashes
00:52 BinaryBench SonOf: do you know if there is a way to tell git to iterate through a ignored file?
00:52 YourTheBombNow23 joined #git
00:52 libman Found https://api.github.com/repos/lbmn/nim/contents/
00:52 SonOf BinaryBench: iterate and output...?
00:53 kejoma joined #git
00:53 BinaryBench SonOf: I'm unsure what you mean?
00:53 phatypus joined #git
00:54 phatypus I've just run: git checkout -b feature/abc-123
00:54 phatypus and then made changes. Now I'm ready to push to origin. Considering I'm already on the branch "feature/abc-123", what command do i run to push the branch to origin?
00:54 SonOf ... and he quit
00:54 SonOf BinaryBench: iterate it and do what with it?
00:54 phatypus All examples seem to indicate: git push -u origin feature/abc-123
00:54 phatypus but do i really need to include the branch name if I'm on the branch?
00:55 BinaryBench SonOf: And figure out which ones are "not ignored"
00:55 SonOf phatypus: you need to tell git to which branch on origin to push it. If you don't, there probably is a default of using the same name (try, can't do much damage)
00:56 SonOf BinaryBench: so you mean, iterate through an ignored dir, not an ignored file, right?
00:56 BinaryBench SonOf: correct, my mistake
00:57 SonOf BinaryBench: I already showed you one: tell it "dir/*", or "ignore any path that starts with dir/"
00:57 SonOf BinaryBench: remember that git has no concept of directories, only files and paths
00:58 SonOf joined #git
00:58 mingrammer joined #git
00:59 BinaryBench SonOf: Wait, I'm confused.  I want to ignore all files in Servers/ except !Server/*/file.xyz but when I do status file.xyz doesn't show up.
01:00 SonOfLilit ok, simple steps:
01:00 SonOfLilit 1. add the rules a/* and !a/b
01:00 kingpinlz_ joined #git
01:00 SonOfLilit 2. create files a/b and a/c
01:01 SonOfLilit 3. git add a/b (git will now walk things in a/)
01:01 SonOfLilit done
01:01 arescorpio joined #git
01:01 mehola joined #git
01:01 BinaryBench So there is no way of doing this without having to manually do "add a/b"?
01:01 Zamyatin joined #git
01:02 SonOfLilit to prove this actually works, lets modify it a bit: make the rules a/* and a/b*, then create a/bx, a/by and a/c, and only git add a/bx, git status will show a/by
01:02 SonOfLilit BinaryBench: you need to add one thing once to make git treat the dir as "in the repo"
01:03 SonOfLilit is that a problem?
01:03 BinaryBench SonOfLilit: not a "problem" but kinda annoying.
01:04 BinaryBench Atm I have Servers/Game and Servers/Hub but I might add other servers later.
01:05 BinaryBench And I have !Servers/*/file.xyz
01:05 SonOfLilit I see
01:05 BinaryBench I'd like to be able to just add a server e.g. Servers/Game2/
01:06 SonOfLilit yes, you will always need to do git add Servers/Game2, as far as I understand
01:06 BinaryBench And I'd like to automatically not ignore Servers/Game2/file.xyz without manually adding it once
01:06 SonOfLilit (from the research we did together - I didn't know anything about this in advance)
01:06 Rodya_ joined #git
01:06 SonOfLilit but this would be true even if you didn't have an ignore rule at all, right?
01:07 SonOfLilit in a clean repo, add a a/eula.txt and do git status, git won't show a/eula.txt as untracked, it will show a/ as untracked
01:07 finalbeta joined #git
01:08 DolpheenDream joined #git
01:09 monoprotic joined #git
01:09 monoprotic joined #git
01:09 johnmilton joined #git
01:10 ahr3n joined #git
01:11 pks joined #git
01:11 BinaryBench SonOfLilit: So, if instead of doing: a/, !a/b I did a/c then it does work
01:11 SonOfLilit you still need to git add
01:12 BinaryBench SonOfLilit: But "git add ." works in that case
01:12 SonOfLilit oh, I see
01:12 SonOfLilit let me test, I would expect it to work in both
01:14 SonOfLilit x/*,!x/a*, with files x/aaa,x/aab,x/y, if I git add ., will add x/aaa and x/aab
01:15 SonOfLilit so... I think this is what you want?
01:15 e14 joined #git
01:20 BinaryBench Here: http://puu.sh/rcIMX/1a217d78f6.zip In this example doing "git add ." does not add servers/Game/aaab
01:20 BinaryBench SonOfLilit:
01:20 BackEndCoder joined #git
01:21 SonOfLilit servers/*
01:21 BinaryBench SonOfLilit: that doesn't seem to fix it
01:22 YourTheBombNow23 joined #git
01:22 SonOfLilit to be fully identical to what we did it has to be servers/*/*
01:22 SonOfLilit does servers/ have any files, or only dirs?
01:23 BinaryBench SonOfLilit: hum, ok that works...
01:23 SonOfLilit :)
01:24 SonOfLilit does servers/ have any files, or only dirs?
01:24 SonOfLilit if only dirs, you're lucky and we solved it 100%
01:24 SonOfLilit btw, do you understand why it had to be servers/*/* to mirror the experiment?
01:24 BinaryBench SonOfLilit: only dirs, but what if it did have files? (just curious)
01:25 BinaryBench SonOfLilit: I understand why that mirrors the experiment, I don't understand why it doesn't work with just servers/*
01:25 SonOfLilit then we'd need to improvise to ignore them without ignoring subdirs
01:25 SonOfLilit because we're not allowed to ignore servers/*, as that is like ignoring x/
01:26 SonOfLilit only it says "ignore all x-es"
01:26 SonOfLilit x/-es
01:26 phatypus left #git
01:26 SonOfLilit servers/x/ is one thing that matches servers/*, and we already established that that's a problem
01:27 SonOfLilit and we need to ignore servers/x/* without ignoring servers/x/
01:27 SonOfLilit hence servers/*/*
01:27 SonOfLilit one solution could maybe be ignore servers/* allow servers/*/ ignore servers/*/*
01:27 BinaryBench SonOfLilit: Hum, weird, I feel like there should be away to be like "ignore everything in this dir, but loop through it anyway to check if there are "!file" files"
01:27 SonOfLilit not sure if it would work or not, but I bet it would
01:27 Gsham joined #git
01:28 SonOfLilit lets see if my solution works
01:28 nonconvergent joined #git
01:29 SonOfLilit yup, this works: a/*,!a/*/,a/*/*,!a/*/x
01:30 SonOfLilit a/* !a/*/ means "ignore all the things in a/ except if they end with /, i.e. ignore all the things except the dirs"
01:30 BinaryBench SonOfLilit: Hum, I still find it weird that it works that way, but I understand
01:30 SonOfLilit it's just... all very literal
01:31 SonOfLilit git only walks dirs that are not ignored
01:31 SonOfLilit and a dir is simply a path that ends with /
01:31 SonOfLilit I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv. Anyone know how to debug it?
01:32 Noldorin joined #git
01:32 zacts joined #git
01:32 netj joined #git
01:32 BinaryBench SonOfLilit: Well, thank you for your help and insight, I'm going to grab a byte to eat ttyl
01:33 SonOfLilit :)
01:34 harish joined #git
01:34 HoierM joined #git
01:35 efco joined #git
01:36 d^sh joined #git
01:38 Peuc joined #git
01:40 BackEndCoder joined #git
01:40 michaelsharpe joined #git
01:40 kingpinlz_ joined #git
01:41 blackwind_123 joined #git
01:46 mizu_no_oto joined #git
01:46 Apteryx joined #git
01:46 dreiss joined #git
01:46 romerocesar joined #git
01:47 Vampire0 joined #git
01:48 e14 joined #git
01:52 rominronin joined #git
01:53 mingrammer joined #git
01:54 Goplat joined #git
01:56 romerocesar joined #git
01:56 kus joined #git
01:59 NeverDie joined #git
02:01 budre joined #git
02:04 fstd joined #git
02:06 madebymarkca joined #git
02:06 SonOfLilit I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv. Anyone know how to debug it?
02:09 rchavik joined #git
02:11 kaldoran joined #git
02:11 BackEndCoder joined #git
02:12 NeverDie joined #git
02:14 Geo joined #git
02:15 nilg joined #git
02:15 romerocesar joined #git
02:16 Geo Hi, I did a git pull of my develop branch, and everything was up to date. I then locally merged in some changes from a different branch with --no-ff, made a change, and commited the whole thing to develop. Now when I push to develop, it tells me my local branch is behing the remote. From that workflow I described, does anything seem obviously wrong that would have generated that error?
02:16 chachasmooth joined #git
02:18 themayor joined #git
02:21 rsukla joined #git
02:23 mizu_no_oto joined #git
02:25 Motoko joined #git
02:25 ojdo joined #git
02:25 ayjay_t joined #git
02:26 romerocesar joined #git
02:28 jon-mac joined #git
02:30 rgrinberg joined #git
02:32 dustinm joined #git
02:33 plouescat joined #git
02:33 tabakhase joined #git
02:36 romerocesar joined #git
02:36 kadoban Geo: Did someone else push something?
02:39 ayjay_t joined #git
02:40 jason237 joined #git
02:43 Ryanar joined #git
02:44 rgrinberg joined #git
02:46 romerocesar joined #git
02:47 mingrammer joined #git
02:49 Mobutils joined #git
02:53 raijin joined #git
02:53 AnarchyAo joined #git
02:53 Geo kadoban: nope
02:53 rominronin joined #git
02:54 Geo but, as I explore this further... why would I want to use --no-ff ? what are the implications of this?
02:54 mizu_no_oto joined #git
02:55 Geo in a simple system where only one user is making changes and merges are rare (for use in this explanatory case, at least) - would I ever want to force a no-ff rule for merges?
02:56 romerocesar joined #git
02:56 kadoban Geo: To insert explicitly insert a record of when a branch was merged, even if the branch being merged-into hasn't moved at all.
02:57 ShapeShifter499 joined #git
02:57 pR0Ps joined #git
02:57 kadoban It just creates a merge commit instead of, not, when there's a choice.
02:57 xlegoman joined #git
02:58 lagothrix joined #git
02:58 romerocesar joined #git
02:59 Geo that went right over my head :)
02:59 Geo I kinda get it, but mostly don't
03:00 thiago joined #git
03:02 aavrug joined #git
03:02 mehola joined #git
03:03 amdi_ joined #git
03:04 ayjay_t joined #git
03:04 hexagoxel joined #git
03:08 e14_ joined #git
03:15 Vortex34 joined #git
03:20 Mobutils joined #git
03:20 e14 joined #git
03:21 danielmahon joined #git
03:27 michaelsharpe joined #git
03:28 romerocesar joined #git
03:36 SonOfLilit joined #git
03:38 SonOfLilit I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv. Anyone know how to debug it?
03:39 romerocesar joined #git
03:39 sdothum joined #git
03:40 nonconvergent joined #git
03:41 NeverDie joined #git
03:41 mingrammer joined #git
03:42 sangy joined #git
03:44 NeverDie joined #git
03:47 BSab joined #git
03:49 romerocesar joined #git
03:49 robotroll joined #git
03:50 thiago SonOfLilit: what do you mean by "listens to the *.tsv"
03:50 thiago ?
03:51 Dazzyp joined #git
03:55 rominronin joined #git
03:55 mehola joined #git
03:56 nonconvergent joined #git
03:56 mehola joined #git
03:59 romerocesar joined #git
04:02 zivester joined #git
04:02 sanketdg joined #git
04:06 _W_ joined #git
04:07 chrisshattuck joined #git
04:09 romerocesar joined #git
04:10 linux_hot joined #git
04:12 dhollinger joined #git
04:15 romerocesar joined #git
04:16 phanimahesh joined #git
04:17 SonOfLilit thiago: two identical lines of configuration except for the file extension, git acts according to one (invoking an external differ or textconv) and completely ignores the other
04:17 vigilvindex joined #git
04:20 Andrew_K joined #git
04:24 narendraj9 joined #git
04:26 daey_ joined #git
04:26 BrianJ joined #git
04:27 CEnnis91 joined #git
04:28 cisco87 joined #git
04:30 a_thakur joined #git
04:33 oldrobot- joined #git
04:36 crayon joined #git
04:37 crayon joined #git
04:37 oldrobot joined #git
04:38 NeverDie_ joined #git
04:39 circ-user-E1qCe joined #git
04:40 michaelsharpe joined #git
04:41 dan2k3k4 joined #git
04:42 SonOfLilit I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv. Anyone know how to debug it?
04:46 bmahe joined #git
04:46 oldrobot joined #git
04:46 romerocesar joined #git
04:47 cagedwisdom joined #git
04:53 cisco87 joined #git
04:54 finalbeta1 joined #git
04:56 rominronin joined #git
04:56 lindenle joined #git
04:58 augg joined #git
04:58 Cabanossi joined #git
04:58 subhojit777 joined #git
04:59 nilg joined #git
05:01 dec0n joined #git
05:02 dsdeiz_ joined #git
05:03 red joined #git
05:03 bket joined #git
05:05 eivarv joined #git
05:06 kingarmadillo joined #git
05:08 ayjay_t joined #git
05:12 romerocesar_ joined #git
05:12 ShapeShifter499 joined #git
05:15 lindenle joined #git
05:15 budre joined #git
05:16 budre joined #git
05:16 |ifei5good joined #git
05:17 narkewoody joined #git
05:17 ayjay_t joined #git
05:21 SonOfLilit I have identical .gitattributes for *.csv and *.tsv, but git only listens to the *.tsv. Anyone know how to debug it?
05:21 oldrobot joined #git
05:23 shams joined #git
05:24 abhinav joined #git
05:26 narkewoody joined #git
05:29 Akimb joined #git
05:29 mingrammer joined #git
05:33 Rodya_ joined #git
05:34 efco joined #git
05:34 mangolisk joined #git
05:35 eivarv joined #git
05:37 afuentes joined #git
05:37 qt-x joined #git
05:37 dsdeiz joined #git
05:37 dsantiago joined #git
05:37 zeroed joined #git
05:37 zeroed joined #git
05:41 lindenle joined #git
05:41 prawnzy joined #git
05:41 BinaryBench joined #git
05:42 Vampire0_ joined #git
05:43 jalopez joined #git
05:43 inflames joined #git
05:45 ThomasLocke joined #git
05:46 jknetl joined #git
05:47 a_thakur_ joined #git
05:48 drizztbsd joined #git
05:48 stuh84 joined #git
05:53 Guest13669 left #git
05:53 Guest13669 joined #git
05:55 Guest13669 left #git
05:57 rominronin joined #git
05:58 bocaneri joined #git
06:00 xlegoman joined #git
06:04 nwkj86 joined #git
06:07 kingarmadillo joined #git
06:07 nonconvergent joined #git
06:08 romerocesar joined #git
06:09 monoprotic joined #git
06:09 mda1_ joined #git
06:12 vuoto joined #git
06:12 ayjay_t joined #git
06:12 lb1c joined #git
06:12 romerocesar_ joined #git
06:12 kyan joined #git
06:13 finalbeta joined #git
06:13 _ng joined #git
06:15 lindenle joined #git
06:15 chele joined #git
06:16 a_thakur joined #git
06:18 _ikke_ SonOfLilit: man git check-attr
06:18 gitinfo SonOfLilit: the git-check-attr manpage is available at http://jk.gs/git-check-attr.html
06:18 SonOfLilit thanks!
06:23 ap0c joined #git
06:23 alcohol joined #git
06:24 mingrammer joined #git
06:26 ap0c joined #git
06:26 abhinav anyone usintg emacs projectile?
06:26 abhinav i suspect that my .projectile conflicts with .gitignore.
06:28 MattMaker joined #git
06:29 michaelsharpe joined #git
06:29 BackEndCoder joined #git
06:31 JeroenT joined #git
06:35 romerocesar joined #git
06:36 djb-irc joined #git
06:37 MattMaker joined #git
06:37 _ikke_ SonOfLilit: Does it help
06:37 _ikke_ ?
06:38 SonOfLilit in the sense that my problem was magically solved
06:38 _ikke_ Oh, lovely
06:38 SonOfLilit I suspect that it has to do with the previous .gitattributes being in a subdir and the new one being in the root of the repo
06:38 Vaelatern joined #git
06:38 _ikke_ Ah, that does make a difference, yes
06:39 SonOfLilit not on purpose - I just read "anywhere .gitignore can be" and didn't question it
06:39 _ikke_ It only applies to the current dir and deeper
06:39 SonOfLilit and since .tsv *did* work...
06:39 peepsalot joined #git
06:41 Alienpruts joined #git
06:42 MattMake_ joined #git
06:45 elect joined #git
06:47 MattMaker joined #git
06:47 ShekharReddy joined #git
06:48 eduardas_m joined #git
06:51 Sceorem joined #git
06:52 MattMaker joined #git
06:57 MattMake_ joined #git
06:57 watabou joined #git
06:58 rominronin joined #git
07:00 overlord_tm joined #git
07:00 danslo joined #git
07:01 blahdodo joined #git
07:02 MattMaker joined #git
07:04 romerocesar joined #git
07:07 rominronin joined #git
07:11 arnlmth joined #git
07:11 MattMaker joined #git
07:11 watabou joined #git
07:13 blahdodo joined #git
07:14 arnlmth ➜  myrepository git:(dev) ✗ git pull
07:14 arnlmth error: Your local changes to the following files would be overwritten by merge:
07:14 arnlmth .meteor/versions
07:14 arnlmth Please, commit your changes or stash them before you can merge.
07:14 arnlmth Aborting
07:14 arnlmth .. I am getting this error. I wonder how could I override all my local files by remote files
07:15 preaction you can remove the changes you made with "git checkout"
07:16 MattMaker joined #git
07:16 arnlmth ➜  my repository git:(dev) ✗ git checkout dev
07:16 arnlmth M.meteor/versions
07:16 arnlmth Already on 'dev'
07:16 arnlmth Your branch and 'origin/dev' have diverged,
07:16 arnlmth and have 1 and 1 different commit each, respectively.
07:16 arnlmth (use "git pull" to merge the remote branch into yours)
07:16 arnlmth @preaction: that gives me the same error
07:17 _ikke_ arnlmth: He means checkout on the files that have changes
07:17 preaction that looks like a different error to me, and stop pasting into the channel
07:17 _ikke_ git checkout .meteor/versions
07:17 arnlmth @preaction. Ok, sorry for pasting.. First time here
07:18 preaction or even just "git checkout ." to restore everything
07:19 SteffanW joined #git
07:19 Oebele joined #git
07:21 MattMaker joined #git
07:21 fees joined #git
07:21 watabou joined #git
07:21 arnlmth looks like I still get the same error
07:21 a_thakur joined #git
07:22 GavinMagnus joined #git
07:22 prawnzy joined #git
07:22 encod3 joined #git
07:22 satya4ever joined #git
07:22 arnlmth cause locally I have an indentical file with the remote, but still it's bouncing an error that my file is modified
07:23 ReenignE joined #git
07:23 ReenignE joined #git
07:23 SteffanW joined #git
07:24 arnlmth When I go to git repository in the browser .meteor/versions file is identical with my local .meteor/versions
07:24 danslo arnlmth: what's the output of 'git status'?
07:24 arnlmth and I am looking at the same branch in the both
07:25 SteffanW joined #git
07:25 arnlmth Your branch and 'origin/dev' have diverged, and have 1 and 1 different commit each, respectively. modified:   .meteor/versions
07:25 mekeor joined #git
07:26 antispaceman joined #git
07:26 danslo anyway, 'git reset --hard origin/dev' should get you back on track
07:26 MattMake_ joined #git
07:26 tristanp joined #git
07:26 heftig safe to do if "git diff origin/dev" is empty
07:26 circ-user-E1qCe joined #git
07:27 jceb joined #git
07:27 Balliad joined #git
07:27 arnlmth @danslo: I reckon that command, but what does it do? Is it indentical to removing my local folder and cloning again and checking out the branch
07:28 mariuscc joined #git
07:28 danslo "Resets the index and working tree. Any changes to tracked files in the working tree since <commit> are discarded."
07:28 arnlmth so it does affect just locally and it will not affect to remote
07:29 danslo indeed
07:29 arnlmth ^^
07:31 MattMaker joined #git
07:32 ahmedelgabri joined #git
07:32 neanderslob joined #git
07:32 Fuzzmz in git most things don't affect remotes unless you push
07:33 arnlmth I will sound then... The actual case is that when there are several developer working for the branch, but in different files, I always face this problem when they commit changes on some files that have not edited..  I would of imagined that if they are committing changes on files that I have not edited, I just have to use git pull origin
07:33 mekeor i just created a git-repository on github. now i realized i'd like to change the local user.name configuration-variable of this repository -- not only for future commits but also for the single and last one. what i did so far is "git config --replace-all --local user.name foobar". how do i push the changed config now (to take effect for the single past commit)?
07:34 osse mekeor: git config cannot change existing commits.
07:34 osse and git config simply edits a couple of files on your system. it's not something you push
07:34 Fuzzmz @mekeor, see https://help.github.com/articles/changing-author-info/
07:35 danslo arnlmth: well for starters, I recommend you use feature branches and only merge to some common (dev/master) branch when you are done with that bug/feature/whatever. as for what you describe, you can regularly do a 'git pull --rebase <common branch>' to keep your own changes cleanly on top of a recent version of that common branch
07:36 kanyeezy joined #git
07:36 mekeor thank you, osse & Fuzzmz (:
07:36 MattMake_ joined #git
07:36 vlt Hello. After years of using git as a single developer I now have a coworker and need some help getting started. I created a bare repo we both can reach, pushed my main repo there, he cloned it, made commits and pushed them. Now a fetched to analyze what he’s done. Too many changes per commit and some to the wrong branches. How to continue now?
07:36 vlt What can he change in his repo without messing up mine?
07:36 vlt Or how can I make corrections so that he also gets them?
07:37 a_thakur joined #git
07:37 osse if you want to make corrections you need push --force
07:37 osse not sure that's a good idea with a new developer
07:38 Fuzzmz vlt, how I'd handle this is have him make a branch on which he'll cherry pick his changes to have them backed up
07:38 Fuzzmz then force push your repo to overwrite his changes
07:38 Fuzzmz then work with him on his feature branch to have the commits like you want them
07:38 vlt Fuzzmz: Sounds good! Thank you.
07:38 Fuzzmz even without some sort of repository management tool (gitlab, github, whatever) I'd still say that feature branches are a good idea
07:39 vlt What happens to the stuff my repo already knows (because I fetched, but not merged)?
07:39 Fuzzmz just tell him every time he wants to make a change to create a new branch and work on that, and that's what you'll review
07:39 arnlmth @danslo: So  'git pull --rebase <common branch>' will do the trick that I will pull the new version for the files that I have not changed?
07:40 Fuzzmz @vlt, as long as you don't pull everything should be ok
07:40 danslo arnlmth: git pull --rebase if basically a 'git fetch' followed by 'git rebase FETCH_HEAD'. it will temporarily put away your local commits, grab the newest stuff from your remote, then try to apply your local commits on top of that
07:40 Fuzzmz the fetch just tells your local repo that 'hey, something changed on the remote'
07:40 vlt Fuzzmz: Ok, thanks. I’ll try that.
07:41 Fuzzmz @vlt, also make a copy of your repository as it is now just to have a backup
07:41 danslo arnlmth: if you do that once in a while, you'll greatly reduce the size of merge conflicts (if there are any at all)
07:42 Fuzzmz @arnlmth, you can also set git config --global pull.rebase true
07:42 Fuzzmz which always rebases on a pull
07:42 arnlmth @danslo: error: Cannot pull with rebase: You have unstaged changes.
07:42 arnlmth @danslo: That is what I got out of the rebase
07:42 Fuzzmz it means you have modified files which are not yet added/committed
07:43 Fuzzmz either stash the changes, or add and commit the modified files, then the pull with rebase should work
07:43 vlt Fuzzmz: Backup, sure :-)   Is it correct that the "fetch" does not only _tell_ my local repo that something has changed but also, well, fetches all the changes so that it is then able to merge without a new remote connection, right?
07:43 asti joined #git
07:43 Fuzzmz vlt, yep; pull is basically a fetch + merge
07:45 MattMaker joined #git
07:46 baxx joined #git
07:47 stasmakarov joined #git
07:47 Raging_Hog joined #git
07:48 arnlmth @Fuzzmz: I cannot commit. When I do git add <files> and git commit -m "changes" it bounces back the same error
07:49 arnlmth @Fuzzmz: Changes not staged for commit: modified: .meteor/versions
07:49 prillian5 joined #git
07:49 arnlmth @Fuzzmz: And this is not the file I have not edited and it should be up to date
07:50 Fuzzmz arnlmth, could you try a git rebase --abort
07:50 MattMake_ joined #git
07:50 Fuzzmz then do a git checkout -- .meteor/versions
07:50 Fuzzmz so that you have the versions file that is in git
07:50 Fuzzmz the add + commit your changes
07:50 Fuzzmz and then try to do a git pull --rebase <common branch>
07:51 romerocesar joined #git
07:52 arnlmth @Fuzzmz: No rebase in progress?
07:52 arnlmth git checkout -- .meteor/versions ==> gives the same error as previously
07:52 Fuzzmz hmm, ok, I'm a bit out of ideas
07:52 arnlmth Me too
07:53 laumann joined #git
07:53 arnlmth I weird, because it looks that I have the same version of this "versions"-file with remote
07:53 arnlmth I wonder that can I force the pull this file from the remote
07:55 MattMaker joined #git
07:55 Fuzzmz @arnlmth, what does git status say?
07:56 submitnine joined #git
07:57 arnlmth ➜  myResository git:(dev) git status  On branch dev Your branch and 'origin/dev' have diverged, and have 1 and 1 different commit each, respectively. (use "git pull" to merge the remote branch into yours) Changes not staged for commit: (use "git add <file>..." to update what will be committed)  (use "git checkout -- <file>..." to discard changes in working directory)modified:   .meteor/versions no changes added to commit (use "git add" and/or "g
07:57 arnlmth it commit -a")
07:58 JeroenT joined #git
07:58 Fuzzmz ok, that's weird as hell, I don't see why it wouldn't let you git checkout -- .meteor/versions :/
07:59 arnlmth I wonder that why this is not showing the files that I have actually change
07:59 arnlmth d
08:00 MattMaker joined #git
08:01 arnlmth This is damn frustrating as looks like I am forced to remove the whole local folder and then clone again
08:02 Fuzzmz hmm
08:02 pikajude joined #git
08:02 Fuzzmz you can try a git reset --hard HEAD
08:03 arnlmth so this safe command? :)
08:03 arnlmth so doing the pull of origin dev
08:04 cdown joined #git
08:04 Fuzzmz it should discard all changes and bring your repo to be identical to HEAD
08:04 Fuzzmz and it also removes all modified files from your directory
08:04 Fuzzmz it's more or less a nuke
08:05 Fuzzmz instead of deleting the folder and cloning again
08:05 MattMake_ joined #git
08:07 arnlmth still same error
08:07 kurkale6ka joined #git
08:07 PCatinean joined #git
08:07 Fuzzmz ok, now I'm really out of ideas
08:07 GodEater` left #git
08:08 danslo arnlmth: can you post the full output of 'git status' to pastebin? just copy paste it entirely
08:08 kingarmadillo joined #git
08:09 mikecmpbll joined #git
08:09 ProbabilityMoon joined #git
08:09 finalbeta joined #git
08:09 tlaxkit joined #git
08:10 MattMaker joined #git
08:10 arnlmth which Syntax Highlighting?
08:11 arnlmth http://pastebin.com/2iv60eAQ
08:12 mingrammer joined #git
08:12 danslo 'git checkout .meteor/versions', not 'git checkout dev'
08:13 romerocesar_ joined #git
08:14 a_thakur joined #git
08:14 arnlmth http://pastebin.com/e6HwQvCj
08:15 dangerousdave joined #git
08:15 Fuzzmz hmm, this might be a long shot, but @arnlmth can you close your terminal and after that open a new one
08:15 danslo arnlmth: ah I think I've seen this before on a coworker's machine
08:15 Fuzzmz and then try the git checkout -- .meteor/versions
08:16 Fuzzmz and check if that still gives you that error
08:16 danslo what does 'git diff' say?
08:16 danslo im suspecting autocrlf issue
08:16 arnlmth git diff says a lot more
08:16 Fuzzmz @arnlmth, try just deleting the file :D
08:16 arnlmth git diff tracker the changes I have made
08:16 Fuzzmz like regular delete
08:16 arnlmth if I delete the file M => D
08:16 Fuzzmz ok
08:17 arnlmth and same issue
08:17 michaelsharpe joined #git
08:17 Fuzzmz with the file deleted try git checkout -- .meteor/versions
08:17 arnlmth what does -- means in this?
08:18 Mobutils_ joined #git
08:18 ferr joined #git
08:18 Fuzzmz it discards changes in the working directory for that file
08:19 Fuzzmz git status even says to use that
08:19 MattMaker joined #git
08:19 Fuzzmz from your http://pastebin.com/e6HwQvCj : (use "git checkout -- <file>..." to discard changes in working directory)
08:21 cbreak arnlmth: -- is a separator
08:21 cbreak it separates ref parameters from path parameters
08:21 cbreak arnlmth: git checkout foo -- bar would checkout the file bar from the branch foo
08:24 MattMake_ joined #git
08:26 jceb joined #git
08:29 MattMaker joined #git
08:29 mischat joined #git
08:29 Mobutils joined #git
08:31 arnlmth deleting the version file + checking it out again remove the conflict, but still cannot do anything
08:34 MattMake_ joined #git
08:35 nonconvergent joined #git
08:35 Mobutils joined #git
08:35 romerocesar joined #git
08:36 Phlogistique joined #git
08:37 Andrew_K joined #git
08:38 BadSnappa joined #git
08:39 MattMaker joined #git
08:44 MattMaker joined #git
08:44 choki joined #git
08:48 SimonNa joined #git
08:49 Anticom joined #git
08:51 dangerousdave joined #git
08:52 Anticom joined #git
08:53 MattMaker joined #git
08:54 ayjay_t joined #git
08:56 Orion3k joined #git
08:56 pikpok_ joined #git
08:56 wget joined #git
08:56 SteffanW joined #git
08:58 MattMaker joined #git
08:58 argol joined #git
09:00 argol join #git
09:00 _ikke_ already there :P
09:00 argol ok :)
09:01 Mobutils joined #git
09:01 JeroenT_ joined #git
09:02 argol hi how to force git hosted on apache 2.2 to work in smart mode. when i curl http://my.repo/info/ref?service=git-receiver-pack i get in conten-type header text/plain ?
09:03 MattMaker joined #git
09:05 irqq joined #git
09:05 canton7 argol, and man git-http-backend
09:05 gitinfo argol: the git-http-backend manpage is available at http://jk.gs/git-http-backend.html
09:05 canton7 https://git-scm.com/book/en/v2/Git-on-the-Server-Smart-HTTP
09:06 mingrammer joined #git
09:07 ShekharReddy joined #git
09:08 MattMaker joined #git
09:09 argol I have all set. git-http-backed, GIT_HTTP_EXPORT_ALL, and root of git repos. Can i check if that request gone through git-http-backed cgi ?
09:09 kingarmadillo joined #git
09:10 mariuscc joined #git
09:10 usm4n joined #git
09:11 magicturtle joined #git
09:11 achadwick joined #git
09:13 MattMaker joined #git
09:14 pijiu joined #git
09:14 ntonjeta joined #git
09:14 argol PROPFIND http verb is used only in dump mode ?
09:14 ntonjeta left #git
09:17 wrouesnel1 joined #git
09:18 jqypau joined #git
09:18 MattMake_ joined #git
09:18 jast argol: if your request to .../info/refs/service?git-receive-pack (your example had a few typos, check that) doesn't return a first line that goes like "001f# service=git-receive-pack", you're not using smart HTTP. and yes, smart HTTP doesn't use PROPFIND.
09:18 mlittlej joined #git
09:18 jast bah, now I typoed the URL, too. the right one is .../info/refs?service=git-receive-pack
09:18 jast sorry :)
09:20 rafalcpp joined #git
09:20 aidalgol joined #git
09:20 Emperor_Earth joined #git
09:21 aavrug joined #git
09:22 finalbeta joined #git
09:22 usm4n_ joined #git
09:22 aavrug joined #git
09:22 kingarmadillo joined #git
09:24 mlittlej joined #git
09:24 argol I have set logging SET GIT_CURL_VERBOSE=1 on git client. I don't see  "001f# service=git-receive-pack". I have empty repository and run git update-server-info. File my.repo/info/refs is empty it is ok?
09:24 Dazzyp joined #git
09:26 jast update-server-info is not needed with smart HTTP
09:27 MattMaker joined #git
09:27 jast have you added git-http-backend to your apache config? can you paste your apache config for the repo virtualhost? (!paste)
09:27 gitinfo Please use a pastebin for snippets longer than one line. https://gist.github.com and https://vomitb.in both work well.
09:27 argol yes i know by hook/post-update was not working
09:27 argol have 404 in apache access log
09:29 tristanp joined #git
09:29 ahmedelgabri joined #git
09:30 argol !past SetEnv GIT_PROJECT_ROOT "c:\var\git" SetEnv GIT_HTTP_EXPORT_ALL SetEnv REMOTE_USER=$REDIRECT_REMOTE_USER ScriptAlias /git/ "C:\Program Files\Git\mingw64\libexec\git-core\git-http-backend.exe" ScriptLog "C:\Apache22\logs\cgi_errors_logs.txt"  DocumentRoot "c:\var\git"  <Directory "c:\var\git">    Options ExecCGI Indexes    Order allow,deny    Allow from all </Directory>   <Directory "C:\Program Files\Git\mingw64\libexec\git-core*">
09:30 gitinfo [!paste] Please use a pastebin for snippets longer than one line. https://gist.github.com and https://vomitb.in both work well.
09:32 MattMake_ joined #git
09:32 sonikspin joined #git
09:33 argol !past https://gist.github.com/rscipien/73be7a441459db6431df35526a2c7e83
09:33 gitinfo [!paste] Please use a pastebin for snippets longer than one line. https://gist.github.com and https://vomitb.in both work well.
09:33 jast okay, according to that, your repo URL should be: http://yourhostname/git/reponame.git (assuming there's a reponame.git in c:\var\git). is that what you've been trying?
09:34 ShapeShifter499 joined #git
09:34 jast by the way, the "! paste" thing was just to get the bot to tell you which way of pasting we prefer, you didn't have to type it yourself :)
09:36 argol Yes i try that way but it turned out to by http://host/repo.git :)
09:36 Borkr joined #git
09:36 jast that doesn't fit your ScriptAlias setting, though
09:37 MattMaker joined #git
09:38 jast if you want to have your repo URLs at / instead of /git/, you have to change the apache config so it says '/' instead of '/git/'
09:39 jast that affects your first LocationMatch block, too
09:39 jast you have a stray '*' in your <Directory ...>
09:39 a_thakur joined #git
09:40 chll_ joined #git
09:42 MattMaker joined #git
09:42 mkopriva joined #git
09:43 roelmonnens joined #git
09:44 argol how to change it to /git. Change only DocumentRoot to c:\var ?
09:44 jast with your current configuration, what do you get in your access/error logfiles if you try *with* /git/?
09:46 argol "GET /git/ HTTP/1.1" 500
09:47 argol error  fatal: GIT_PROJECT_ROOT is set but PATH_INFO is not
09:47 MattMaker joined #git
09:47 jast okay, so it's already set up to let you use /git/ but it's not working correctly yet
09:48 jast you do have to include the repo subdir in the URL, though
09:48 argol when i go to browser and type http:\\myhost i get list of repos in c:\var\git
09:48 jast unless you have a single git repo right inside c:\var\git...?
09:49 jast yeah, that's because of DocumentRoot. smart HTTP lives in /git/ only, the way the config is set up now. any other requests goes through normal delivery of files by apache
09:49 jast for your current smart HTTP setup it doesn't matter what happens outside /git/
09:50 elsevero joined #git
09:51 argol ok i get it. So first param of ScriptAlias /git/ must match part of url
09:52 MattMaker joined #git
09:52 jast exactly, and the first part of the LocationMatch blocks (^/git/.*/...), too
09:55 leeN joined #git
10:00 King_Hual joined #git
10:01 mingrammer joined #git
10:01 MattMaker joined #git
10:01 argol jast: thanks for help i will try to figure it out
10:05 michaelsharpe joined #git
10:06 Raansu joined #git
10:06 MattMaker joined #git
10:07 Raansu joined #git
10:11 overlord_tm joined #git
10:11 MattMaker joined #git
10:13 dan2k3k4 joined #git
10:14 Mobutils joined #git
10:15 JeroenT joined #git
10:15 stamina joined #git
10:16 MattMaker joined #git
10:16 JeroenT joined #git
10:18 kanyeezy joined #git
10:18 kanyeezy joined #git
10:20 Hounddog joined #git
10:21 MattMake_ joined #git
10:22 cdown joined #git
10:24 Akimb joined #git
10:26 MattMaker joined #git
10:30 liuzhen joined #git
10:31 rnsanchez joined #git
10:31 prawnzy joined #git
10:33 Hounddog_ joined #git
10:34 a_thakur joined #git
10:35 MattMaker joined #git
10:40 MattMake_ joined #git
10:40 kejoma joined #git
10:45 Silenced_v2 joined #git
10:45 MattMaker joined #git
10:50 MattMake_ joined #git
10:52 choki joined #git
10:52 NeverDie joined #git
10:55 MattMaker joined #git
10:55 mingrammer joined #git
10:57 Sadeukko joined #git
10:58 Darren__ joined #git
10:59 rmrrn joined #git
10:59 rmrrn joined #git
11:00 MattMaker joined #git
11:01 JeroenT joined #git
11:03 JeroenT_ joined #git
11:04 JeroenT__ joined #git
11:05 bernardio joined #git
11:06 berndl joined #git
11:06 DolphinDreamm joined #git
11:09 MattMaker joined #git
11:11 notebox joined #git
11:12 kingarmadillo joined #git
11:14 Hirsemeier joined #git
11:14 rossome joined #git
11:14 MattMake_ joined #git
11:14 ams__ joined #git
11:17 CussBot joined #git
11:19 tmg joined #git
11:19 tmg is there any way to cache the log --graph output?
11:19 MattMaker joined #git
11:19 johnmilton joined #git
11:19 Lomex joined #git
11:20 dvl joined #git
11:20 mehola joined #git
11:20 rastos2 joined #git
11:21 argol I have error in apache log. [fatal: GIT_PROJECT_ROOT is set but PATH_INFO is not]. In documentation https://jk.gs/git-http-backend.html  I see that PATH_INFO should be set by server
11:22 argol GIT_PROJECT_ROOT may be different then DocumentRoot ?
11:24 MattMake_ joined #git
11:25 dvl left #git
11:25 Hounddog joined #git
11:25 nilg joined #git
11:27 fahadash joined #git
11:29 MattMaker joined #git
11:30 a_thakur joined #git
11:32 tristanp joined #git
11:34 MattMaker joined #git
11:34 nilg` joined #git
11:35 jpic joined #git
11:35 mohabaks joined #git
11:39 a_thakur joined #git
11:41 __main__ joined #git
11:41 Repox joined #git
11:43 MattMaker joined #git
11:44 sdothum joined #git
11:45 Gsham joined #git
11:47 Lomex joined #git
11:47 Borkr joined #git
11:47 Hirsemeier left #git
11:48 MattMaker joined #git
11:48 Hounddog_ joined #git
11:49 sanketdg joined #git
11:49 mingrammer joined #git
11:49 ahr3n joined #git
11:51 Hounddog__ joined #git
11:52 Sceorem joined #git
11:53 MattMaker joined #git
11:53 michaelsharpe joined #git
11:53 Hounddog joined #git
11:54 Viddax joined #git
11:55 Viddax I want to give developer access to a repo but I don't want developer to have acess all the files - only files they need. Is that possible?
11:56 Fuzzmz any simple way to list all tags on a branch?
11:56 daey Viddax: maybe with submodules?
11:56 daey just something that crossed my mind, im a git newb*
11:57 Viddax daey Is submodules base on folders or filess?
11:57 harish joined #git
11:57 daey Viddax: afaik you make simple projects. and include the git projects into your main git project
11:57 newbie1 joined #git
11:57 Fuzzmz Viddax, a submodule is practically a separate git repo
11:58 daey Viddax: you end up with multiple .git directories in your overall project
11:58 dminuoso joined #git
11:58 Fuzzmz so you'd split the files you want the other dev to have access to, commit them to a separate repo, delete them from your main repo and then use a submodule to reference the new repo
11:58 MattMaker joined #git
11:58 dminuoso Hi. I have a modified file in my working directory and want to undo a chunk of it. Is there an immediate way to do this?
11:58 Fuzzmz the thing is that as far as I know, you can't have the files in the submodule at the same level with the others (so they'll be under a folder)
11:58 dminuoso Or do I have to stage the chunk and then use git checkout?
11:58 daey yea i would assume that as well
11:59 Gsham joined #git
11:59 Seveas dminuoso: chechout -p
11:59 daey besides if your dev access level changes its gonna be a nightmare
11:59 Hounddog_ joined #git
11:59 dminuoso Seveas: Ah neat.
12:02 LBV_User joined #git
12:02 daey Viddax: out of curiousity. why do you want to hide code?
12:02 cbreak Fuzzmz: tags are not on a branch
12:02 JeroenT joined #git
12:02 Fuzzmz cbreak, well, yeah, if you want to get technical, all tags on commits which share a common parent
12:02 Fuzzmz :P
12:03 Fuzzmz more or less
12:03 synthroid joined #git
12:03 cbreak Fuzzmz: git tag --contains XXX
12:03 cbreak that'll list all tags that can be used to reach that commit
12:03 MattMake_ joined #git
12:03 cbreak (it does NOT list all tags that can be reached from that commit)
12:03 enleeten joined #git
12:04 Fuzzmz thanks, will give it a try
12:08 MattMaker joined #git
12:08 stamina joined #git
12:09 Viddax daey Not sure i fully understand how it work. Basically I have  main project repo. I would create a folder in a main project with git repo?
12:09 JeroenT joined #git
12:09 Viddax in folder, I can duplicate from main project and remove files/folders they don't need.
12:10 JeroenT joined #git
12:10 cbreak Viddax: no
12:11 cbreak Viddax: in git, everyone that can clone has the full history of all files
12:11 JeroenT joined #git
12:11 cbreak if you want to hide things, put them into separate repositories
12:11 Viddax Ah seperate repo
12:11 cbreak you can glue separate repositories together with submodules
12:11 cbreak but they are still mostly separate
12:11 Viddax main-dev-repo and mini-dev-repo
12:12 Viddax I give developer a mini-dev-repo
12:13 Viddax cbreak Would I need to setup submodules in both repo main-dev-repo and mini-dev-repo?
12:13 Fuzzmz no, just in main-dev-repo
12:13 Viddax ok
12:13 cbreak Viddax: what ever
12:13 Fuzzmz check out https://github.com/blog/2104-working-with-submodules
12:13 cbreak Viddax: it depends on which repo you want to glue into the other
12:15 Viddax cbreak Let say Developer-A made some changes on file.php on main-dev-repo. Can Developer-B from mini-dev-repo  pull latest code? new changes of file.php
12:15 Fuzzmz no
12:15 cbreak Viddax: they are separate repositories
12:15 Fuzzmz the repositories are separate
12:15 cbreak Viddax: they have unrelated files
12:16 cbreak merging them makes zero sense
12:16 choki_ joined #git
12:16 Viddax Hmm that is a problem.
12:16 cbreak git obviously allows merging them
12:16 cbreak but it won't make much sense
12:17 Viddax That is a shame git does not support it.
12:17 Fuzzmz Viddax, basically what you want is for a dev to see all files, but only be able to modify some of them, correct?
12:17 MattMaker joined #git
12:17 Viddax No
12:18 kusfedora joined #git
12:18 Fuzzmz from what you said above, it means that dev-b needs file.php, so that one should live in mini-dev-repo, not in main-dev-repo
12:19 Fuzzmz it's all about how flexible/modular your project is
12:19 Viddax Fuzzmz Well, there is two options. Option-1:  they can see all files but they can only read/edit some files they have access to (granted). or Option 2: Only show some files and have access to read/edit.
12:19 GodGinrai joined #git
12:20 Fuzzmz yeah, you only get option 2 in git via submodules or subtrees, and that is only if your project can be split in multiple repos
12:20 Fuzzmz if you have a monolithic website I imagine that'd be hard to do
12:20 Viddax Laravel Project
12:21 Fuzzmz sorry, no clue about php
12:21 Viddax Fuzzmz In regarding option 2. Can developers from both repo pull latest code even its seperate repos?
12:22 Fuzzmz so, how submodules work is like this:
12:22 MattMake_ joined #git
12:22 Fuzzmz main-dev-repo is the main repository
12:22 Fuzzmz mini-dev-repo is another repository (self-sustaining)
12:22 Fuzzmz you then link mini-dev-repo to main-dev-repo as a submodule
12:23 Fuzzmz this means that people with access to main-dev-repo also see stuff coming from mini-dev-repo
12:23 Viddax Oh cool
12:23 Fuzzmz basically mini-dev-repo is a shortcut inside main-dev-repo
12:23 Fuzzmz the flow to update things is generally like this:
12:24 Viddax What about if a developer update client.php in main-dev-repo and mini-dev-repo have also have client.php file.
12:24 Silenced joined #git
12:24 fguillot joined #git
12:24 Fuzzmz it shouldn't have it
12:24 ahmedelgabri joined #git
12:24 Fuzzmz it basically can't, as submodules live in subfolders inside the main repo
12:24 Fuzzmz you shouldn't have the same file in both repos
12:25 johnmilton joined #git
12:25 shams joined #git
12:25 osse !subproject
12:25 gitinfo [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.
12:26 quackgyver joined #git
12:26 jpic left #git
12:26 is_null joined #git
12:26 JCaesar joined #git
12:26 kingarmadillo joined #git
12:26 is_null hi all, i don't understand why git submodule update --depth 1 fails to find a submodule commit ( log http://dpaste.com/06YVQ9H ) which is definitely available upstream ( https://github.com/ansible/ansible-modules-core/commit/48d932643b5ad2c66261e8a11cbf402d53f8d83b )
12:27 Viddax Fuzzmz I see, not perfect though :(
12:27 MattMaker joined #git
12:27 naw joined #git
12:27 Viddax Fuzzmz There are chance we all developers working on many same files.
12:28 Fuzzmz question is, as daey said, why do you want to hide code?
12:28 Viddax Boss said.
12:28 Viddax :/
12:28 Fuzzmz let me guess, boss isn't a developer, right? :)
12:28 Viddax Correct
12:29 Geo left #git
12:29 Fuzzmz to be honest, I don't know of any source code management system that lets you do what you want
12:29 Viddax because we have private repo on github and we like to give some codes to a developer (remote work) and he don't want deveoloper to have access all the codes.
12:29 Viddax Fuzzmz no version control does that?
12:29 Fuzzmz not that I know of
12:30 Fuzzmz not without being able to split your projects into self-contained parts
12:30 johnmilton joined #git
12:30 Seveas Viddax: how often are you going to ask the same question?
12:30 Seveas It wasn't possible the last time you asked, it isn't suddenly magically possible now.
12:31 Fuzzmz what I'd do in your case is set up a separate repository with the file the dev needs to have access to + the minimal sanitized framework needed for him to do his job
12:31 Fuzzmz and then just have him send over patches via email
12:31 Seveas Fuzzmz: he knows this, has asked it a few times already this week.
12:31 Fuzzmz oh, ok
12:32 MattMaker joined #git
12:35 LBV_User joined #git
12:36 ahmedelgabri joined #git
12:37 MattMaker joined #git
12:38 liquidjamm joined #git
12:39 pasadinhas joined #git
12:39 rastos2 joined #git
12:40 tubaman joined #git
12:40 dangerousdave joined #git
12:41 griffindy joined #git
12:42 trebuh joined #git
12:42 MattMaker joined #git
12:42 trebuh Hello, is rebasing a branch a commonly accepted good practice or should I do differently?
12:43 dendazen joined #git
12:43 rethus joined #git
12:44 mingrammer joined #git
12:44 mooru joined #git
12:45 anonymuse joined #git
12:45 tabrez joined #git
12:47 Seveas trebuh: depends on the situation
12:48 Seveas if it's a branch you collaborate on with others, frequent rebasing will be annoying. If it's just your dev things, rebasing is just fine.
12:50 trebuh Seveas: I have a master in state 1. I create a branch (state 1B) Commits are push onto master (state 2) If I want to incorporate those changes from master into my branch (and have it in state 2B) what would the best practice be knowing that no common files were modified between master and branch?
12:51 dminuoso trebuh: Why do you care whether common files were modified? If they are, you get merge conflicts that you can simply dael with.
12:51 MattMaker joined #git
12:53 actinide joined #git
12:53 mizu_no_oto joined #git
12:56 afuentes joined #git
12:56 MattMaker joined #git
12:57 armenzg joined #git
12:58 marc_v92 joined #git
12:59 rewt trebuh, git is commit-based, not file-based like some other versioning systems, so for branching/merging it doesn't really matter what files are touched
12:59 jaafar joined #git
12:59 endiruna joined #git
13:01 rewt if you haven't pushed your 1B commits yet, rebasing should be ok; if you've pushed them, then others may have fetched them so they may have issues so merging may be a better strategy in that case
13:01 MattMake_ joined #git
13:02 eizua joined #git
13:02 tcsc joined #git
13:02 jeffreylevesque joined #git
13:03 endiruna joined #git
13:04 fmcgeough joined #git
13:05 notebox joined #git
13:06 cdg joined #git
13:06 mooru joined #git
13:06 MattMaker joined #git
13:07 SteffanW joined #git
13:08 ExoUNX joined #git
13:11 MattMaker joined #git
13:11 d0nn1e joined #git
13:12 ItSANgo_ joined #git
13:12 ItSANgo joined #git
13:14 endiruna joined #git
13:14 zivester joined #git
13:16 MattMake_ joined #git
13:16 rgrinberg joined #git
13:19 KaibutsuX_ joined #git
13:19 KaibutsuX_ can anyone tell me if there is a way while rebasing -i to just trash a commit?
13:21 trebuh Thanks for the answers
13:21 avar joined #git
13:21 trebuh left #git
13:21 mehola joined #git
13:22 andlabs joined #git
13:22 bremner KaibutsuX_: just delete it from the todo list
13:23 lindenle joined #git
13:24 avar joined #git
13:25 cbreak KaibutsuX_: during a rebase: git rebase --skip
13:25 MattMaker joined #git
13:26 rgrinberg joined #git
13:26 endiruna joined #git
13:27 dimi1947 joined #git
13:27 prawnzy joined #git
13:29 ejb joined #git
13:30 le_melomane joined #git
13:30 MattMake_ joined #git
13:33 mizu_no_oto joined #git
13:33 xcesariox joined #git
13:33 pasadinhas joined #git
13:34 CEnnis91 joined #git
13:34 itosh joined #git
13:34 m0viefreak joined #git
13:35 tristanp joined #git
13:35 MattMaker joined #git
13:38 Snopandan joined #git
13:38 ahmedelgabri joined #git
13:38 mingrammer joined #git
13:38 Snopandan joined #git
13:40 LukeLR joined #git
13:40 MattMaker joined #git
13:41 al-damiri joined #git
13:41 MacroMan joined #git
13:41 rnsanchez_ joined #git
13:41 michaelsharpe joined #git
13:42 MacroMan in .gitignore, how do I ignore everything except the .gitignore file itself? !.gitignore doesn't seem to work
13:44 lss8 joined #git
13:44 MacroMan This is the entirety of my .gitignore file: https://p.ngx.cc/3b74aa5fbb5785de
13:44 nonconvergent joined #git
13:45 MattMaker joined #git
13:45 finalbeta1 joined #git
13:47 MacroMan Oh ok. Just put the * first and the !.gitignore after
13:48 barteks2x joined #git
13:48 cbreak MacroMan: indeed.
13:48 cbreak MacroMan: or just ignore everything and force-add .gitignore
13:49 AaronMT joined #git
13:50 MattMake_ joined #git
13:50 NeverDie joined #git
13:50 Snopandan joined #git
13:50 dsdeiz joined #git
13:50 dsdeiz joined #git
13:51 nonconve1gent joined #git
13:51 ejb joined #git
13:53 everamzah joined #git
13:56 tristanp joined #git
13:57 crose joined #git
13:58 ash_workz joined #git
13:59 raijin joined #git
13:59 MattMaker joined #git
14:01 eivarv joined #git
14:01 dan2k3k4 joined #git
14:02 niels__ joined #git
14:04 MattMake_ joined #git
14:06 gyre007 joined #git
14:06 nidr0x joined #git
14:06 bibiki joined #git
14:07 umrwpr joined #git
14:07 gyre007 Hi there. I have a n00b question about git. I have rebased my branch against master and resolved some merge conflicts . When I push now it looks like my push adds like 40 commits
14:08 gyre007 Which is kinda nuts to me - but that could be because of my misunderstanding?
14:08 gyre007 Basically, it looks like the rebase added all commits since my branch was created
14:09 MattMaker joined #git
14:10 bibiki I set git config --global core.compression 0
14:10 bibiki yet, git push reports: "Delta compression... compressing objects: 100%"
14:10 bibiki gyre007, you likely committed before rebased, no?
14:10 gyre007 bibiki: yeah, is there a way I can fix this?
14:10 cbreak gyre007: look at !lol
14:10 gitinfo gyre007: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
14:11 cbreak and pastebin it
14:11 bibiki not sure that you can undo commits, unless you reinitialize the repo... but that'd generate more hassles
14:12 cbreak bibiki: you can undo commits
14:12 marc_v92 joined #git
14:13 bibiki thanks cbreak, I aint known that
14:13 eivarv joined #git
14:13 cbreak git reset --hard HEAD~ nukes away the most recent commit. git rebase -i is more surgical
14:14 cbreak note that both are history rewriting
14:14 bibiki cbeak, do you also happen to know why is git compressing though I set config --global compression to 0
14:14 MattMake_ joined #git
14:14 bibiki or at least, it is reporting "compressing 100%"
14:16 notebox joined #git
14:16 cbreak it's probably compressing with a compression algorithm for no compression
14:17 LukeLR joined #git
14:17 CheckDavid joined #git
14:17 rgrinberg joined #git
14:18 sangy joined #git
14:19 e14_ joined #git
14:19 MattMaker joined #git
14:19 mikecmpbll joined #git
14:20 wget Hi guys. Is there any built-in solution in git allowing me to remove local branches that have been removed from the tracked remote?
14:20 wget I have come along with the following command, but I think this is weird, there is no built-in solution AFAIK:
14:20 wget cleanup = "!git branch --merged | grep -v '^* master$' | grep -v '^  master$' | xargs git branch -d"
14:21 le_melomane joined #git
14:21 cbreak wget: git remote prune should do that
14:22 mischat joined #git
14:23 wget cbreak: Yep I know about this command, but applying it, I still have locally ther branches that have been removed from the tracked origin remote :-/
14:23 cbreak git remote prune origin doesn't remove them?
14:23 _ikke_ It doesn't
14:24 _ikke_ It only removes remote tracking branches
14:24 khfeng joined #git
14:24 wget indeed, not those locally fetched
14:24 MattMaker joined #git
14:24 _ikke_ wget: But no, there is no built-in way
14:25 _ikke_ wget: git branch -vv would show which are gone
14:25 wget _ikke_: damn. so the solution I came along with seems the only one ossible :-(
14:25 wget but my xargs command are not robust enough
14:25 wget *s
14:25 _ikke_ There is a better way
14:25 _ikke_ git update-ref
14:25 hobodave joined #git
14:26 phaleth joined #git
14:26 _ikke_ wget: can you show the output of git branch -vv ?
14:26 Sonderblade joined #git
14:27 michaelsharpe joined #git
14:28 cbreak _ikke_: yes, that's what I mean
14:29 cbreak it should remove the local remote tracking branches for origin
14:29 cbreak i.e. the local remotes/origin/foobar if foobar was removed on origin
14:29 _ikke_ Yes, but he want to move the local branches that tracked a remote branch that has been removed
14:29 cbreak sounds dangerous :/
14:30 UTAN_dev joined #git
14:30 dsdeiz joined #git
14:30 dsdeiz joined #git
14:32 zeroed joined #git
14:32 zeroed joined #git
14:33 lolmac joined #git
14:33 MattMaker joined #git
14:33 _rnsanchez_ joined #git
14:34 mizu_no_oto joined #git
14:35 irqq joined #git
14:36 kusfedora joined #git
14:36 kusfedora joined #git
14:36 rahtgaz joined #git
14:36 terminal_echo joined #git
14:38 MattMaker joined #git
14:38 Hounddog_ left #git
14:38 Rodya_ joined #git
14:39 isysd joined #git
14:40 _ADN_ joined #git
14:42 wget _ikke_: sorry was afk. here you are:
14:42 wget https://gist.github.com/wget/c000a50c5f2ae37005d985eca4aa01ce
14:42 _ikke_ wget: You see the "gone" part?
14:42 wget yep
14:42 _ikke_ Those are branches where the remote tracking branch is not there anymore
14:43 finalbeta1 joined #git
14:43 MattMake_ joined #git
14:43 wget maybe I should parse that output instead of taking all the branches that have been merged and deleted except master
14:44 wget because it isn't always master
14:44 _ikke_ right
14:44 apotry joined #git
14:44 _ikke_ there might be comming patches that add this to plumbing tools so the output is more stable / parseble
14:46 dminuoso joined #git
14:47 subhojit777 joined #git
14:47 Noldorin joined #git
14:48 MattMaker joined #git
14:51 endiruna joined #git
14:51 whitby joined #git
14:52 ghostlight joined #git
14:53 MattMake_ joined #git
14:54 thebope__ joined #git
14:55 e14_ joined #git
14:55 satbyy joined #git
14:56 robotroll joined #git
14:58 MattMaker joined #git
14:58 eivarv joined #git
14:59 thebope_ joined #git
15:00 shinnya joined #git
15:01 thebope_ joined #git
15:02 diogenese joined #git
15:02 NeverDie joined #git
15:04 nikivi joined #git
15:04 stomplee joined #git
15:04 Vinnie_win joined #git
15:04 lindenle joined #git
15:06 rubyonrailed joined #git
15:06 e14_ joined #git
15:07 MattMaker joined #git
15:12 MattMake_ joined #git
15:14 ResidentBiscuit joined #git
15:15 p4trix joined #git
15:15 ecsel joined #git
15:17 madewokherd joined #git
15:17 chrisshattuck joined #git
15:17 rgrinberg joined #git
15:17 MattMaker joined #git
15:18 rubie joined #git
15:22 MattMake_ joined #git
15:24 e14_ joined #git
15:25 mooru joined #git
15:26 mischat_ joined #git
15:26 mingrammer joined #git
15:26 peepsalot joined #git
15:27 MattMaker joined #git
15:27 Zamyatin joined #git
15:28 thiago joined #git
15:28 dreiss joined #git
15:32 MattMaker joined #git
15:32 godzirra joined #git
15:33 fdsafdsa joined #git
15:34 ahmedelgabri joined #git
15:38 synthroid joined #git
15:38 dimi1947 joined #git
15:40 ejb joined #git
15:41 romerocesar joined #git
15:41 rbr joined #git
15:41 abit2 joined #git
15:41 MattMaker joined #git
15:42 abit2 git branch -D branch-name    does??
15:42 berndl Deletes teh branch.
15:42 berndl *the
15:42 NeverDie_ joined #git
15:43 _ikke_ force delete
15:44 _ikke_ (deletes even when not merged)
15:45 GodGinrai joined #git
15:46 MattMake_ joined #git
15:48 raijin joined #git
15:49 Electrometro joined #git
15:50 newbie1 joined #git
15:50 Electrometro joined #git
15:50 venmx joined #git
15:51 Lomex joined #git
15:51 nonconvergent joined #git
15:51 MattMaker joined #git
15:52 e14_ joined #git
15:53 cbaylis joined #git
15:55 cbaylis why does "git cat-file -t <sometagname>" return "tag" in some repos, and "commit" in some others?
15:55 e14 joined #git
15:56 MattMake_ joined #git
15:56 editshare-ericf joined #git
15:56 ilmari annotated vs. lightweight tag
15:57 Gsham joined #git
15:57 ahmedelgabri joined #git
15:57 danielmahon joined #git
15:59 cbaylis ilmari: thanks, I didn't know the difference before, but that makes things clearer
15:59 cbaylis what's the recommended way to test if a tag name exists? "git rev-parse refs/tags/sometagname" or something else?
16:00 e14_ joined #git
16:01 MattMaker joined #git
16:02 grawity `git rev-parse --verify`
16:04 Prisoner-2460_1 joined #git
16:06 phanimahesh joined #git
16:06 MattMaker joined #git
16:06 leeN joined #git
16:10 rgrinberg joined #git
16:11 JanC_ joined #git
16:12 InfoTest joined #git
16:13 thiago cbaylis: what are you going to do if it exists? Or if it doesn't?
16:14 cbaylis thiago: decide whether or not to automatically do "git pull"
16:15 nd joined #git
16:15 MattMaker joined #git
16:16 l4v2 joined #git
16:16 l4v2 left #git
16:17 thiago cbaylis: ok, rev-parse is probably the right solution.
16:20 dsdeiz joined #git
16:20 MattMake_ joined #git
16:20 synthroid joined #git
16:21 synthroid joined #git
16:24 a_thakur joined #git
16:25 MattMaker joined #git
16:27 rgrinberg joined #git
16:30 MattMake_ joined #git
16:31 ansimation joined #git
16:31 Emperor_Earth_ joined #git
16:32 a_thaku__ joined #git
16:33 raijin joined #git
16:34 nikivi joined #git
16:35 nitric joined #git
16:35 MattMaker joined #git
16:35 cdown joined #git
16:35 LukeLR joined #git
16:39 chrisshattuck joined #git
16:40 MattMaker joined #git
16:41 tyreld joined #git
16:43 Silenced_v2 joined #git
16:44 SwiftMatt joined #git
16:46 pijiu joined #git
16:46 encod3 joined #git
16:47 nonconvergent joined #git
16:47 pijiu joined #git
16:48 Borkr joined #git
16:49 MattMaker joined #git
16:52 mutex joined #git
16:52 rbr joined #git
16:54 MattMaker joined #git
16:54 actinide joined #git
16:56 nonconve1gent joined #git
16:59 MattMaker joined #git
16:59 shgysk8zer0 joined #git
17:00 Darren__ joined #git
17:00 godzirra joined #git
17:00 diogenese joined #git
17:01 mikecmpbll joined #git
17:01 mingrammer joined #git
17:03 eivarv joined #git
17:03 nonconve1gent left #git
17:04 Gnut joined #git
17:04 thebope joined #git
17:04 MattMake_ joined #git
17:04 cdown joined #git
17:06 vuoto joined #git
17:09 MattMaker joined #git
17:11 cdown joined #git
17:12 Lomex joined #git
17:13 cagedwisdom joined #git
17:13 Gnut Hey everyone I have a question about somehting a bit complicated. We have a branch that someone decided to merge master into, then a rebase occurred so all the commits it grabbed from master were re-committed.
17:14 robotroll joined #git
17:14 MattMake_ joined #git
17:14 eivarv joined #git
17:15 Gnut What's the best way to create a new branch off master and rebase only the commits that weren't merged into the other branch on top
17:15 cdown_ joined #git
17:17 mizu_no_oto joined #git
17:17 rorro joined #git
17:17 jost joined #git
17:18 EvilPeng1 joined #git
17:19 grawity at first thought, this seems like a regular `git checkout mybranch && git rebase master`
17:19 Limes left #git
17:19 grawity I remember rebase being smart enough to skip duplicates... sometimes
17:20 Limes joined #git
17:20 chrisshattuck joined #git
17:20 rory Gnut: does anyone have a copy of the repo from before it happened?
17:20 grawity the person who did the rebasing probably does – the reflog.
17:21 Limes left #git
17:21 rory Gnut: Are you familiar with interactive rebase?
17:21 bernardio joined #git
17:21 subhojit_ joined #git
17:21 Limes joined #git
17:21 Limes left #git
17:22 shgysk8zer0 joined #git
17:22 Limes joined #git
17:23 irqq joined #git
17:23 Limes left #git
17:23 Limes joined #git
17:23 MattMaker joined #git
17:23 Gnut rory: the branch hasn't been merged into master yet, it just had master merged into it
17:24 cdown joined #git
17:24 Gnut rory: yes, i use it for cleaning up my branch before i push it for a PR
17:24 mehola joined #git
17:24 lolmac joined #git
17:24 rory Gnut: interactive rebase the branch to master, and skip the merge commit?
17:25 thiago joined #git
17:26 thebope joined #git
17:26 rbern joined #git
17:28 Sasazuka joined #git
17:28 salamanderrake_ joined #git
17:29 dreiss joined #git
17:30 roentgen joined #git
17:30 roentgen joined #git
17:31 Gnut rory: I'll give that a try
17:32 cdown_ joined #git
17:32 ayjay_t joined #git
17:34 SteffanW joined #git
17:34 isysd joined #git
17:36 sdothum joined #git
17:38 chrisshattuck joined #git
17:38 MattMaker joined #git
17:40 antipsychiatry joined #git
17:40 salamanderrake joined #git
17:41 rubyonrailed joined #git
17:41 LionsMane joined #git
17:43 jungsubk joined #git
17:46 ahmedelgabri joined #git
17:48 MattMaker joined #git
17:50 tabrez joined #git
17:56 mingrammer joined #git
17:56 mooru joined #git
17:57 multi_io joined #git
17:57 romerocesar joined #git
17:58 mingrammer joined #git
17:58 MattMaker joined #git
17:58 drizztbsd joined #git
17:59 z8z joined #git
18:00 salamanderrake_ joined #git
18:03 rgrinberg joined #git
18:04 pandeiro joined #git
18:04 Vivekananda joined #git
18:07 MattMaker joined #git
18:08 leeN joined #git
18:10 cdown joined #git
18:11 netj_ joined #git
18:11 chrisshattuck joined #git
18:12 Milos joined #git
18:15 salamanderrake joined #git
18:15 dsantiago joined #git
18:15 EvilPenguin joined #git
18:17 MattMaker joined #git
18:18 glebihan__ joined #git
18:19 lindenle joined #git
18:19 iota_pi Is it possible for me to pull a zipfile of a specific commit hash from github without using the git binary? Maybe through curl or wget?
18:19 iota_pi ...or scp?
18:19 mohabaks joined #git
18:21 MacWinner joined #git
18:21 Lauven joined #git
18:22 ToxicFrog iota_pi: of course. Just look at the request the browser makes to do that and replicate it.
18:22 ToxicFrog I think it's just a simple URL, no POST wackiness
18:23 ToxicFrog iota_pi: yeah, it's just https://github.com/${USER}/${REPO}/archive/${COMMIT}.zip
18:24 iota_pi oh it's easier than I thought. Awesome. Thanks.
18:24 Lauven joined #git
18:25 ToxicFrog (breadcrumbs: go to project in browser, commit history, click on a commit, then "browse files", "download", "download as zip")
18:28 afuentes joined #git
18:29 gkatsev joined #git
18:29 eivarv joined #git
18:31 MattMaker joined #git
18:36 MattMaker joined #git
18:38 kingarmadillo joined #git
18:42 dviola joined #git
18:43 nikivi joined #git
18:45 MattMake_ joined #git
18:48 atomicb0mb joined #git
18:49 atomicb0mb joined #git
18:50 GodGinrai joined #git
18:52 mingrammer joined #git
18:53 satbyy joined #git
18:53 satbyy joined #git
18:54 theskillwithin joined #git
18:56 ruskie joined #git
18:57 chibbol joined #git
18:58 rubyonrailed joined #git
18:58 jknetl joined #git
18:59 tax joined #git
19:00 rgrinberg joined #git
19:00 stamina joined #git
19:01 dreiss joined #git
19:02 shinnya joined #git
19:05 e14_ joined #git
19:06 robotroll joined #git
19:08 FB joined #git
19:08 rbr joined #git
19:08 Vinnie_win_f joined #git
19:10 duderonomy joined #git
19:12 choki joined #git
19:12 apotry joined #git
19:17 barteks2x joined #git
19:20 Faylite joined #git
19:21 wget iota_pi: By a single zip, you mean the entire repository like it was at that commit id or only the changes introduced by that commit id (like a patch)?
19:21 iota_pi The entire repo as it was at that time.
19:21 Repox joined #git
19:22 fahadash joined #git
19:22 dsdeiz joined #git
19:23 mizu_no_oto joined #git
19:23 michaelsharpe joined #git
19:28 dsdeiz joined #git
19:34 tgummerer joined #git
19:34 cilkay joined #git
19:35 Gsham joined #git
19:38 cilkay Hello. I've converted a Mercurial repo to Git using the hg-fast-export tool. It looks fine when I browse it with gitg or GitDAG. I then did "git clone --bare myrepo myrepo.git" and used scp to copy myrepo.git to a server on which we have hg repos that the team accesses via ssh.
19:39 mkoskar joined #git
19:39 cilkay I then did "git remote add user@remoteServer:/absolut/path/to/myrepo.git" in myrepo.
19:40 grawity I think you should've used `git clone --mirror` instead, which implies --bare but also preserves the exact same branch names (instead of translating to origin/whatever)
19:41 cilkay No matter which tool I use, I can't push to origin/master. I get a "does not appear to be a git repository" error. I'm certain that the repo is there.
19:42 grawity can you fetch from it, or `git ls-remote` it?
19:42 cilkay grawity: That's easy enough to change. It takes 2 minutes to convert.
19:42 grawity can you at least `ssh user@remoteServer ls -l /path/to/myrepo.git`?
19:43 cilkay I can do the latter since I used scp to transfer myrepo.git there
19:48 jorj joined #git
19:48 sagerdearia joined #git
19:49 grawity can you also do `ssh user@remoteServer git-upload-pack /path/to/myrepo.git
19:50 cilkay It gets curiouser and curioser. ls -l /home/cds/git-repos/api-server.git works. ls -l /home/cds/git-repos/api-server⁠⁠⁠⁠.git does not. I suspect a wayward Unicode character somewhere.
19:51 cilkay The issue must be the difference between those two strings.
19:51 cilkay They look the same but they obviously aren't.
19:52 grawity there's a stray U+FEFF (byte order mark) before ".git"
19:54 tabrez joined #git
19:55 ojacobson zero-width non-breaking space, or, the "fuck you" character
19:56 grawity it's still also a BOM
19:56 ojacobson yeah
19:58 romerocesar joined #git
20:01 stamina joined #git
20:02 cilkay I just saw that when I compared the two strings in a hex editor. Aaargh!
20:02 e14_ joined #git
20:06 Jellyg00se joined #git
20:06 ghostlight joined #git
20:11 mikecmpbll joined #git
20:13 cilkay I've been able to push but there are a bunch of other issues now, which I suspect might be addressed by the --mirror option. It's choking on various tags and heads saying that they don't point a valid object. It's probably due to the fact that the name of the default branch was changed to master.
20:13 cilkay ... on conversion.
20:14 abit2 not able to find a  branch in my local copy of my computer ??
20:14 Vaelatern joined #git
20:15 Narwhaal joined #git
20:15 Electrometro joined #git
20:16 re1 joined #git
20:16 Lowell joined #git
20:17 [diecast] joined #git
20:18 emilyrose joined #git
20:19 rubyonrailed joined #git
20:19 axl__ joined #git
20:20 actinide joined #git
20:20 re1 joined #git
20:21 le_melomane joined #git
20:21 mohabaks joined #git
20:22 blahdodo joined #git
20:23 cagedwisdom joined #git
20:23 mohabaks joined #git
20:24 gigq joined #git
20:24 JSharpe joined #git
20:25 danimal2 joined #git
20:25 johnmilton joined #git
20:27 boombatower joined #git
20:28 e14_ joined #git
20:29 shgysk8zer0 joined #git
20:29 axl__ joined #git
20:33 mischat joined #git
20:35 nwkj86 joined #git
20:35 liefer joined #git
20:38 netj joined #git
20:39 aidalgol joined #git
20:40 mingrammer joined #git
20:40 hebele joined #git
20:41 sq-one joined #git
20:44 dendazen joined #git
20:44 ningu joined #git
20:45 ningu suppose I make a feature branch and want to incorporate it back into the master branch. how do I decide whether to use merge or rebase? what are the pros/cons?
20:47 ningu this seems like a good overview: https://www.atlassian.com/git/tutorials/merging-vs-rebasing/conceptual-overview
20:49 ahmedelgabri joined #git
20:52 elastix joined #git
20:53 prawnzy joined #git
20:55 prawnzy joined #git
20:58 qsx Seveas: plan to attend 33c3?
21:01 neunon joined #git
21:02 ningu left #git
21:02 editshare-ericf joined #git
21:02 Silenced joined #git
21:04 Shentino joined #git
21:05 michaelsharpe joined #git
21:05 romerocesar joined #git
21:06 bongjovi joined #git
21:06 e14 joined #git
21:07 prawnzy joined #git
21:07 cyphase joined #git
21:07 Silenced joined #git
21:09 Seveas qsx: I won't be able to attend ccc unless they move it to a non-stupid date.
21:12 terminal_echo joined #git
21:15 realz joined #git
21:16 netj joined #git
21:17 izalove joined #git
21:17 izalove does git keep track of who pushed?
21:18 sagerdearia joined #git
21:19 qsx Seveas: it's so that the nerds have an excuse not to visit their family
21:19 SimonNa joined #git
21:20 faheem joined #git
21:20 blahdodo joined #git
21:20 Hobbyboy|BNC joined #git
21:20 `mx joined #git
21:21 Electrometro joined #git
21:21 kadoban joined #git
21:21 asdofindia joined #git
21:21 [cust] joined #git
21:21 KaibutsuX_ joined #git
21:21 rnsanchez_ joined #git
21:22 blackjid joined #git
21:22 krillr joined #git
21:25 blahdodo joined #git
21:26 netj joined #git
21:28 King_Hual joined #git
21:29 romerocesar joined #git
21:31 Gsham joined #git
21:32 romerocesar_ joined #git
21:35 mingrammer joined #git
21:35 re1 joined #git
21:36 godzirra joined #git
21:36 tgunr joined #git
21:37 nettoweb joined #git
21:39 Seveas qsx: yes, and I actually like my family, so screw the nerds
21:42 anuxivm joined #git
21:45 tgunr joined #git
21:48 nettoweb1 joined #git
21:50 nettoweb1 joined #git
21:52 spacelord joined #git
21:53 actinide joined #git
21:54 inflames joined #git
21:54 rgrinberg joined #git
21:54 |ifei5g00d joined #git
21:56 kingarmadillo joined #git
21:58 jrnieder joined #git
22:02 ShapeShifter499 joined #git
22:03 QwertyDragon joined #git
22:06 andlabs joined #git
22:09 Jellyg00se joined #git
22:10 brandonw joined #git
22:11 p4trix joined #git
22:18 satbyy joined #git
22:23 jrnieder joined #git
22:26 prawnzy joined #git
22:28 jeffreylevesque joined #git
22:29 mingrammer joined #git
22:32 Shentino joined #git
22:35 aidalgol joined #git
22:36 GT4066 joined #git
22:39 eaudi joined #git
22:39 sagerdearia joined #git
22:44 chrisshattuck joined #git
22:44 elsevero joined #git
22:45 aidalgol joined #git
22:46 shgysk8zer0 joined #git
22:47 thiago joined #git
22:51 kusfedora joined #git
22:56 johnmilton joined #git
22:58 jrnieder joined #git
23:01 texasmynsted joined #git
23:12 fullstack joined #git
23:14 mxmasster joined #git
23:15 swiftgeek joined #git
23:17 swiftgeek i'm a bit confused about Jeff King quote about escaping quotes
23:17 swiftgeek Git does not support it via git-am, but nor does it via
23:17 swiftgeek git-commit, etc.
23:17 mxmasster joined #git
23:17 swiftgeek ↑ it works for git commmit for me somehow
23:19 mxmasster left #git
23:20 King_Hual joined #git
23:21 * duarteper Hi all
23:21 fguillot joined #git
23:23 mingrammer joined #git
23:25 thiago swiftgeek: what quotes?
23:26 swiftgeek "[PATCH] mailinfo: unescape quoted-pair in header fields"
23:26 swiftgeek like From: "Author \"The Author\" Name" <somebody@example.com>
23:27 jrnieder joined #git
23:27 swiftgeek i committed something like that successfully to coreboot
23:27 romerocesar joined #git
23:28 Seveas swiftgeek: Only "\"The author\"" doesn't preserve quotes
23:28 Seveas that's what jeff is saying, and I just verified that with 2.10
23:29 mkopriva joined #git
23:30 swiftgeek somehow it worked for me on 2.9.3
23:30 swiftgeek Seveas: https://review.coreboot.org/cgit/coreboot.git/commit/?id=7e516fbb47daf2affc72135db68f2c9bd8b97dd5
23:30 swiftgeek the git commit, not git am
23:31 thiago swiftgeek: did that commit properly with git am?
23:32 swiftgeek just git commit, not with git am
23:32 thiago so it worked with git commit?
23:32 swiftgeek yes
23:32 thiago and that From: is what git format-patch produced?
23:33 swiftgeek git format-patch is fine
23:33 thiago it produced that From: line, right?
23:33 thiago and git am did not properly interpret it?
23:33 swiftgeek yep
23:33 thiago just plain bug then
23:39 swiftgeek i'm just not sure whether i understand jeff king correctly
23:45 NeverDie joined #git
23:46 thiago I don't know who jeff king is and what he said
23:47 tgunr joined #git
23:47 * swiftgeek neither, got message from mainlinglist
23:47 ayjay_t joined #git
23:51 NeverDie_ joined #git
23:53 xlegoman joined #git
23:54 justanotheruser joined #git

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