| 2009-12-24 |
| → bigmack83 joined | 00:01 |
|
subsonic
| I don't understand that when I swtich to a branch, modify a file, and then commit, it says I need to git add or git commit -a | 00:01 |
| ← WinterWeaver left | 00:01 |
|
cbreak
| what do you want to commit? | 00:01 |
|
| the changed file? | 00:01 |
|
subsonic
| yeah | 00:02 |
|
cbreak
| then add the file | 00:02 |
|
subsonic
| I'm just running a test | 00:02 |
| → jablko joined | 00:02 |
|
subsonic
| WHy does it need to be added when it's marked as modified? | 00:02 |
|
cbreak
| because git doesn't commit all changes | 00:02 |
|
| only those you added | 00:02 |
|
subsonic
| ohhhh, does it need to be "added" to the commit list? | 00:02 |
|
cbreak
| that way, you can commit a small bugfix while working at something else | 00:02 |
|
shd
| subsonic: git commit -a adds all modified files | 00:02 |
|
brian_panneton
| I have set up the server to host all the files and it uses git for collaboration. Each dev is going to make their own clone of the files. They edit them and want to send them back to the server. However, I don't want to push this? I don't understand why. | 00:02 |
|
shd
| subsonic: or git commit file | 00:02 |
|
subsonic
| ahh, okay | 00:03 |
|
| it's like the reverse of SVN in that way | 00:03 |
|
shd
| subsonic: the logic is that git requires commits to be "staged" | 00:03 |
|
subsonic
| yeah | 00:03 |
|
shd
| subsonic: staging means the user indicates which files will go to the commit. you can have multiple files modified, but you can commit only some of it in logical pieces | 00:04 |
|
subsonic
| right. is there a way to list which files are staged? | 00:04 |
|
shd
| subsonic: git status | 00:04 |
| ← Sky[x] left | 00:05 |
|
subsonic
| ahh, okay | 00:05 |
|
jablko
| i cloned a repo on one machine and made some commits | 00:05 |
|
shd
| # Changes to be committed: | 00:05 |
|
jablko
| now i cloned the same repo on another machine | 00:05 |
|
shd
| <list of files to be committed> | 00:05 |
| → vital joined | 00:05 |
|
shd
| Changed but not updated: | 00:05 |
|
jablko
| and wanna pull the commits from the first machine to this one | 00:05 |
|
shd
| <list of files that are modified but that will not be committed> | 00:06 |
|
jablko
| but git pull says you asked me to pull w/o telling me which branch | 00:06 |
|
shd
| subsonic: staging allows one to pick-and-choose those parts of the tree should go in on a single commit | 00:06 |
| ← schacon left | 00:06 |
|
subsonic
| shd: yep :) | 00:07 |
| ← DavidKlein left | 00:07 |
|
shd
| subsonic: it's pretty convenient actually, once you get used to it | 00:07 |
| → arod__ joined | 00:07 |
|
subsonic
| I have files added (staged) how can I see the diffs of those? | 00:07 |
|
cbreak
| jablko: you need to pull a branch | 00:07 |
|
| which branch do you want to merge? | 00:07 |
|
| git diff --cached | 00:07 |
|
shd
| git commit -a more or less emulates the svn default behaviour | 00:07 |
|
| subsonic: git diff HEAD | 00:07 |
|
cbreak
| difference between what is staged and what is HEAD | 00:07 |
|
shd
| subsonic: or, git diff HEAD -- filelist | 00:07 |
|
cbreak
| no | 00:07 |
|
| git diff HEAd will show you the difference between what is modified and HEAD | 00:08 |
|
| that is not neccessarily the same | 00:08 |
|
zoran119
| hi people, i have a master bare repo and two developers pushing to it. one developer made a new branch and pushed it, the other one made a few commits on top of that same branch, then merged it into master and pushed it. how do we now clean up and delete this merged branch from the central bare repo and the developers local repos? | 00:08 |
| → loincloth joined | 00:08 |
|
jablko
| cbreak: thanks - not sure - on both machines i just ran "git clone ... git commit ..." | 00:08 |
|
| w/o noticing the branch | 00:08 |
|
shd
| git diff HEAD is what svn people expect | 00:08 |
|
cbreak
| maybe master? | 00:08 |
|
jablko
| cbreak: master worked | 00:09 |
|
| thanks | 00:09 |
| ← loinclot_ left | 00:09 |
| ← arod__ left | 00:09 |
|
cbreak
| good :) master's the default branch | 00:13 |
| ← Geralt left | 00:13 |
| ← tbuehlmann left | 00:15 |
| ← brett_h left | 00:16 |
| → idletask joined | 00:21 |
|
idletask
| Hello | 00:21 |
| ← subsonic left | 00:21 |
| ← erlehmann left | 00:21 |
| → fynn joined | 00:28 |
| ← kithpom left | 00:29 |
|
fynn
| Yo. How do I set core.excludesfile to be ~/.excludesfile? The following in ~/.gitconfig doesn't work | 00:29 |
| → loinclot_ joined | 00:29 |
|
fynn
| [core] | 00:29 |
|
| excludesfile = $HOME/.gitexcludes | 00:29 |
|
Pieter
| s/$HOME/~/ | 00:30 |
| ← Bryson left | 00:30 |
|
sitaram
| RobertLaptop: pingg | 00:30 |
|
RobertLaptop
| Hi Sitaram | 00:30 |
|
sitaram
| RobertLaptop: what does "user key and git key dont match" even mean? (just woke up, need help understanding what you meant!) | 00:30 |
| ← maxter left | 00:30 |
|
fynn
| Pieter: doesn't work, unfortunately :/ | 00:31 |
| ← mbroeker left | 00:31 |
|
RobertLaptop
| I was looking at this line of code. | 00:32 |
|
| + next if $user =~ /^(gitweb|daemon|\@all)$/ or ( $user_list{$user} || '' ) eq 'has pubkey' or ( $internal{$user} || 0 ) == 1; | 00:32 |
|
| I take that back. It was this message "print STDERR "WARNING: pubkey $pubkey exists but $user already used as an external name\n"" | 00:32 |
|
idletask
| Feature request: when you git pull, git will automatically fetch + merge from the correct remote repository and then branch, as known in the config file | 00:33 |
|
sitaram
| RobertLaptop: let rephrase my question. What does "git key" even mean? | 00:33 |
| → mithro joined | 00:33 |
|
sitaram
| s/let/let me | 00:33 |
|
idletask
| However, suppose you want to dissociate fetch and merge: you need to git fetch theremote (or --all), but you need to specify the remote branch for git merge, even if you want the default | 00:33 |
|
fynn
| Looks like a patch for this might have gotten rejected? | 00:34 |
|
| http://n2.nabble.com/PATCH-Support-core-excludesfile-gitignore-td742043.html | 00:34 |
|
RobertLaptop
| It means a Key used inside ssh to access git instead of a key used inside ssh to access a shell. | 00:34 |
|
idletask
| The feature would be; when on a tracking branch, "git merge" merges from the appropriate remote branch | 00:34 |
|
| I mean, "git merge" with no other argument | 00:35 |
| ← Yuuhi` left | 00:35 |
|
RobertLaptop
| Sitaram my typing sucks really bad and I likely missed a few words. | 00:35 |
| ← amirite left | 00:36 |
| → Paraselene_ joined | 00:37 |
|
sitaram
| RobertLaptop: ok you meant a gitolite key versus shell key. Anyway that code has nothing to do with that, and I have no checks anywhere to make sure they dont match. | 00:37 |
|
charon
| idletask: how does 'git pull' not cover that case already? | 00:38 |
|
sitaram
| RobertLaptop: and with the new @SHELL feature, it may not even be needed after the first install | 00:38 |
|
RobertLaptop
| sitaram It means I miss read the patch. Wouldn't be the first time, | 00:38 |
|
sitaram
| np :) | 00:38 |
|
fynn
| OK, does anyone know whether it's possible to use $HOME or ~ when specifying the path to core.excludefiles? | 00:40 |
|
| if not, seems like .gitconfig's would become non-portable. | 00:40 |
|
idletask
| charon: git pull will fetch | 00:41 |
|
fynn
| the docs says it is possible, but it's not working on 1.6.5 | 00:42 |
|
| err, 1.6.4 | 00:42 |
|
felipec
| hmm, git ls-files -x $patern doesn't work | 00:42 |
| → amirite joined | 00:43 |
|
charon
| idletask: oh, apparently i missed the first half of your explanations. but how is that bad? admittedly someone might sneak commits into your merge, but you can just as well 'git pull; git log HEAD@{1}..' to look at it after the merge (and possibly 'git reset --merge HEAD@{1}' if you're not happy) instead of holding off the merge until you're sure you want to do it | 00:43 |
| → hummingkid joined | 00:43 |
| ← Sonderblade left | 00:44 |
|
zoran119
| how do i tell if a remote branch has been deleted? | 00:44 |
|
idletask
| charon: well, I just thought that since git pull is smart enough to know which remote branch you track, git merge could be as well | 00:44 |
| ← loincloth left | 00:45 |
|
charon
| felipec: read http://thread.gmane.org/gmane.comp.version-control.git/135219 | 00:46 |
|
| idletask: true | 00:46 |
|
| charon shrugs | 00:46 |
|
charon
| idletask: you can probably even write a small alias using git for-each-ref --format="%(upstream)" $(git symbolic-ref HEAD) | 00:47 |
|
| modulo error checking, and note that git-pull has its own logic | 00:47 |
|
idletask
| charon: uh, that's much too advanced git usage for me right now | 00:48 |
| → unomi1 joined | 00:48 |
|
| idletask tries this command | 00:48 |
| ngw → ngw|away | 00:49 |
| ← aconbere left | 00:50 |
|
idletask
| git for-each-ref --format="%(upstream:short)" $(git symbolic-ref HEAD) indeed shows the argument which could be used in this case | 00:51 |
|
felipec
| charon: ah :( | 00:53 |
| ngw|away → ngw | 00:54 |
| → zoran119_ joined | 00:54 |
| ← JakeSays left | 00:54 |
| ← hummingkid left | 00:55 |
|
RobertLaptop
| sitaram you could update the install script to add the new user into the shell group and not need the 2nd key at all anymore. :) | 00:55 |
| → loincloth joined | 00:56 |
| ← vbgunz left | 00:58 |
| ← fynn left | 00:58 |
| ← zoran119_ left | 00:58 |
| jelmer_ → jelmer | 00:59 |
| → multi_io joined | 00:59 |
| → aquabu joined | 00:59 |
| ← JasonWoof left | 01:00 |
| ← unomi left | 01:00 |
| ← loinclot_ left | 01:01 |
| ngw → ngw|away | 01:01 |
|
multi_io
| hm...so... if I push a branch that is checked out in the target repo I'm pushing to, then the HEAD (i.e. the (pointer to the) checked-out commit) in that repo is advanced to the most recent pushed commit on that branch? | 01:01 |
| ← jablko left | 01:02 |
| → jablko joined | 01:03 |
| ← cousine left | 01:07 |
| ngw|away → ngw | 01:08 |
| → exygen joined | 01:09 |
| → jinzhu joined | 01:09 |
| ← brian_panneton left | 01:10 |
| ← tedoc2000 left | 01:11 |
| ← nsussman left | 01:12 |
|
sitaram
| RobertLaptop: catch-22 | 01:17 |
| ← zoran119 left | 01:18 |
| → JasonWoof joined | 01:18 |
| ← loincloth left | 01:18 |
| → loinclot_ joined | 01:18 |
| ← nadavoid left | 01:18 |
| → dgarr joined | 01:18 |
|
sitaram
| RobertLaptop: possible to break the catch-22 but downside: if something goes wrong within gitolite they might be locked out completely, so I have to make sure they have password access for emergencies | 01:19 |
|
| RobertLaptop: nothing should go wrong, but being worried it might is a good thing :) | 01:19 |
| → nadavoid joined | 01:20 |
| ← vital left | 01:21 |
|
RobertLaptop
| Yea but then again. I somehow managed to make both the key's the same and was locked out anyway. Lukely for me that account was create just for git so I could SU in and fix things | 01:21 |
| → vital joined | 01:21 |
| ← cenuij left | 01:21 |
| ← felipe left | 01:21 |
| → babilen_ joined | 01:21 |
|
RobertLaptop
| sitaram ^^ | 01:21 |
| ← nadavoid left | 01:23 |
|
RobertLaptop
| sitaram If you don't have admin access and you manage to cross the keys you are in bad shape. | 01:23 |
|
sitaram
| RobertLaptop: not sure how you did that. Gitolite requires a "shell access without password" key and it does not touch that -- the worst that can happen is that gitolite won't work for you, but the shell will (because even if you use the same key, the "shell" line comes first) | 01:25 |
|
| RobertLaptop: I'd like to see a repeatable case of starting with a "shell access" key and locking yourslef out of that while installing gitolite | 01:25 |
| → cenuij joined | 01:26 |
| → tekky joined | 01:26 |
| ← jspiros left | 01:26 |
| → jspiros joined | 01:27 |
|
sitaram
| assuming ssh-add -l quirkiness is taken care of...! | 01:29 |
|
RobertLaptop
| It wasn't gitolite. It was me. | 01:29 |
|
sitaram
| RobertLaptop: and if I do what you first asked, it *could* be gitolite. I *never* want that :) | 01:30 |
|
RobertLaptop
| Sitaram It was about 10 crises ago so I can't remember what broken and how. | 01:30 |
|
sitaram
| I can look it up; but it was something to do with ssh-add -l IIRC | 01:31 |
|
| were you the one who did everything as root? no that was someone else I think...! | 01:31 |
| ← jspiros left | 01:32 |
| → jspiros joined | 01:32 |
|
RobertLaptop
| I think it was more about me tiring to manage both a testing gitolite repo and a production gitolito repo | 01:32 |
|
| from the same users | 01:32 |
|
sitaram
| that should work fine! It's even documented how to do that | 01:32 |
|
| in particular, that should not result in any lock outs | 01:33 |
|
RobertLaptop
| Well I get my keys all mixed up. | 01:33 |
| ← idletask left | 01:34 |
|
sitaram
| aah well.. anyway, you see my problem? being able to say "gitolite cant lock you out" is important to me... so... I wont break that initial install's dependence on 2 keys | 01:34 |
| → idletask joined | 01:35 |
|
RobertLaptop
| It had nothing to do with gitolite and more with me tiring to do stuff manually. | 01:35 |
|
sitaram
| :) | 01:35 |
|
Tronic
| Should git filter-branch --tree-filter "script that modifies some files" automatically notice the modifications? | 01:35 |
| ← babilen left | 01:35 |
|
Tronic
| Because here it doesn't appear to do so. | 01:35 |
| → totty joined | 01:35 |
| → mugwump joined | 01:35 |
|
Tronic
| The documentation says that it will notice any new or deleted files, but it says nothing about modified files. | 01:36 |
| ← totty left | 01:36 |
|
RobertLaptop
| I took care of the issue by just using 2 users and not having to keep track of a bunch of keys. | 01:36 |
| ← jspiros left | 01:37 |
| → javatexan joined | 01:37 |
| → LiamH joined | 01:38 |
| ← bdowning left | 01:38 |
| → hipe joined | 01:39 |
| ← hipe left | 01:39 |
|
RobertLaptop
| siteram but in the end passwords would still have worked I just choice to su in and look around to see what I did. | 01:40 |
| → bdowning joined | 01:41 |
| → jspiros joined | 01:42 |
| ← hyperair left | 01:43 |
| → thiago joined | 01:44 |
| → godsflaw joined | 01:44 |
| → Tureba joined | 01:45 |
| → aconbere joined | 01:45 |
| ← justin-george left | 01:47 |
| ← tfarina left | 01:47 |
| ← peper left | 01:47 |
| → peper joined | 01:47 |
|
Tronic
| After the operation is complete, it says WARNING: Ref 'refs/heads/master' is unchanged | 01:47 |
|
| ... and nothing in the repository has changed, the changes are applied to the working copy, though. | 01:48 |
| → hipe joined | 01:50 |
| → tfarina joined | 01:50 |
| ← jspiros left | 01:51 |
| → tfarina_ joined | 01:52 |
| ← dreiss_ left | 01:53 |
| → JakeSays joined | 01:55 |
| → jspiros joined | 01:55 |
| ← Tank` left | 01:58 |
| ← hipe left | 01:58 |
| → hyperair joined | 01:58 |
| ← jessep left | 02:00 |
| ← jspiros left | 02:00 |
| → Tank` joined | 02:01 |
| ← tfarina_ left | 02:03 |
| ← nitrate left | 02:05 |
| → loincloth joined | 02:06 |
| → justin-george joined | 02:06 |
| ← intripoon left | 02:06 |
| ← thiago left | 02:08 |
|
Tronic
| I am starting to think that filter-branch is only suitable for adding/removing/moving files and not actually modifying them. | 02:09 |
| ← tfarina left | 02:09 |
| → nitrate joined | 02:09 |
| → hummingkid joined | 02:11 |
| → jspiros joined | 02:11 |
| → thiago_home joined | 02:11 |
|
me22
| Tronic: so remove one then add it back? | 02:12 |
|
Tronic
| Add as a new commit after the processing, you mean? | 02:13 |
| → benbeecher joined | 02:14 |
|
pigeon
| is there a git command to show the top level git tree path? | 02:16 |
|
thiago_home
| git rev-parse --show-cdup | 02:16 |
| → jessep joined | 02:16 |
|
benbeecher
| I'm new to git and trying to understand the distrubted part of it - if I clone a repository and issue a git branches I only see master - shouldn't I have pulled over the remote repositories branches with the clone? | 02:17 |
|
thiago_home
| benbeecher: you did | 02:18 |
|
| you cloned everything | 02:18 |
|
| you have to understand the distinction between branches on the remote, your branches, and the local copies of the remote branches | 02:18 |
|
| (the latter are often called "remote branches") | 02:18 |
|
| do a git branch -r | 02:18 |
|
benbeecher
| aha | 02:19 |
| ← loinclot_ left | 02:22 |
|
benbeecher
| ok - this is probably a good place to drop what I'm doing and read the manual. :) Thanks! | 02:22 |
| ← iron left | 02:23 |
| ← hobodave left | 02:24 |
|
pigeon
| thiago_home: thanks | 02:26 |
| → VVSiz_ joined | 02:26 |
| → upul joined | 02:26 |
| → MetaCosm_ joined | 02:28 |
| ngw → ngw|away | 02:32 |
| ← RobertMeta left | 02:33 |
| ngw|away → ngw | 02:33 |
| → loinclot_ joined | 02:36 |
| → felipe joined | 02:37 |
| → hipe joined | 02:38 |
| ← parasti left | 02:39 |
| → whaley joined | 02:39 |
| ← VVSiz left | 02:44 |
| ← yzlin left | 02:45 |
| ← jlilly left | 02:47 |
| → yzlin joined | 02:47 |
| ← spearce left | 02:47 |
| ← jessep left | 02:50 |
| ← loincloth left | 02:52 |
| → AAA_awright joined | 02:52 |
| ngw → ngw|away | 02:57 |
| ngw|away → ngw | 02:57 |
| ← zackattack left | 02:58 |
| ngw → ngw|away | 03:02 |
| ngw|away → ngw | 03:02 |
| ← LiamH left | 03:04 |
| → dreiss_ joined | 03:05 |
| ← j416 left | 03:08 |
| → kpreid_ joined | 03:11 |
| ← javatexan left | 03:12 |
| ← Tureba left | 03:13 |
| → jlilly joined | 03:18 |
| ← coolcat left | 03:18 |
| → idle_task joined | 03:18 |
| → cytrinox_ joined | 03:20 |
| ← aziz left | 03:27 |
| → LiamH joined | 03:28 |
| ← warthog9 left | 03:28 |
| ← t0rc left | 03:28 |
| ← Paraselene_ left | 03:30 |
| ← thiago_home left | 03:31 |
| ← idletask left | 03:33 |
|
| offby1 would never make a joke about sitaram's age | 03:35 |
| → t0rc joined | 03:35 |
|
sitaram
| would hit too close to home, huh? | 03:35 |
|
| :) | 03:35 |
| ← t0rc left | 03:36 |
| ← cytrinox left | 03:36 |
| cytrinox_ → cytrinox | 03:36 |
| → tarski joined | 03:43 |
| ← kpreid_ left | 03:45 |
| → kpreid_ joined | 03:45 |
|
| offby1 glances around nervously. | 03:46 |
| → loincloth joined | 03:46 |
| ← tarski left | 03:48 |
| ← loinclot_ left | 03:50 |
| → tjaway joined | 03:52 |
| ← ereslibre left | 03:54 |
| → ajonat joined | 03:57 |
| → jessep joined | 03:59 |
| ← hummingkid left | 04:00 |
| ← kukks left | 04:00 |
| ← ngw left | 04:00 |
| ← LRN left | 04:04 |
| → LRN joined | 04:05 |
| ← mithro left | 04:06 |
| ← JM left | 04:08 |
| ← tjafk left | 04:08 |
|
| Gitzilla wonders what offby1 has to be nervous about. | 04:11 |
| ← tekky left | 04:11 |
|
offby1
| nothing, nothing at all; why, I ... uh ... | 04:14 |
|
| <.-.< | 04:14 |
|
| >.-.> | 04:14 |
| ← ajonat left | 04:21 |
| ← benbeecher left | 04:23 |
| → ajonat joined | 04:29 |
| → sumangurung joined | 04:35 |
| ← loincloth left | 04:36 |
| → loincloth joined | 04:36 |
| ← LiamH left | 04:37 |
| → meanburrito920_ joined | 04:38 |
| → j416 joined | 04:39 |
| → munichlinux joined | 04:40 |
| → pinnacle joined | 04:40 |
| → loinclot_ joined | 04:42 |
| ← whaley left | 04:44 |
| → ybloc joined | 04:48 |
| → hobodave joined | 04:49 |
| → fcarrijo joined | 04:49 |
| ← fcarrijo left | 04:50 |
| → owen1 joined | 04:51 |
| → tekky joined | 04:52 |
|
owen1
| can i get rid of master and name one of my branches as the 'new' master? | 04:52 |
| ← Jak_o_Shadows left | 04:52 |
|
owen1
| or cut some commits from the master and then merge it with a branch i have? | 04:54 |
| ← jablko left | 04:54 |
| → rchavik joined | 04:55 |
| ← loincloth left | 04:57 |
|
j416
| owen1: man git-branch | 04:58 |
|
Gitbot
| owen1: the git-branch manpage can be found at http://git.or.cz/man/git-branch | 04:59 |
|
j416
| search for flag -m | 04:59 |
|
| it will let you rename a branch | 04:59 |
|
| not sure what you mean by "cut some commits", but maybe man git-reset will help you | 04:59 |
|
Gitbot
| j416: the git-reset manpage can be found at http://git.or.cz/man/git-reset | 05:00 |
|
ybloc
| lol ^ | 05:00 |
|
| handy little guy | 05:00 |
| → hummingkid joined | 05:00 |
| ← FunkeeMonk left | 05:00 |
| → lockers joined | 05:04 |
| ← ashokie_bear left | 05:05 |
| ← lockers left | 05:05 |
| ← pinnacle left | 05:05 |
| → loincloth joined | 05:06 |
| ← owen1 left | 05:06 |
| → psankar joined | 05:07 |
| ← loinclot_ left | 05:09 |
| → owen1 joined | 05:09 |
|
owen1
| i lost connection guys. anynoe replied? | 05:09 |
| ← aconbere left | 05:09 |
| ← madewokherd left | 05:11 |
| ← sumangurung left | 05:11 |
| → javatexan joined | 05:12 |
| → krynnotaur joined | 05:12 |
| ← dreiss_ left | 05:12 |
|
ybloc
| 20:58 j416: owen1: man git-branch | 05:13 |
|
Gitbot
| ybloc: the git-branch manpage can be found at http://git.or.cz/man/git-branch | 05:13 |
|
ybloc
| 20:59 j416: not sure what you mean by "cut some commits", but maybe man git-reset will help you | 05:13 |
|
Gitbot
| ybloc: the git-reset manpage can be found at http://git.or.cz/man/git-reset | 05:13 |
|
j416
| thank you ybloc | 05:13 |
|
ybloc
| j416: np | 05:13 |
|
j416
| 13:59 j416: search for flag -m | 05:14 |
|
| 13:59 j416: it will let you rename a branch | 05:14 |
|
| (git branch) | 05:14 |
| ← fcuk112 left | 05:17 |
| ← dgarr left | 05:18 |
| → roop joined | 05:19 |
| → cytrinox_ joined | 05:20 |
| → mithro joined | 05:20 |
| → FunkeeMonk joined | 05:23 |
| ← eletuchy left | 05:24 |
| ← godsflaw left | 05:25 |
|
owen1
| the reason i need this is i have a good branch that i want to use. if i'll merge it with master things will not work as i want. i just want to get rid of this master. | 05:25 |
| → ankit9 joined | 05:26 |
|
owen1
| ybloc: thanks, i'll look at git-branch manpage | 05:26 |
|
ybloc
| owen1: Be sure to thank j416, I only repeated what he had said after you left. | 05:27 |
|
j416
| owen1: perhaps git-reset is what you are after then | 05:27 |
|
owen1
| j416: i'll look into it. thanks | 05:28 |
|
j416
| owen1: I take it that you want to make master look like something else? | 05:28 |
|
| in which case it is very simple | 05:28 |
| ← MetaCosm_ left | 05:31 |
| → aconbere joined | 05:31 |
| → MetaCosm_ joined | 05:31 |
| ← MetaCosm_ left | 05:31 |
| ← aquabu left | 05:31 |
| ← javatexan left | 05:31 |
| → D[a]vey joined | 05:31 |
| → MetaCosm_ joined | 05:31 |
| ← MetaCosm_ left | 05:32 |
| → MetaCosm_ joined | 05:32 |
| ← MetaCosm_ left | 05:32 |
| ← justin-george left | 05:32 |
| → MetaCosm_ joined | 05:33 |
|
owen1
| j416: yeah | 05:33 |
| ← MetaCosm_ left | 05:33 |
|
j416
| owen1: git checkout master ; git reset <sha1> ; | 05:34 |
|
| where <sha1> is what you want your master to be | 05:34 |
| → MetaCosm_ joined | 05:34 |
| → dgarr joined | 05:34 |
|
j416
| it can be a sha1 of any commit, or the name of a branch | 05:34 |
|
| for instance. | 05:34 |
|
ybloc
| Ah, neat, that was going to be my question. | 05:34 |
|
j416
| :) | 05:35 |
|
ybloc
| Commit or branch sha1. | 05:35 |
|
j416
| ybloc: any valid ref | 05:35 |
| ← cytrinox left | 05:35 |
| cytrinox_ → cytrinox | 05:35 |
|
ybloc
| I see, neat tip. Thanks. | 05:35 |
|
j416
| git reset HEAD^ ; < undo your last commit, for example | 05:35 |
| ← MetaCosm_ left | 05:35 |
|
j416
| (won't touch work dir) | 05:36 |
| → MetaCosm_ joined | 05:36 |
|
owen1
| j416: works. easy! | 05:36 |
|
j416
| owen1: good good :) | 05:36 |
|
| notice your work dir hasn't changed | 05:36 |
|
owen1
| yeah | 05:36 |
|
j416
| git status ; will report changes | 05:36 |
|
| if you want to reset that as well, you can pass the flag --hard to git reset. | 05:37 |
|
| you can reset it now by just doing: git reset --hard ; | 05:37 |
|
| be sure to not have any changes you need | 05:37 |
| → vbgunz joined | 05:38 |
| ← vinse left | 05:40 |
| → xlistener joined | 05:41 |
| → charlenopires joined | 05:41 |
| ← vbgunz left | 05:43 |
| → nadavoid joined | 05:44 |
| ← nadavoid left | 05:44 |
| → brizly joined | 05:46 |
| → nadavoid joined | 05:48 |
| ← ybloc left | 05:48 |
| ← nadavoid left | 05:51 |
| → nadavoid joined | 05:52 |
| ← kmap left | 05:52 |
| ← nadavoid left | 05:52 |
| → becomingGuru joined | 05:54 |
| ← orafu left | 05:55 |
| → orafu joined | 05:56 |
| ← dgarr left | 06:02 |
| ← hobodave left | 06:02 |
| ← FunkeeMonk left | 06:03 |
| → hobodave joined | 06:03 |
| → gitster joined | 06:05 |
| ← wolog left | 06:07 |
| ← hobodave left | 06:07 |
| → amerine joined | 06:08 |
| → hobodave joined | 06:08 |
| ← tekky left | 06:11 |
| → ajonat_ joined | 06:11 |
| ← ajonat left | 06:12 |
| ← krawek left | 06:13 |
| → loinclot_ joined | 06:14 |
| → DavidKlein joined | 06:15 |
| → dreiss_ joined | 06:16 |
| → sh1m joined | 06:16 |
| ← loinclot_ left | 06:21 |
| → loinclot_ joined | 06:22 |
| ← dreiss_ left | 06:22 |
| → arod joined | 06:23 |
| → sis joined | 06:23 |
| ← sh1mmer left | 06:24 |
| sh1m → sh1mmer | 06:24 |
| ← brettgoulder left | 06:24 |
| ← arod left | 06:25 |
| ← bigmack83 left | 06:26 |
| ← charlenopires left | 06:26 |
| ← hipe left | 06:26 |
| ← loincloth left | 06:30 |
| → d0ugal joined | 06:32 |
| → tekky joined | 06:33 |
| ← d0ugal left | 06:33 |
| ← segher left | 06:36 |
| → roop_ joined | 06:37 |
| ← D[a]vey left | 06:37 |
| ← roop left | 06:39 |
| → solofight joined | 06:40 |
| → calin joined | 06:41 |
| ← AAA_awright left | 06:41 |
| ← solofight left | 06:41 |
| → AAA_awright joined | 06:41 |
| → solofight joined | 06:42 |
| ← jessep left | 06:44 |
| → jessep joined | 06:45 |
| ← hummingkid left | 06:45 |
| ← j416 left | 06:45 |
| → j416 joined | 06:45 |
| ← j416 left | 06:47 |
| → j416 joined | 06:47 |
| ← ajonat_ left | 06:48 |
| → warthog9 joined | 06:50 |
| → mbroeker joined | 06:53 |
| → numberGrey joined | 06:58 |
| babilen_ → babilen | 06:58 |
|
numberGrey
| Has anyone gotten this error before? git: 'pull' is not a git-command. See 'git --help'. | 06:58 |
|
| Did you mean this? | 06:58 |
|
| shell | 06:58 |
| → sis_ joined | 06:59 |
|
numberGrey
| I was trying to pull down to a small slice I had and am getting this on the pull command. I am pretty confused | 06:59 |
| ← Adlai left | 07:00 |
| → doener joined | 07:00 |
|
offby1
| me too | 07:03 |
|
| maybe the directory that holds the subcommands is missing, or unreadable, or something | 07:03 |
|
| I'd try "strace -f -o yow git pull", then examine the file "yow" for clues | 07:04 |
|
| ... if you're using Linux, anyway | 07:04 |
|
| on OS X you'd use "ktrace", but I forget the details. On Windows it's different again. | 07:04 |
| ← sis left | 07:05 |
| → mextex joined | 07:06 |
| mextex → mutex | 07:06 |
| → hummingkid joined | 07:07 |
| → bigmack83 joined | 07:18 |
| ← meanburrito920_ left | 07:20 |
| ← hummingkid left | 07:20 |
| → AMorozov joined | 07:21 |
|
AMorozov
| Hi! | 07:21 |
|
| How to configurate a branch to push with a different name to a remote?Right now I use smth like this: git push origin master:waved_processor to achieve what I need, but I would like to avoid explicit specification of branch names from command line. | 07:23 |
|
| I have already set up branch master as this:[branch "master"] remote = origin merge = waved_processorand it actually works for pull'ing. Now I need this for pushing too. | 07:24 |
| ← JackWinter left | 07:25 |
|
doener
| AMorozov: you can set push.default to "tracking", then instead of pushing all branch heads that exist locally and in the remote repo, push will only push the checked out branch head, to the configured upstream branch | 07:25 |
|
| AMorozov: of course, that means that you can only push a single branch head at a time | 07:26 |
| ← datechman left | 07:27 |
| ← JoeConyers left | 07:27 |
|
doener
| AMorozov: alternatively, you can set remote.origin.push to master:waved_processor, then pushes to "origin", will always default to pushing that branch head and renaming it | 07:27 |
|
| AMorozov: and you can have multiple config entries for that key, so you can push more than one ref | 07:28 |
| → wolog joined | 07:28 |
| → icwiener joined | 07:29 |
| → JoeConyers joined | 07:29 |
|
AMorozov
| doener: thank you. push.default seems to be the solution for my case. | 07:30 |
| ← j416 left | 07:30 |
|
rtyler
| blast, where's spearce when I want to bother him :D | 07:33 |
| ← bigmack83 left | 07:36 |
| → zackattack joined | 07:38 |
| → bigmack83 joined | 07:39 |
| ← davido left | 07:40 |
| ← tekky left | 07:41 |
| ← AMorozov left | 07:41 |
| → tekky joined | 07:47 |
| → tehbaut joined | 07:50 |
| → patrickmclaren_ joined | 07:55 |
| → jceb joined | 08:01 |
| → tek joined | 08:02 |
| ← xlistener left | 08:02 |
| ← tekky left | 08:03 |
| tek → tekky | 08:03 |
| → Sigma joined | 08:04 |
| → dreiss_ joined | 08:06 |
| → hobodave_ joined | 08:06 |
| ← hobodave left | 08:07 |
| → pantsman joined | 08:10 |
| ← mbernstein left | 08:11 |
| → RobertMeta joined | 08:17 |
| ← AkumaStreak left | 08:18 |
| → MetaCosm joined | 08:19 |
| ← psankar left | 08:19 |
| → Adlai joined | 08:19 |
| ← patrickmclaren_ left | 08:19 |
| ← numberGrey left | 08:20 |
| → loincloth joined | 08:20 |
| → fcuk112 joined | 08:20 |
| ← bigmack83 left | 08:22 |
| → maxter joined | 08:25 |
| ← RobertMeta left | 08:25 |
| → numberGrey joined | 08:26 |
| → droden joined | 08:26 |
| ← numberGrey left | 08:27 |
| → numberGrey joined | 08:28 |
| ← dreiss_ left | 08:33 |
| ← loinclot_ left | 08:35 |
| ← MetaCosm_ left | 08:36 |
| ← MetaCosm left | 08:39 |
| ← hobodave_ left | 08:40 |
| → Sonderblade joined | 08:40 |
| → MetaCosm joined | 08:40 |
| ← MetaCosm left | 08:41 |
| → MetaCosm joined | 08:41 |
| ← QshelTier left | 08:42 |
| → ngw joined | 08:48 |
| ← kristian42 left | 08:52 |
| → loinclot_ joined | 08:54 |
| ← sis_ left | 08:54 |
| → incredible joined | 08:55 |
| → kristian42 joined | 08:56 |
| ← loincloth left | 08:58 |
| → shruggar joined | 08:58 |
| → psankar joined | 09:06 |
| → j416 joined | 09:10 |
| → jelmer_ joined | 09:14 |
| → __iron joined | 09:16 |
| ← pantsman left | 09:16 |
| → corni joined | 09:17 |
| → Oejet joined | 09:20 |
| → idletask joined | 09:21 |
| → Patabugen joined | 09:21 |
| ← idle_task left | 09:22 |
|
idletask
| Hello everyone | 09:22 |
|
Patabugen
| Hey | 09:22 |
|
idletask
| I want to add an alias isremote that does git for-each-ref --format="%(upstream:short)" $(git symbolic-ref HEAD) | 09:22 |
|
| But whatever combination I try in the config file, I can't seem to get it right :( The command by itself works, the alias doesn't... | 09:23 |
| → intripoon joined | 09:23 |
| ← nkuttler left | 09:23 |
|
Patabugen
| Can anyone explain how git clone --mirror works? I think I'd like to use it to backup my git, by having a cron job git fetch every X minutes, am I correct? And how does that differ from --bare or a regular clone? | 09:25 |
| → hummingkid joined | 09:26 |
|
Patabugen
| messed up the colours there eh | 09:26 |
|
idletask
| Well, OK, found | 09:27 |
|
| You have to use the '!' alias form | 09:27 |
| ← loinclot_ left | 09:30 |
| → loincloth joined | 09:30 |
| → sgh joined | 09:30 |
| → Paraselene_ joined | 09:31 |
| ← fcuk112 left | 09:35 |
| → fcuk112 joined | 09:36 |
|
idletask
| Argh, but the ! form won't take arguments :( | 09:39 |
| ← JEEB left | 09:46 |
| ← unomi1 left | 09:49 |
|
wereHamster
| Patabugen: regular clones put the refs from the remote repository into refs/remotes/origin/, --mirror creates an exact mirror (refs/heads/foo in the remote repo is copied into refs/heads/foo in the local repo) | 09:49 |
| ← zackattack left | 09:50 |
| → unomi1 joined | 09:50 |
| ← j416 left | 09:50 |
| → rudi_s joined | 09:53 |
| ← jceb left | 09:54 |
|
Patabugen
| thanks wereHamster | 09:54 |
| → idle_task joined | 09:55 |
|
idle_task
| Argh, disconnected :( | 09:55 |
|
| Anyone has seen my question? | 09:55 |
|
Patabugen
| I did, but haven't a clue! sorry | 09:56 |
|
| by alias you mean a bash alias? | 09:56 |
| ← hummingkid left | 09:58 |
| → mtkd joined | 09:58 |
|
idle_task
| No | 09:59 |
|
| What you can configure in git with alias.* | 09:59 |
|
Patabugen
| Then I'm afraid I'm completly helpless! | 10:00 |
| ← rchavik left | 10:00 |
|
idle_task
| I have a favorite alias of mine: git config --global alias.cdiff "diff --cached" | 10:00 |
|
| But I want to know what remote branch, if any, a local branch tracks | 10:01 |
|
| Right now I can do that only if I'm on the current branch using the lengthy command above with HEAD instead of $1 | 10:01 |
|
Patabugen
| Would it be acceptably to use git clone --mirror to create a backup, then git fetch to keep it up to date? And if the original gets borked I can git clone --mirror in the opposite direction to get it working again? | 10:02 |
| ← jessep left | 10:02 |
|
wereHamster
| idle_task: you want to know which remote branch your current HEAD tracks, right? | 10:02 |
|
| Patabugen: yes | 10:02 |
|
Patabugen
| =D thanks! | 10:03 |
|
idle_task
| wereHamster: well, no, in fact, I'd like, say, "git showremote <thebranchhere>" to show what remote branch <thebranchhere> tracks | 10:04 |
|
| For the current HEAD, I can do that already using a ! alias | 10:04 |
|
| The problem is that the ! form won't take arguments, otherwise I'd just do refs/heads/$1 and it would work | 10:05 |
| → Axius joined | 10:05 |
|
idle_task
| I have tried this: whatremote = !git for-each-ref --format=\"%(upstream:short)\" refs/heads/$1 | 10:05 |
| ← Axius left | 10:06 |
|
idle_task
| Unfortunately, $1 is not expanded and this becomes refs/heads/ | 10:06 |
|
| Well, it is expanded to nothing | 10:06 |
| ← icwiener left | 10:06 |
| ← idletask left | 10:09 |
| ← rhett left | 10:10 |
|
wereHamster
| idle_task: you have to use !sh -c '...' | 10:13 |
|
| see http://git.or.cz/gitwiki/Aliases for examples | 10:14 |
|
idle_task
| Aiie, quoting hell :p | 10:14 |
|
| wereHamster: thanks! | 10:14 |
| ← Twisted_Mentat left | 10:15 |
| ← jelmer_ left | 10:15 |
| → tbuehlmann joined | 10:21 |
|
idle_task
| YES | 10:23 |
|
| whatremote = !sh -c 'git for-each-ref --format=\"%(upstream:short)\" $([ -z "$1" ] && git symbolic-ref HEAD || echo refs/heads/$1)' - | 10:23 |
|
| Thanks a lot, wereHamster | 10:23 |
| → khmarbaise joined | 10:23 |
| → loinclot_ joined | 10:24 |
|
wereHamster
| np | 10:27 |
| → jrockway joined | 10:27 |
|
idle_task
| It even takes the current branch if no argument is given | 10:29 |
|
| Lovely | 10:29 |
| ← nitrate left | 10:30 |
| → sgh_ joined | 10:30 |
| → loinclo__ joined | 10:30 |
|
doener
| idle_task: I'd use: git rev-parse --symbolic-full-name --default HEAD $1 | 10:31 |
|
| idle_task: that even won't explode when you're on a detached HEAD (symbolic-ref will, obviously, as HEAD isn't symbolic in that case) | 10:32 |
|
| idle_task: ... that's to replace the stuff in the subshell, the result of course still needs to be fed to for-each-ref | 10:33 |
| → d0k joined | 10:34 |
| GustavoR|aweh → GustavoRamos | 10:35 |
| ← mtkd left | 10:35 |
| → ustunozgur joined | 10:36 |
|
idle_task
| doener: trying that | 10:38 |
| → ixti joined | 10:38 |
| ← sgh left | 10:38 |
| ← doener left | 10:38 |
| → doener joined | 10:39 |
|
idle_task
| doener: yep, that works fine | 10:39 |
|
doener
| idle_task: :-) makes things a bit more readable (IMHO) | 10:39 |
| ← loincloth left | 10:39 |
|
idle_task
| I agree | 10:40 |
| ← ankit9 left | 10:43 |
|
idle_task
| I'm no advanced git user, unfortunately | 10:44 |
|
| But I learn | 10:44 |
| → hipe joined | 10:44 |
| ← psankar left | 10:45 |
|
doener
| idle_task: no problem with that. I didn't mean to complain that you did it "wrong", just show how it can be done better | 10:45 |
| → nitrate joined | 10:46 |
|
idle_task
| doener: I didn't take it as a complaint, don't worry :p | 10:46 |
| ← loinclot_ left | 10:46 |
|
idle_task
| But I think this could be in the default git, somehow... Like git branch --whattracks or something | 10:47 |
|
| Right now you have to either see the config file or do git remote show [-n] <remotename> | 10:47 |
|
doener
| idle_task: there was some discussion about master@{upstream} or so, but that somehow diead, I don't recall the reasons | 10:47 |
|
| s/diead/died/ | 10:48 |
|
idle_task
| Well, anyway - it's now in my toolbox and my git guide | 10:48 |
|
| :p | 10:48 |
|
| More annoyingly though, push lines have to be configured by hand | 10:50 |
|
| It'd be nice, on a git branch -t, to say that the push side is to be configured along with the pull side | 10:50 |
|
doener
| idle_task: well, there's push.default = tracking, but that limits you to "push the checked out branch head" | 10:51 |
|
idle_task
| Err | 10:51 |
|
| idle_task checks git-config manpage again | 10:51 |
|
doener
| idle_task: needs 1.6.3 | 10:52 |
| → hummingkid joined | 10:52 |
|
idle_task
| push.default is empty by default | 10:52 |
|
| Well, I have 1.6.5 | 10:52 |
|
doener
| which means the same as "push.default = matching" | 10:52 |
| → erlehmann joined | 10:53 |
|
idle_task
| So, if I git branch -t foo bar/baz and push.default = tracking, it means that not only will git pull will pull from bar.baz, but git push will actually push to bar/baz as well? | 10:54 |
| ← mithro left | 10:54 |
| → zackattack joined | 10:54 |
|
doener
| idle_task: yes, if you run "git push" (no refspec, not sure whether giving a remote disables it, too) | 10:54 |
|
| idle_task: ... and you have foo checked out, of course | 10:55 |
|
| idle_task: i.e. "git push" will act like "git push bar foo:baz" | 10:55 |
|
| idle_task: I'm pretty sure that remote.bar.push will override it though | 10:56 |
| ← vital left | 10:57 |
| → vital joined | 10:57 |
|
idle_task
| doener: well, that's something already | 10:57 |
|
| doener: as a side note, I saw that if you configure a push line but rename the branch with git branch -m, the push line is not updated | 10:58 |
|
| Is that on purpose? | 10:58 |
| ngw → ngw|away | 10:59 |
|
doener
| idle_task: I guess it's "nobody thought that that be useful" | 10:59 |
|
idle_task
| Well, setting push.default to tracking is better anyway | 11:00 |
|
| (imho) | 11:00 |
|
doener
| idle_task: dunno, I dislike "tracking" and "current" | 11:01 |
|
| idle_task: too error prone. git checkout -b foo origin/master; *work*; git checkout maste; git push *DAMN!* | 11:02 |
| ngw|away → ngw | 11:02 |
|
idle_task
| doener: ah, yes, it's true that you can have local refs with slashes in them | 11:02 |
|
doener
| idle_task: that's not what I meant. The important part was the typo in the second checkout (maste instead of master) | 11:03 |
|
| idle_task: so the push ended up pushing "foo" to origin's master | 11:03 |
|
| (instead of pushing master, as intended) | 11:03 |
|
idle_task
| Oh, so origin/master is actually refs/remotes/origin/master | 11:04 |
|
doener
| idle_task: the default "matching" thing is IMHO quite good. You have a set of branch heads you prepare to be good, and then push them out, and regardless of where you are, "git push" will only push out the "good" set | 11:04 |
|
| idle_task: renames are only needed what you have more than one remote, and they have conflicting names. And then I prefer push.default = nothing, with manual default push refspecs | 11:05 |
|
idle_task
| doener: well, it becomes problematic if you have several read/write remotes, as I do - although, yes, "git push <theremote> <localref>:<remoteref>" solves the problem easily | 11:05 |
|
doener
| idle_task: as you usually have only a few public branch heads (say master, maint, next), that's not that troublesome | 11:06 |
|
| idle_task: example: http://git.pastebin.com/m35ebb1c2 | 11:06 |
| → til joined | 11:06 |
|
doener
| idle_task: a "minimal risk" setup. You can only push the wrong thing with explicit refspecs, or when you mess up your local branch heads (and you can't really protect against that) | 11:07 |
|
idle_task
| doener: why the -f to git remote add, line 3? | 11:09 |
|
doener
| idle_task: that's "fetch", not "force" | 11:09 |
|
| ... yes, it's pretty damn misleading | 11:09 |
|
idle_task
| I just saw that in the manpage | 11:09 |
|
| Should be -F, I guess :p | 11:09 |
|
doener
| idle_task: i.e. it saves a "git fetch public" | 11:09 |
| → ph^_ joined | 11:09 |
| → tsal joined | 11:11 |
| ← wolog left | 11:12 |
| ← jinzhu left | 11:12 |
| → ayaz joined | 11:13 |
| ← tsal left | 11:15 |
| ← khmarbaise left | 11:23 |
| → Yuuhi joined | 11:23 |
| → thiago_home joined | 11:24 |
| ngw → ngw|away | 11:26 |
| ← loinclo__ left | 11:26 |
| → loincloth joined | 11:26 |
| ← ayaz left | 11:27 |
| ← ngw|away left | 11:28 |
| GustavoRamos → GustavoR|aweh | 11:28 |
| ← hummingkid left | 11:30 |
| → kumbayo joined | 11:31 |
| → icwiener joined | 11:32 |
| → JEEB joined | 11:32 |
| → Helios joined | 11:38 |
| ← brizly left | 11:39 |
| → Alfonzo joined | 11:40 |
| → ngw joined | 11:41 |
| fujin → xmasfu | 11:42 |
| ← hyperair left | 11:43 |
| ← Helios left | 11:46 |
| → Helios-- joined | 11:47 |
| → vbgunz joined | 11:48 |
| ngw → ngw|away | 11:50 |
| → WinterWeaver joined | 11:53 |
| ← unomi1 left | 11:53 |
| ngw|away → ngw | 11:53 |
| → unomi1 joined | 11:58 |
| → loinclot_ joined | 11:58 |
| ← sunil left | 12:01 |
| ← calin left | 12:02 |
| → bentob0x joined | 12:03 |
| ← loincloth left | 12:04 |
| → cousine joined | 12:04 |
| → khmarbaise joined | 12:05 |
| ← doener left | 12:07 |
| → doener joined | 12:07 |
| ← upul left | 12:09 |
| ← khmarbaise left | 12:12 |
| ← amerine left | 12:13 |
| → ereslibre joined | 12:15 |
| ngw → ngw|away | 12:18 |
| ← zackattack left | 12:22 |
|
magcius
| When trying to use msysgit, I get: "No supported authentication methods available" | 12:23 |
| → sumangurung joined | 12:24 |
| ← numberGrey left | 12:24 |
| → kmartiin joined | 12:24 |
| ← ia left | 12:27 |
| → whaley joined | 12:27 |
| ← sgh_ left | 12:31 |
| ngw|away → ngw | 12:31 |
| → x|ixti|x joined | 12:34 |
| ← sumangurung left | 12:35 |
| ← ixti left | 12:37 |
| ← shruggar left | 12:37 |
| ngw → ngw|away | 12:38 |
| ngw|away → ngw | 12:38 |
| ← roop_ left | 12:40 |
| → Octalot joined | 12:41 |
| → ia joined | 12:43 |
| → khmarbaise joined | 12:44 |
|
magcius
| huh. | 12:44 |
|
| $ git remote show origin | 12:44 |
|
| fatal: The remote end hung up unexpectedly | 12:44 |
| → sde_antrix1 joined | 12:45 |
| ← multi_io left | 12:46 |
|
idle_task
| magcius: what does git remote show -v says? | 12:48 |
|
| Err | 12:48 |
|
| git remote -v | 12:48 |
| ← selckin left | 12:49 |
| ← marcoecc left | 12:49 |
|
idle_task
| Question: a branch deletion can fail with "the branch you want to delete is not an ancestor of your current HEAD" - while I've read the git guide about this, I still don't understand what that means exactly | 12:50 |
|
| Where does that message come from exactly? Where is the danger in deleting such a branch? | 12:50 |
| → kukks joined | 12:50 |
|
magcius
| idle_task, origin [email@hidden.address] | 12:50 |
|
thiago_home
| idle_task: well, it's just a warning to tell you that it's not an ancestor of your current HEAD | 12:51 |
|
| i.e., it's not merged | 12:51 |
|
| if you want to ignore the warning, use -D instead of -d | 12:52 |
|
idle_task
| magcius: are you absolutely sure about this URL? If you have gitweb on this repo, is this the actual push URL for this branch? | 12:52 |
|
magcius
| idle_task, yeah, it's what github says to use. | 12:52 |
|
| hmm... cygwin didn't install ssh, do I need it? | 12:52 |
|
idle_task
| Well, yes you do :p | 12:52 |
|
| That's the problem, I guess | 12:53 |
|
| thiago_home: ah, so it's to avoid "dangling" heads? | 12:53 |
|
thiago_home
| idle_task: yes, and to let you know you haven't merged that branch yet | 12:53 |
|
idle_task
| thiago_home: OK, I _think_ I get it, but I have to make some experiments to be sure about that | 12:55 |
| → rchavik joined | 12:58 |
| ← jlilly left | 12:59 |
|
Ilari
| magcius: Remote end hung up unexpectedly immediately with ssh urls could indicate that executing ssh goes wrong. | 13:01 |
|
Blub\0_
| if I have a remote named A, a branch named B which is supposed to be pushed into A as master | 13:02 |
|
Ilari
| magcius: Also incorrect values of $GIT_SSH can cause that (with cygwin, it should usually be blank). | 13:02 |
|
Blub\0_
| like when I do: git push A B:refs/heads/master | 13:02 |
|
Ilari
| magcius: s/blank/not set/ | 13:02 |
|
Blub\0_
| how would I configure git to do that automatically that way? | 13:02 |
|
| git config remote.A.push <what>? | 13:02 |
|
Ilari
| Blub\0_: 'remote.A.push B:refs/heads/master'? | 13:03 |
|
Blub\0_
| oh right, I trued using 'git push B' then, but doing 'git push A' works | 13:03 |
|
idle_task
| I though that needed to be refs/heads/B? | 13:03 |
|
Blub\0_
| s/trued/tried/ | 13:03 |
| Blub\0_ → Blub\0 | 13:03 |
|
magcius
| Ilari, same thing. | 13:04 |
|
| Ilari, still can't git remote show origin | 13:04 |
|
Blub\0
| ty | 13:04 |
|
magcius
| even though ssh is installed, my keys are correct (I can ssh to [email@hidden.address] | 13:04 |
|
cbreak
| did you ssh-add the key? | 13:05 |
|
magcius
| yeah | 13:05 |
|
| hold on | 13:05 |
|
Ilari
| magcius: 'echo $GIT_SSH'? Blank? | 13:05 |
|
magcius
| Ilari, how do I unset an environment variable? | 13:05 |
|
thiago_home
| unset | 13:05 |
| → mtkd joined | 13:06 |
|
Ilari
| magcius: Yeah 'unset $GIT_SSH'. | 13:06 |
|
magcius
| $GIT_SSH or GIT_SSH? | 13:06 |
|
thiago_home
| without the $ | 13:06 |
| → loincloth joined | 13:07 |
| → aziz joined | 13:07 |
|
Ilari
| Yeah, no $ there. | 13:08 |
| → Bass10 joined | 13:09 |
| ← loinclot_ left | 13:10 |
| → fermion_ joined | 13:11 |
| → _Krieger_1 joined | 13:11 |
| → loinclot_ joined | 13:13 |
| → sgh joined | 13:14 |
| ← jepler left | 13:15 |
| ← TDJACR left | 13:16 |
| → jepler joined | 13:16 |
| ← cousine left | 13:17 |
| fermion_ → capndiesel | 13:17 |
| → LiamH joined | 13:18 |
| → bigmack83 joined | 13:20 |
| → Adlai` joined | 13:21 |
| → hummingkid joined | 13:22 |
| ngw → ngw|away | 13:22 |
| ← Adlai left | 13:22 |
| Adlai` → Adlai | 13:22 |
| ← drizzd left | 13:24 |
| → _buno_ joined | 13:24 |
| → drizzd joined | 13:24 |
| ngw|away → ngw | 13:25 |
| ← _buno_ left | 13:26 |
|
_Krieger_1
| i have a question about merging, i've described it here http://pastebin.ca/1725914 - please help me | 13:26 |
|
Ilari
| _Krieger_1: $faq non-bare may be relevant. | 13:28 |
|
Gitbot
| _Krieger_1: Pushing to non-bare repositories is discouraged. See http://git.or.cz/gitwiki/GitFaq#non-bare | 13:28 |
| ← loincloth left | 13:28 |
| → Sky[x] joined | 13:28 |
|
cbreak
| yeah, it seems you pushed to a repository with working directory | 13:29 |
|
| so the working directory looks changed compared to the new head | 13:29 |
|
| (that you pushed) | 13:29 |
|
| because it's still the old one | 13:29 |
| → TDJACR joined | 13:32 |
| ← bigmack83 left | 13:32 |
| ← tg left | 13:33 |
| → tg` joined | 13:34 |
| ← nanothief left | 13:34 |
| → bigmack83 joined | 13:34 |
| → pantsman joined | 13:34 |
| → parasti joined | 13:35 |
| → petrux joined | 13:37 |
| ← jjuran left | 13:38 |
| ← petrux left | 13:38 |
| → comawhite joined | 13:38 |
|
_Krieger_1
| Ilari, thanks, it helped | 13:38 |
|
comawhite
| I'm having issues with merging two branches. I created a branch called integration and pulled the merge request. I had some conflicts so I fixed them then did git add ., switched back to master and did git merge integration. but got Already up-to-date | 13:40 |
|
_Krieger_1
| and what is recommended practice for setting up local repositories between dev group and upstream code storage? | 13:40 |
|
comawhite
| did I do something wrong? | 13:40 |
|
Ilari
| comawhite: You committed in order to conclude the merge? | 13:40 |
|
comawhite
| eh? | 13:40 |
|
Ilari
| comawhite: If you didn't, there's your mistake. | 13:41 |
|
cbreak
| git add just adds things to the index | 13:41 |
|
comawhite
| okay I'm confused now (n00b) | 13:42 |
| ← mtkd left | 13:42 |
|
Ilari
| comawhite: Presumably the state of repo is wonky, so 'git reset --hard' first (this is why you shouldn't merge with uncommitted changes). | 13:42 |
|
comawhite
| in integration or master? | 13:43 |
|
cbreak
| first I'd get to the correct branch | 13:43 |
|
Ilari
| comawhite: Doesn't matter much. | 13:43 |
|
| comawhite: Then check out integration, and fetch + merge again. | 13:43 |
|
comawhite
| HEAD is now at fd3bb0d Changed tooltip to table (which was my last commit) | 13:44 |
| erlehmann → self | 13:44 |
|
Ilari
| comawhite: 'git checkout integration', 'git pull <remote> <branch>'. | 13:44 |
|
| comawhite: That should give some conflicts (since it did the last time). | 13:44 |
| → hobodave joined | 13:45 |
|
comawhite
| okay | 13:45 |
|
Ilari
| comawhite: Then resolve the conflicts, do 'git add' on conflicted files and finally 'git commit'. | 13:45 |
|
comawhite
| done | 13:45 |
|
Ilari
| comawhite: (no arguments to git commit). | 13:46 |
| → loincloth joined | 13:47 |
|
zorzar
| hey, is there a way to checkout severel tags or branches into subdirecotries of a working-copy? | 13:47 |
| self → erlehmann | 13:48 |
|
Ilari
| zorzar: No sane way to do that. | 13:48 |
| → mtkd joined | 13:48 |
|
drizzd
| doesn't git-new-workdir do something like that? | 13:49 |
|
| albeit not as a subdirectory | 13:49 |
| → iTron4 joined | 13:49 |
| ← khmarbaise left | 13:52 |
|
comawhite
| Ilari, hehe, I did the changes and commit but got error: Error building trees | 13:52 |
|
_Krieger_1
| and what is recommended practice for setting up local repositories between dev group and upstream code storage? what is wrong in scheme i've set up? | 13:52 |
|
cbreak
| just make it bare | 13:52 |
|
| make everything you want to push to bare | 13:52 |
|
_Krieger_1
| how to do that? | 13:53 |
|
Ilari
| comawhite: Did you add all the conflicted files? | 13:53 |
|
comawhite
| yes | 13:53 |
|
cbreak
| with --bare when you clone | 13:53 |
|
Ilari
| comawhite: 'git status' shouldn't list any anymore. | 13:53 |
|
_Krieger_1
| can i do that on already existing repo? | 13:53 |
| ← orafu left | 13:53 |
|
cbreak
| just clone again | 13:53 |
| → orafu joined | 13:53 |
|
_Krieger_1
| but my local_repo already stores info about my own changes to sources | 13:54 |
|
comawhite
| Ilari, hehe forgot the . at the end of git add (heres output of commit) http://pastebin.ca/1725937 | 13:54 |
|
Ilari
| _Krieger_1: Mv the .git out of it and nuke the rest. | 13:54 |
|
cbreak
| _Krieger_1: can't you delete it, clone it, and just push again? | 13:54 |
|
Ilari
| _Krieger_1: 'mv foo/.git foo.git' or so. | 13:54 |
|
cbreak
| if you do what Ilari said, you might have to change the config file too | 13:55 |
|
Ilari
| comawhite: 'Ctrl-X' to exit the editor? | 13:55 |
|
cbreak
| there is a flag for bareness, but I am not sure how relevant that is | 13:55 |
|
Ilari
| Well, that's easy to change. | 13:55 |
|
comawhite
| :P | 13:56 |
|
| okay now I do the merge yes? | 13:56 |
| → remi joined | 13:56 |
|
Ilari
| comawhite: Then after exiting it, 'git checkout master', 'git merge integration'. That shouldn't produce conflicts. | 13:56 |
|
comawhite
| thank you Ilari | 13:57 |
|
| :D | 13:57 |
|
| 19 files changed, 1906 insertions(+), 696 deletions(-) | 13:57 |
|
_Krieger_1
| bare = false | 13:57 |
|
| gonna to hack that... | 13:57 |
|
| thank you gyus for support | 13:57 |
| ← Paraselene_ left | 13:59 |
| ← til left | 13:59 |
| → mbernstein joined | 14:00 |
| → Paraselene_ joined | 14:01 |
| ← hummingkid left | 14:01 |
|
_Krieger_1
| git-pull cannot be used without a working tree :( | 14:02 |
|
| and how should i get changes from upstream? git fetch? as i get it, it'll receive recent commits from upstream and set HEAD to last commit. and what about my changes pushed to this local repo? how should i preserve them? | 14:02 |
|
Ilari
| _Krieger_1: Yeah, fetch. | 14:02 |
|
cbreak
| _Krieger_1: pull means merge | 14:02 |
|
| without working tree, there's nothing to merge :) | 14:02 |
| → vbgunz__ joined | 14:03 |
|
cbreak
| I think you could get problems with resolving conflicts otherwise | 14:03 |
| → jelmer_ joined | 14:03 |
|
_Krieger_1
| cbreak, i'm noob and i worry how to get consistent local repo with both upstream contributions and my | 14:03 |
|
| i am tuned to upstream branch 'kamailio_3.0'; so clones of local_repo do so also | 14:04 |
| ← loinclot_ left | 14:04 |
|
cbreak
| why don't you push and pull directly to upstream? | 14:05 |
|
_Krieger_1
| i push to local repo, and where will my changes go? as i understand, to untitled local branch? | 14:05 |
|
cbreak
| it should end up in a branch with the same name | 14:05 |
|
_Krieger_1
| i simply don't have write access to upstream repo | 14:05 |
|
cbreak
| unless you change it | 14:05 |
| ← hipe left | 14:06 |
| ← becomingGuru left | 14:06 |
|
_Krieger_1
| so when i push to local repo and then in local_repo if i `git fetch` i'll end up with simply that same one branch? | 14:07 |
|
| but wait, there must be also `git merge` call after git fetch, as i have read somewhere | 14:07 |
|
| what must be argument to git merge in my case? | 14:08 |
|
cbreak
| I don't think you can merge in a bare repo | 14:08 |
|
| I've never tried | 14:08 |
|
_Krieger_1
| let's assume then i still have non-bare repo | 14:09 |
|
cbreak
| then you can't push to it | 14:09 |
|
_Krieger_1
| hm... feels you said sth very wrong | 14:10 |
| → rlb3 joined | 14:10 |
| ← mtkd left | 14:10 |
|
cbreak
| feel free to read the man page :) | 14:10 |
| → sumangurung joined | 14:10 |
|
_Krieger_1
| you say i can't push to non-bare local repo :) but i have done it several times. sorry, i just didn't know that i can't :) | 14:11 |
|
cbreak
| oh... it's not that it's not possible | 14:11 |
|
| it's just that the working directory will get out of sync | 14:11 |
|
| and that will cause you problems | 14:11 |
|
_Krieger_1
| git reset --hard helped me | 14:11 |
|
cbreak
| yeah | 14:11 |
|
| it'll fix the working directory | 14:11 |
|
| but it will also irrevocably nuke any local changes | 14:12 |
|
| that aren't commited yet | 14:12 |
|
| anyway, why do you even want to pull from a bare? | 14:12 |
|
| can't you just pull from your normal repository, the one you work with? | 14:12 |
|
_Krieger_1
| but, if you have a couple of minutes more for me, i want to ask for nearest usage problem that i foresee | 14:12 |
|
cbreak
| I have one working repository with working directory, and a few bares I push to | 14:13 |
|
| (+ some more with working directories on other computers/vms) | 14:13 |
|
_Krieger_1
| what i call local_repo, i need for storing my changes and to tell co-workers "to build a package you do `git clone /home/local_repo my_workcopy && cd my_workcopy && ./build.sh`" | 14:14 |
|
cbreak
| yeah, then you never need to pull from that | 14:15 |
|
| you just push to it | 14:15 |
|
| I have a few such repositories for syncing, I call them "web" | 14:15 |
|
_Krieger_1
| no, i have to pull in it, to check out last commits from upstream | 14:15 |
|
cbreak
| why? | 14:15 |
| → nitrate_ joined | 14:15 |
|
cbreak
| why don't you pull form your normal repository? | 14:15 |
|
_Krieger_1
| emm i don't understand you quite well | 14:16 |
|
cbreak
| you have a repository you work in, right? | 14:16 |
|
| the one where you write your code, make your changes, and so on | 14:16 |
| ← jelmer left | 14:17 |
|
_Krieger_1
| i do `cd /home/local_repo; git reset --hard; git pull` and thus i get synced with upstream development | 14:17 |
|
cbreak
| just pull from there | 14:17 |
|
_Krieger_1
| then i can go to workcopy indented for package building, `git pull` and build newer package | 14:17 |
|
cbreak
| why don't you pull form your normal repository? | 14:17 |
|
_Krieger_1
| from upstream community repo? | 14:18 |
|
cbreak
| yeah | 14:18 |
|
| from everywhere you want | 14:18 |
|
_Krieger_1
| because i have my own commits | 14:18 |
|
cbreak
| yeah, git pull is fetch and merge | 14:18 |
|
| or if you really want to pull from an other repository than you work with | 14:19 |
| ← bigmack83 left | 14:19 |
|
cbreak
| just make one to pull and an other to push into | 14:19 |
|
| the pull repo has a working directory | 14:19 |
|
_Krieger_1
| if you can answer, `git pull` == `git fetch && git merge`, right? but merge takes an argument, what arg must be used to emulate git pull? | 14:19 |
|
cbreak
| and the other you push to doesn't | 14:19 |
|
| _Krieger_1: pull also takes arguments | 14:19 |
|
_Krieger_1
| but i use pull without them | 14:20 |
|
cbreak
| default args | 14:20 |
|
| they get set up when you clone | 14:20 |
| → selckin joined | 14:20 |
|
cbreak
| you can use others if you want | 14:20 |
|
| by default it pulls from the remote "origin" and merges depending on the branch's config | 14:21 |
| → bigmack83 joined | 14:21 |
|
_Krieger_1
| ok, but merge requires mandatory args, what are they in case of invoking `git pull`? | 14:21 |
|
cbreak
| depends on the branch | 14:21 |
|
wereHamster
| git reads the config file to find out which branch to fetch and merge | 14:21 |
|
cbreak
| anyway | 14:21 |
|
_Krieger_1
| so it is sth like `git fetch origin && git merge branch_name_workcopy_is_switched_to`, right? | 14:22 |
|
cbreak
| is there a reason why you can't pull from your normal repository? | 14:22 |
|
| no | 14:22 |
|
_Krieger_1
| yes, the reason is that I HAVE MY OWN COMMITS | 14:22 |
|
cbreak
| so what? | 14:22 |
|
| that's not a reason | 14:22 |
| ← nitrate left | 14:23 |
| jelmer_ → jelmer | 14:23 |
| ← sumangurung left | 14:23 |
|
cbreak
| merge merges. | 14:23 |
|
wereHamster
| d'oh :) | 14:23 |
| ← selckin left | 14:24 |
|
cbreak
| if you want, you can make different local tracking branches for the remote | 14:24 |
|
| but if you really really want a separate repository to pull with, just make one. but don't push to it :) | 14:24 |
| → selckin joined | 14:24 |
|
_Krieger_1
| i wonder how do you assume i'll build a package with my changes. maybe i should clone upstream repo, and...? | 14:25 |
|
| apply my changes stored on filesystem in patch files? | 14:25 |
| ngw → ngw|away | 14:25 |
| ngw|away → ngw | 14:25 |
|
cbreak
| why? | 14:25 |
| ← solofight left | 14:25 |
|
cbreak
| sounds like a big waste of time | 14:25 |
|
_Krieger_1
| i just can't understand how exactly you advise me to do | 14:26 |
| ← vbgunz__ left | 14:26 |
|
kukks
| _Krieger_1: have a look at the --track option e.g. for git branch | 14:26 |
|
cbreak
| you have a remote thingie, right? | 14:26 |
|
| and a local repository you work with | 14:26 |
|
| you make some commits there | 14:26 |
|
| then you want to get the changes someone else made, so just pull from the remote thingie | 14:27 |
|
| if you just want the remote thingie, make a branch from it | 14:27 |
|
_Krieger_1
| no, i'd prefer making commits in workcopies cloned from local repo | 14:27 |
|
cbreak
| and you have only the remote changes without anything you did | 14:27 |
|
wereHamster
| why? | 14:27 |
| → project2501a joined | 14:27 |
|
cbreak
| well, it's your time :) | 14:27 |
|
_Krieger_1
| because it can be not only me making changes to push to local repo | 14:27 |
|
cbreak
| ... and? | 14:28 |
|
| you can pull from both if you want | 14:28 |
| ← vbgunz left | 14:28 |
| → khmarbaise joined | 14:28 |
| ← cesario_ left | 14:28 |
|
cbreak
| but anyway, just make a second repository to pull with. then you have three :) | 14:29 |
|
_Krieger_1
| guys, as i understand, git is distributes system, so such tree-like commits flows should be normal, i can't get why you shame me for making my local repo for both pulling and pushing | 14:29 |
| ← JasonWoof left | 14:29 |
|
wereHamster
| as long as this local repo you'll be pushing/pulling to is bare, I don't see any problem | 14:30 |
|
cbreak
| I wouldn't expect merging in a bare repository to make sense | 14:30 |
| → nitrate__ joined | 14:30 |
|
_Krieger_1
| then please teach me how should i merge upstream commits and my local development commits, in bare repo | 14:31 |
|
wereHamster
| have your local bare repo, clone it, make some commits in the clone, maybe pull from upstream to get the latest changes, and push back into your local bare repo | 14:31 |
|
cbreak
| as I said three or four times already | 14:31 |
|
wereHamster
| _Krieger_1: as explained several times, you can't merge in a bare repo | 14:31 |
|
cbreak
| pull from your normal repository | 14:31 |
|
| I am not sure why you don't get that | 14:32 |
|
| there should be something like a pastebin for sketches | 14:32 |
| ← capndiesel left | 14:33 |
|
_Krieger_1
| i think because i'm not used to git, i've used svn for long time, and Mercurial for a little, and it seemed to me that distributed versioning systems allow what i was talking about | 14:34 |
|
cbreak
| they do. | 14:35 |
|
| but you do it wrong :) | 14:35 |
| dwmw2_gone → dwmw2 | 14:36 |
|
_Krieger_1
| ok, anyway thanks for your time :) | 14:36 |
|
_rane
| is there something better than Tab key for inputting long paths for git commands like git checkout, add etc. | 14:36 |
|
| would be cool if I could just write the filename | 14:36 |
|
cbreak
| http://www.twiddla.com/174013 | 14:36 |
|
| I guess that interactive sketch explain stuff doesn't work | 14:38 |
| → amitprakash joined | 14:38 |
|
amitprakash
| how can i export a git repo to svn? | 14:38 |
|
pantsman
| wrong way | 14:38 |
|
| surely :P | 14:38 |
|
amitprakash
| pantsman, heh.. i am using git to save my work .. but the stupid office uses svn | 14:39 |
|
| pantsman, trying to find a way to push my changes across | 14:39 |
|
bremner
| amitprakash: can you use git-svn? | 14:40 |
|
amitprakash
| bremner, i'll look into it, thanks a lot | 14:40 |
| → telemachus joined | 14:40 |
| ← nitrate_ left | 14:40 |
|
cbreak
| I think it is possible that a repository can be a git-repo and an svn-repo at the same time | 14:40 |
| → giallu joined | 14:41 |
| ← multani left | 14:42 |
| ← telemachus left | 14:45 |
| → altrux joined | 14:45 |
|
wereHamster
| cbreak: I'd pull directly into the local work repo when needed, and not use the 'pull repo' at all | 14:46 |
|
cbreak
| yes, me too | 14:46 |
|
| but for some reason, he doesn't want to be cause... OMGLOCALCHANGES! | 14:46 |
| → paltman joined | 14:46 |
|
cbreak
| maybe he wants to have both his local modifications and a pure remote version in working directories | 14:47 |
| → paltman_ joined | 14:47 |
| ← x|ixti|x left | 14:48 |
| ← bigmack83 left | 14:49 |
| ← iTron4 left | 14:50 |
| ← munichlinux left | 14:56 |
| → hummingkid joined | 14:57 |
| → x|ixti|x joined | 14:59 |
|
cbreak
| http://www.dabbleboard.com/draw?b=Guest244239&i=0&c=474e2a593ba76dc6ddfdd890d0a6c9090b527240 seems like a better tool. | 15:00 |
|
| Auto shape recognition! | 15:00 |
| ← DrNick left | 15:00 |
|
_rane
| but it's flash! | 15:00 |
|
cbreak
| hmm. | 15:02 |
|
| it has no text either | 15:02 |
| ← sh1mmer left | 15:02 |
| ← paltman_ left | 15:02 |
| ← Lisimba left | 15:03 |
| → jelmer_ joined | 15:03 |
| ← paltman left | 15:04 |
| ← amitprakash left | 15:05 |
| → amitprakash joined | 15:05 |
| ngw → ngw|away | 15:06 |
| ← warthog9 left | 15:07 |
| ← flaguy48 left | 15:10 |
|
bremner
| it seems to understand a bit of LaTeX | 15:10 |
| → warthog9 joined | 15:11 |
| ngw|away → ngw | 15:11 |
| ← khmarbaise left | 15:13 |
| ← sgh left | 15:17 |
| ← hummingkid left | 15:17 |
| ngw → ngw|away | 15:17 |
| → tychoish joined | 15:17 |
| ← EricInBNE_ left | 15:18 |
| ← kmartiin left | 15:19 |
| → paltman joined | 15:19 |
| → nessundorma joined | 15:21 |
| → voker57 joined | 15:22 |
| → beni- joined | 15:23 |
| → lucapost joined | 15:24 |
| → reprore joined | 15:26 |
| ← jelmer_ left | 15:27 |
| → dreiss_ joined | 15:27 |
|
lucapost
| hi, I have a problem with #git checkout -f, after push on remote repository, in remote repo I have excutable .git/hooks/post-receive with #git checkout -f, can anybody help me? | 15:28 |
| → hipe joined | 15:28 |
| ← BlindOracle left | 15:29 |
| → Xeross joined | 15:30 |
| → schacon joined | 15:30 |
|
Xeross
| Somehow git modified my file permissions if my file has a . at the start, how can I prevent this ? | 15:30 |
|
| for example it makes my .htaccess files hidden | 15:30 |
| ← beni- left | 15:31 |
| → beni- joined | 15:32 |
| ← beni- left | 15:32 |
|
Pieter
| hidden is not a permission in git | 15:32 |
|
bremner
| or in linux | 15:32 |
| → sumangurung joined | 15:33 |
| ← Adlai left | 15:33 |
|
Pieter
| it is a convention for programs not to display files starting with a ., but that is entirely git-unrelated | 15:33 |
|
Xeross
| I'm talking windows | 15:34 |
|
| somehow my .htaccess file gets assigned hidden | 15:34 |
|
| causing my PHP IDE to not save the file | 15:34 |
|
| saying permission denied | 15:34 |
| ← sumangurung left | 15:34 |
| → sumangurung joined | 15:34 |
|
cbreak
| must be some windows related problem | 15:35 |
|
Xeross
| hmm why would MSYSGIT assign it hidden :/ | 15:36 |
|
cbreak
| just make a symlink to it | 15:36 |
|
| then you can see it :) | 15:36 |
| → krawek joined | 15:36 |
|
Xeross
| I can see it | 15:36 |
|
cbreak
| then it's not very well hidden :) | 15:36 |
|
Xeross
| my ide shouldnt complain about not being able to write hidden files either :/ | 15:36 |
|
charon
| Xeross: probably to fool windows into treating the file as hidden, like most *nix tools would | 15:36 |
|
Xeross
| It has the hidden permission in windows | 15:36 |
|
| I want that disabled | 15:36 |
|
cbreak
| does windows have a SetFile command like mac os? | 15:37 |
| → luckyone joined | 15:37 |
|
luckyone
| hello all | 15:37 |
| → patrikf joined | 15:37 |
|
luckyone
| I have a quick question about git-daemon | 15:38 |
| → brizly joined | 15:39 |
|
wereHamster
| luckyone: this is irc, just ask | 15:39 |
|
Pieter
| 4863902 | 15:39 |
|
luckyone
| I am running ubuntu and when I try to run that, it can't find the command | 15:39 |
| ← paltman left | 15:39 |
|
thiago_home
| well, did you install git-daemon? | 15:40 |
|
wereHamster
| it = what, and can't find which command | 15:40 |
|
Pieter
| eh | 15:40 |
|
| Xeross: http://repo.or.cz/w/git/mingw/4msysgit.git/commit/f95b4f00a6b3bfb64dae4a9d27bce4b8998562d1 | 15:40 |
|
cbreak
| shouldn't it be "git daemon"? | 15:40 |
|
luckyone
| thiago_home: there doesn't seem to be a package for it | 15:40 |
|
Pieter
| Xeross: Naturally, we make this behavior optional on core.hideDotFiles. | 15:40 |
|
luckyone
| thiago_home: I would figure it comes in git-core? | 15:40 |
|
thiago_home
| luckyone: where is the binary? | 15:40 |
| → khmarbaise joined | 15:41 |
|
cbreak
| git daemon works here | 15:43 |
|
| and I only installed the basic git.git | 15:43 |
|
luckyone
| yeah - I am getting this this error - "sudo: git-daemon: command not found" | 15:43 |
|
cbreak
| binary is in /usr/local/bin/git | 15:43 |
|
| pass the absolute path | 15:43 |
|
_rane
| OT: what's the opposite of trailing in programming context? | 15:45 |
|
wereHamster
| luckyone: what about 'git daemon' (without the dash) | 15:45 |
|
charon
| _rane: leading? | 15:45 |
|
wereHamster
| _rane: leading.. ? | 15:45 |
|
_rane
| right :P | 15:45 |
|
| thanks | 15:45 |
|
cbreak
| trailing = following I thought | 15:45 |
|
context
| you said my name again | 15:46 |
|
cbreak
| the executable is in /usr/local/libexec/git-core/git-daemon, but I am not sure if you're supposed to call it directly | 15:47 |
|
luckyone
| wereHamster: I tried running "sudo -u git git-daemon --base-path=/home/git/repositories/ --export-all" | 15:48 |
|
cbreak
| lucapost: | 15:49 |
|
| see above :) | 15:49 |
|
luckyone
| when I run it without the dash, just git daemon, no error comes back, but I also don't see a git process running and I can't clone my repo publicly | 15:49 |
| Tommy[D]_ → Tommy[D] | 15:52 |
|
luckyone
| I just used the full path to the binary, and that worked - but I still can't seem to clone my repo | 15:52 |
|
wereHamster
| luckyone: what happens when you run 'git daemon' ? | 15:52 |
|
Ilari
| luckyone: Does 'netstat -an' show process listening? | 15:52 |
|
cbreak
| does it fulfill the conditions? | 15:53 |
|
| like the special file it has to contain? | 15:53 |
| ← remi left | 15:53 |
|
wereHamster
| cbreak: he used --export-all | 15:53 |
|
cbreak
| hmm. right. | 15:54 |
|
| then maybe it's a permission problem with the git user. or a path problem? | 15:54 |
|
| or maybe a com problem | 15:54 |
|
| or some other problem | 15:54 |
|
luckyone
| I have lots of problems :S | 15:55 |
|
cbreak
| type /nick unluckyone | 15:55 |
|
luckyone
| no kidding! | 15:55 |
| ← rchavik left | 15:55 |
| → vbgunz__ joined | 15:55 |
|
Ilari
| luckyone: Try to become git user and run: 'git daemon --base-path=/home/git/repositories/ --export-all --verbose'. Then leave it running and attempt to clone. | 15:56 |
|
| luckyone: The git daemon should print some verbose output. | 15:57 |
|
luckyone
| so - when I run ps -ef | grep git, I see this running : /usr/lib/git-core/git-daemon --verbose --base-path=/var/cache /var/cache/git | 15:57 |
|
| it looks like it is up, but pointed to the wrong base path | 15:57 |
|
cbreak
| maybe a problem with sudo | 15:57 |
|
Ilari
| luckyone: That's invoked from init scripts? | 15:58 |
| → Simoyd joined | 15:59 |
|
cbreak
| try adding an -- in your sudo call between the command and the rest | 16:00 |
|
Simoyd
| what's an easy way to check the state of a repository? How do I tell if I'm in the middle of a merge or rebase? | 16:00 |
|
_rane
| git status | 16:00 |
| → cuco joined | 16:00 |
|
Simoyd
| why does git status do everything and why do I keep forgetting about it? LOL | 16:00 |
|
luckyone
| I switched users to my git user, ran the command and got this: and fatal: unable to allocate any listen sockets on host (null) port 9418 | 16:01 |
|
cbreak
| now that's strange | 16:01 |
|
Ilari
| luckyone: Git daemon is already running. | 16:01 |
|
luckyone
| Ilari: I thougth I killed it | 16:01 |
|
cbreak
| or you have an execute protected environment | 16:01 |
|
| that blocks you from allocating sockets | 16:01 |
| ← amitprakash left | 16:01 |
|
Ilari
| luckyone: Maybe initscripts respawned that git daemon after you killed it? | 16:02 |
|
cbreak
| like mac os x' firewall, or the sandbox-exec jails | 16:02 |
|
luckyone
| Ilari: you bet they did... | 16:02 |
|
| blerg | 16:02 |
|
cbreak
| do you have lsof -i to show which sockets are allocated by which processes? | 16:03 |
|
Ilari
| luckyone: '/etc/init.d/git-daemon stop' (or whatever the name for the service was)? | 16:03 |
|
| luckyone: Or it might use some weird framework... | 16:03 |
|
| luckyone: Or uninstall 'git-daemon-run' package? | 16:04 |
|
luckyone
| I just uninstalled that | 16:06 |
| → vbgunz joined | 16:06 |
|
Ilari
| luckyone: If you still have git-daemon running, try to kill it and see if it comes back. | 16:07 |
|
luckyone
| that worked | 16:08 |
|
| it is up now | 16:08 |
| ← lucapost left | 16:08 |
|
luckyone
| so, now what is the best way for me to make sure this starts correctly in the event of a reboot | 16:08 |
|
| so I don't have manually login, switch to the git user, and start it? | 16:09 |
|
cbreak
| can't you modify the script that started it before? | 16:09 |
|
| seems to have kept it alive ok :) | 16:09 |
|
Ilari
| luckyone: Or put some simple script into /etc/init.d and link it from /etc/rcX.d (or something like that). | 16:11 |
| → jfxberns joined | 16:11 |
|
Ilari
| luckyone: Or even put it into /etc/inittab... | 16:11 |
|
cbreak
| if you'd be on OS X, you'd write a launchd script | 16:11 |
|
| that way, it git would only be started when someone actually requests the service | 16:12 |
|
| and wouldn't run all the time | 16:12 |
|
luckyone
| cbreak: I had installed git-daemon-run before, but it did not work | 16:12 |
| → catalis joined | 16:13 |
| ← catalis left | 16:13 |
| ← WinterWeaver left | 16:13 |
| → catalis joined | 16:13 |
| ← schacon left | 16:14 |
|
luckyone
| cbreak: if I was on OSX, I would have a toliet made of gold too | 16:15 |
| ← shuffle2 left | 16:15 |
| → nessundorma_ joined | 16:15 |
|
cbreak
| :O | 16:15 |
|
| I don't! | 16:15 |
| → ness1d0rma joined | 16:15 |
|
luckyone
| well, maybe that is just my list... rolex, toilet made of gold, mac fanboy | 16:16 |
|
| :P | 16:16 |
|
| I keed, I keed | 16:16 |
| → vbgunz_ joined | 16:17 |
| → datechman joined | 16:17 |
|
_rane
| how do I symlink to a git repo? | 16:18 |
| ← vbgunz_ left | 16:18 |
| ← sumangurung left | 16:18 |
|
_rane
| is it even possible? | 16:18 |
|
| or do I have to symlink from git repo to external location and not the other way around | 16:19 |
|
cbreak
| ln -s? | 16:19 |
| → vbgunz_ joined | 16:19 |
|
_rane
| I did ln -s ~/.profile ~/git/repo/bash_profile and it didn't work as expected | 16:20 |
|
cbreak
| how? | 16:20 |
|
_rane
| I ended up having string '/Users/rane/.profile/' in origin's bash_profile | 16:20 |
|
| I'll try doing it the other way | 16:21 |
|
cbreak
| just make a symlink that points into the repository | 16:21 |
| → shuffle2 joined | 16:22 |
| → madewokherd joined | 16:22 |
|
_rane
| I did, and it didn't work | 16:22 |
|
cbreak
| how did it not work? | 16:22 |
|
| it's like a symlink to a normal file | 16:22 |
|
_rane
| didn't I just explain .............. | 16:22 |
|
cbreak
| because it is a normal file | 16:22 |
| ← krawek left | 16:23 |
|
cbreak
| ln -s whateveryourgitrepois/somefile.txt whereveryouwantthesymlink/somefile.txt | 16:23 |
|
| then you have a symlink where ever you want it and it points wherever your repository is | 16:23 |
|
_rane
| that's the "other way around" I was talking about | 16:23 |
|
cbreak
| no | 16:23 |
|
_rane
| it is | 16:23 |
|
cbreak
| that's what I said | 16:24 |
|
_rane
| < _rane> | or do I have to symlink from git repo to external location and not the other way around | 16:24 |
|
cbreak
| make a symlink that points into the repo | 16:24 |
|
| do NOT make one into the repo that points to the external location | 16:24 |
|
| unless you want git to track the symlink | 16:24 |
|
| and not the file contents | 16:24 |
| tg` → tg | 16:24 |
| ← LiamH left | 16:28 |
| → kmartiin joined | 16:29 |
| → idletask joined | 16:29 |
| ← nessundorma left | 16:30 |
| ← vbgunz__ left | 16:31 |
| ← nessundorma_ left | 16:31 |
| → tedoc2000 joined | 16:32 |
| → LiamH joined | 16:35 |
| → vbgunz__ joined | 16:36 |
| → Pupeno joined | 16:37 |
| ← kmartiin left | 16:37 |
| → fynn joined | 16:38 |
|
fynn
| guys, 1.6.6 has been released. | 16:39 |
| ← whaley left | 16:40 |
| → kmap joined | 16:42 |
| ← idle_task left | 16:44 |
| → sumangurung joined | 16:44 |
| → Adlai joined | 16:45 |
| ← vbgunz left | 16:48 |
| ← vbgunz_ left | 16:51 |
| → cloudhead joined | 16:51 |
| ness1d0rma → nessundorma | 16:55 |
| ← tedoc2000 left | 16:56 |
| → mtkd joined | 16:56 |
| → offby1 joined | 16:57 |
| ← przemoc left | 16:59 |
| ← altrux left | 17:00 |
| ← lu_zero left | 17:01 |
| ← idletask left | 17:02 |
| ← catalis left | 17:03 |
| ← cuco left | 17:05 |
| → amerine joined | 17:08 |
| ← amerine left | 17:13 |
| → babilen_ joined | 17:13 |
| ← babilen_ left | 17:14 |
| → babilen_ joined | 17:14 |
| → mavera joined | 17:17 |
| ← LiamH left | 17:18 |
| ← mavera left | 17:18 |
| → jelmer_ joined | 17:20 |
| ← jelmer_ left | 17:20 |
| ← vbgunz__ left | 17:22 |
| ← fynn left | 17:23 |
| → hummingkid joined | 17:24 |
| → mithro joined | 17:24 |
| → paltman joined | 17:25 |
| ← intripoon left | 17:25 |
| ← _Krieger_1 left | 17:25 |
| → Lisimba joined | 17:26 |
| → beni- joined | 17:27 |
| ← docl left | 17:27 |
| ← Alfonzo left | 17:27 |
| ← kpreid_ left | 17:28 |
| → InTheWings joined | 17:32 |
|
InTheWings
| my git ends on a "fatal: read error: Connection reset by peer" when fetching refs | 17:33 |
|
| a clone works from the same server | 17:33 |
|
thiago_home
| fix the server | 17:33 |
| ← nessundorma left | 17:33 |
|
thiago_home
| is that ssh or git? | 17:34 |
|
InTheWings
| git:// | 17:34 |
| ← mbroeker left | 17:34 |
|
thiago_home
| sounds like a server bug | 17:35 |
|
InTheWings
| Ends up after exchanging ids http://pastebin.com/m650753d3 | 17:35 |
|
thiago_home
| do you have access to it? Can you check if it logged a crash? | 17:35 |
|
InTheWings
| no access | 17:35 |
|
| I'll ask | 17:35 |
| ← erlehmann left | 17:36 |
| → Alagar joined | 17:38 |
| → nsussman joined | 17:40 |
| ← beni- left | 17:40 |
| → beni- joined | 17:41 |
| ← MetaCosm left | 17:41 |
| ← hummingkid left | 17:43 |
| → jsallis joined | 17:44 |
| ← jsallis left | 17:45 |
| → jsallis joined | 17:45 |
| ← jsallis left | 17:46 |
| → tfarina joined | 17:46 |
| ← mithro left | 17:48 |
| → hyperair joined | 17:49 |
| → __skyl__ joined | 17:56 |
| ← Simoyd left | 18:01 |
| ← ia left | 18:03 |
| → przemoc joined | 18:05 |
| ngw|away → ngw | 18:05 |
| ← icwiener left | 18:06 |
| ← hipe left | 18:08 |
| → hipe joined | 18:09 |
| → icwiener joined | 18:09 |
| → docl joined | 18:10 |
| → ia joined | 18:11 |
| → sh1mmer joined | 18:11 |
| ← pantsman left | 18:11 |
| ← rwohleb left | 18:12 |
| → rwohleb joined | 18:12 |
| → lightcap joined | 18:14 |
| ← paltman left | 18:17 |
| ← jelmer left | 18:18 |
| → jelmer joined | 18:19 |
| ← icwiener left | 18:19 |
| ngw → ngw|away | 18:19 |
| → icwiener joined | 18:20 |
| → przemoc86 joined | 18:20 |
| ← rlb3 left | 18:21 |
| → wolog joined | 18:22 |
| ← sh1mmer left | 18:23 |
| ← lightcap left | 18:23 |
| ← przemoc86 left | 18:25 |
| → sh1mmer joined | 18:26 |
| → rootbit joined | 18:26 |
|
rootbit
| Hey there | 18:26 |
|
| i need some help using git | 18:26 |
| → AkumaStreak joined | 18:26 |
|
rootbit
| can someone help me? | 18:26 |
| ngw|away → ngw | 18:26 |
|
selckin
| !ask | 18:26 |
|
rootbit
| well | 18:27 |
|
| whats the workflow i should use with git? | 18:27 |
|
| I wanna create a patch for cyanogenmod | 18:27 |
|
| but i think i didnt understand correctly how do submit the patch... | 18:27 |
|
| i do clone | 18:28 |
|
| now i have the repository | 18:28 |
|
| and now what? | 18:28 |
|
selckin
| you replace return with , | 18:30 |
|
Ilari
| rootbit: First, Make some logical chnage, commit it (if its larger change make multiple commits)... | 18:31 |
|
rootbit
| OK | 18:32 |
|
| and then, push? | 18:32 |
|
| i dont have to branch or anything? | 18:32 |
| ← Xeross left | 18:32 |
| → froschi joined | 18:32 |
| → cloudhea joined | 18:32 |
| ← babilen left | 18:32 |
| babilen_ → babilen | 18:32 |
| → babilen_ joined | 18:33 |
|
Ilari
| rootbit: If you have permissions to. If not, you can either: 1) Put the changes in some public repo and send pull request, or 2) export the changes as patches and email them. | 18:34 |
| → przemoc86 joined | 18:34 |
| ngw → ngw|away | 18:34 |
| ← doener left | 18:36 |
| → doener joined | 18:36 |
|
rootbit
| OK | 18:36 |
|
| so I'm using github | 18:37 |
|
| and doing a pull request | 18:37 |
|
| now, how do i get the latest files from the repo? | 18:37 |
|
| i wanna update all my files | 18:37 |
|
| something like svn checkout | 18:37 |
|
Griffon26
| I don't know github, but svn checkout would be like git clone | 18:38 |
| → coolcat joined | 18:38 |
|
rootbit
| OK | 18:39 |
| ← cedricv left | 18:39 |
| ← przemoc86 left | 18:39 |
|
rootbit
| how about git pull, what does that do? | 18:39 |
|
Ilari
| rootbit: If you have existing repository... | 18:40 |
| → froschi1 joined | 18:40 |
|
Ilari
| rootbit: pull is fetch + merge. | 18:40 |
| → Xeross joined | 18:40 |
|
_rane
| Ilari: maybe take a break? :) eaten any christmas ham yet? | 18:40 |
|
Ilari
| _rane: Yes, I have. :-) | 18:41 |
|
_rane
| great, merry christmas then! | 18:41 |
| → hummingkid joined | 18:42 |
| ← przemoc left | 18:42 |
| → snearch joined | 18:43 |
| → przemoc joined | 18:45 |
| ← cloudhead left | 18:46 |
| → spearce joined | 18:49 |
| → calin joined | 18:50 |
| ← beni- left | 18:52 |
|
rootbit
| Ilari, Thanks! | 18:54 |
|
| fetch? whats that? | 18:54 |
| → segher joined | 18:54 |
|
offby1
| ^-. | 18:55 |
|
| that's supposed to be me raising one eyebrow | 18:55 |
| ← froschi left | 18:55 |
| ← sde_antrix1 left | 18:55 |
|
offby1
| rootbit: "fetch" does more or less what it sounds like: copies stuff from the remote repository into the local one. | 18:56 |
| → lu_zero joined | 18:57 |
| ← tango_ left | 18:58 |
| → erikeldridge joined | 18:58 |
|
rootbit
| ok, how about rebase? when do i use that one? | 18:59 |
| → LiamH joined | 19:00 |
|
rootbit
| one more thing | 19:00 |
|
| say i did i change and i dont like it | 19:01 |
|
| how do i go back? | 19:01 |
|
sh1mmer
| is there a way to cherry-pick from a remote? | 19:01 |
|
Ilari
| rootbit: Rebase tears down the current branch and rebuilds it (from patches saved before tearing it down). | 19:01 |
|
| sh1mmer: Fetch first. | 19:01 |
|
sh1mmer
| no tags | 19:01 |
|
| I can pull, but then I have to unstage a bunch of crap | 19:02 |
|
rootbit
| Ilari, ok, so say i made changes | 19:02 |
|
| and i dont like them and want to go back to the version in the repository | 19:02 |
|
| how do i do that? | 19:02 |
| ← __skyl__ left | 19:02 |
| ← ustunozgur left | 19:03 |
|
Ilari
| rootbit: To reset your changes away: 'git reset --merge origin/master' (reset to master of upstream). | 19:04 |
| ← amirite left | 19:04 |
|
rootbit
| yay | 19:04 |
|
| what does origin/master mean? | 19:04 |
|
Ilari
| rootbit: Refers to remote tracking branch: 'origin/master' meaning 'master' of 'origin'. | 19:05 |
| → chalkbag joined | 19:05 |
|
rootbit
| ok | 19:05 |
|
| and what is considered master of origin? | 19:05 |
| → przemoc86 joined | 19:05 |
|
rootbit
| what i did clone from? | 19:05 |
| ngw|away → ngw | 19:05 |
| ← bentob0x left | 19:06 |
|
Ilari
| rootbit: Origin is what you cloned from (unless you change that). 'master' is master branch. | 19:06 |
|
rootbit
| ok | 19:07 |
|
| every git repository has a master branch? | 19:07 |
|
Ilari
| rootbit: Most do, since it is the default for first branch, but there is no technical requirement. | 19:08 |
|
rootbit
| ok | 19:08 |
|
Ilari
| rootbit: To display names of all known local mirrors: 'git branch -r'. | 19:08 |
|
rootbit
| I'm talking about remote ones | 19:08 |
|
| and I want to merge into cyanogenmod | 19:09 |
|
| http://github.com/cyanogen/android_frameworks_base | 19:09 |
|
| here actually | 19:09 |
|
| so I'm gonna fork it | 19:09 |
|
| commit my change to that | 19:09 |
|
| and do a pull request | 19:09 |
|
| sounds right? | 19:09 |
|
Ilari
| rootbit: Merging happens between branches. And the branch to merge to is always the one that is checked out. | 19:09 |
|
rootbit
| sorry | 19:10 |
|
Ilari
| rootbit: Push your change to that (since you can't directly commit to it). | 19:10 |
|
rootbit
| want to commit to | 19:10 |
|
| yeah, push into the fork | 19:10 |
|
| and then, do a pull request | 19:10 |
|
| right? | 19:10 |
|
Ilari
| rootbit: Commit is local operation. But in order to push, you need to first commit the change that gets pushed. | 19:10 |
| ← dreiss_ left | 19:10 |
|
Ilari
| rootbit: Sounds about right. | 19:10 |
| ← kumbayo left | 19:11 |
|
rootbit
| OK, great | 19:11 |
|
| so ill try this now | 19:11 |
|
| see if it works | 19:11 |
| ← przemoc left | 19:11 |
|
Ilari
| rootbit: If the repo to push to isn't the same you cloned, you should register new remote for it (man git-remote , subcommand 'add'). | 19:12 |
|
Gitbot
| rootbit: the git-remote manpage can be found at http://git.or.cz/man/git-remote | 19:12 |
| ← icwiener left | 19:12 |
| → przemoc joined | 19:14 |
| ← przemoc86 left | 19:14 |
|
rootbit
| can i just change the value at git -config? | 19:15 |
|
Ilari
| rootbit: Yes. | 19:16 |
|
| rootbit: (the URL setting for it). | 19:16 |
| → danopia joined | 19:16 |
|
Ilari
| rootbit: Remote is 1) Shorthand name for another repository and 2) Assigns place to store mirrors of remote branches. | 19:17 |
| → rhett joined | 19:19 |
| ← przemoc left | 19:19 |
|
rootbit
| git push | 19:20 |
|
| fatal: No destination configured to push to. | 19:20 |
|
Ilari
| rootbit: 'git remote' says what? | 19:20 |
| → MetaCosm joined | 19:20 |
|
rootbit
| github | 19:21 |
| → Axius joined | 19:21 |
|
Ilari
| rootbit: 'git push github' then... | 19:21 |
|
rootbit
| tried that as well | 19:21 |
|
Ilari
| rootbit: 'git config remote.github.url'? | 19:22 |
|
rootbit
| http://pastebin.com/d1ce44afb | 19:22 |
| → FunkeeMonk joined | 19:22 |
|
rootbit
| as for your question | 19:23 |
|
| [email@hidden.address] | 19:23 |
|
Ilari
| rootbit: What's the branch on local side called? | 19:23 |
|
rootbit
| no idea | 19:23 |
|
| i didnt branch | 19:23 |
|
| how do i check? | 19:23 |
|
Ilari
| rootbit: 'git branch' will show that. | 19:23 |
|
rootbit
| * (no branch) | 19:24 |
| → przemoc joined | 19:24 |
|
Ilari
| rootbit: 'git checkout -b <name-for-branch>'. | 19:24 |
|
rootbit
| what would the name be? | 19:24 |
| ← przemoc left | 19:24 |
|
Ilari
| rootbit: Pick one. | 19:25 |
|
rootbit
| just something i come up with? | 19:25 |
|
| or the one in the repo | 19:25 |
|
| ? | 19:25 |
|
| i want to submit to http://github.com/omri/android_frameworks_base | 19:25 |
| → przemoc joined | 19:25 |
|
Ilari
| rootbit: It must be new name (or the checkout will complain it exists). | 19:25 |
|
rootbit
| OK | 19:25 |
|
| so omri_donut would work? | 19:25 |
| → loinclot_ joined | 19:25 |
|
Ilari
| rootbit: Maybe something short that describes what it's about? | 19:26 |
|
| rootbit: (the change that is). | 19:26 |
|
rootbit
| ok | 19:26 |
|
| bidi | 19:26 |
|
| its a fix for bidi support | 19:26 |
|
| so bidi could be it i guess | 19:26 |
|
Ilari
| rootbit: Yeah, fairly good one (git checkout -b bidi) | 19:26 |
|
rootbit
| so git checkout -b bidi | 19:26 |
|
Ilari
| rootbit: Now (no branch) should have disappeared from git branch and instead you have '* bidi'. | 19:27 |
|
rootbit
| ok | 19:27 |
| ← loincloth left | 19:27 |
|
rootbit
| yeah | 19:27 |
|
| so now what? | 19:27 |
|
Ilari
| rootbit: 'git push github bidi'. | 19:27 |
|
rootbit
| wouldent that create a new branch there as well? | 19:27 |
|
Ilari
| rootbit: Yes. | 19:27 |
|
| rootbit: (and that explicit what to push should even supress those warnings). | 19:27 |
|
| *suppress | 19:27 |
|
rootbit
| thats not what i want to do | 19:28 |
|
| i want to push it into donut | 19:28 |
|
| dont i? | 19:28 |
|
Ilari
| rootbit: 'git checkout donut', 'git merge bidi'? | 19:28 |
|
rootbit
| oh | 19:28 |
|
| i see | 19:28 |
|
| now? | 19:28 |
|
| or first commit it | 19:29 |
|
Ilari
| rootbit: Or does 'git branch' even show 'donut'? | 19:29 |
|
rootbit
| it dosnt | 19:29 |
|
Ilari
| rootbit: Commit first. | 19:29 |
|
rootbit
| its the branch on the server | 19:29 |
|
| remote branch | 19:29 |
|
| commited already | 19:29 |
|
| i ment push first.... | 19:29 |
|
Ilari
| rootbit: Maybe actually make proper checkout of donut first: 'git checkout -b donut origin/donut'. | 19:30 |
| → __skyl__ joined | 19:30 |
|
Ilari
| rootbit: Ugh. github/donut probably... | 19:30 |
|
| rootbit: I.e. mirror of 'donut' of 'github'. | 19:30 |
|
rootbit
| Branch donut set up to track remote branch donut from github. | 19:31 |
|
| Switched to a new branch 'donut' | 19:31 |
|
| now what? | 19:31 |
|
Ilari
| rootbit: 'git merge bidi'. | 19:31 |
|
rootbit
| great | 19:31 |
|
| worked | 19:31 |
|
| and now what? | 19:31 |
| → przemoc86 joined | 19:32 |
|
Ilari
| rootbit: 'git push github' (throws some warnings but should work). | 19:32 |
|
rootbit
| yay | 19:32 |
|
| worked | 19:32 |
|
| Ill copy paste this chat for me for future refs | 19:32 |
|
| because I'm sure ill forget something here | 19:33 |
| ← Axius left | 19:34 |
|
Ilari
| rootbit: The way to create local branch from remote tracking branch: 'git checkout -b <localbranchname> <remotetrackingbranch>'. | 19:34 |
|
rootbit
| ok | 19:34 |
|
| so i do this | 19:34 |
|
| then commit my change | 19:34 |
|
| and push at the end | 19:34 |
|
| right? | 19:34 |
|
Ilari
| rootbit: Any existing local branch can be switched to by 'git checkout <name>' (if you have uncommitted changes, that either fails or carries changes over). | 19:34 |
|
| rootbit: Pretty much. You can make multiple changes before pushing (and also inspect the series before pushing). | 19:35 |
|
rootbit
| great | 19:35 |
| → ablemike joined | 19:35 |
|
rootbit
| and when i want to update my repository back to the remote version | 19:35 |
|
| what do i do again? | 19:35 |
| → bugfux joined | 19:35 |
| → stamina_ joined | 19:35 |
| → lockd joined | 19:36 |
|
lockd
| how can I get a specific checkout hash? | 19:36 |
|
rootbit
| Thanks alot BTW! | 19:36 |
|
Ilari
| rootbit: 'git reset --merge <remotetrackingbranch>' (--hard also kills uncommitted changes). | 19:36 |
|
| rootbit: Update remote tracking branches: 'git remote update'. | 19:37 |
|
| rootbit: (it goes through all remotes). | 19:37 |
|
rootbit
| cool | 19:37 |
|
| :-D | 19:37 |
|
| rootbit learned | 19:37 |
|
| rootbit is waiting for his patch to be pulled | 19:37 |
|
rootbit
| android is amazing! | 19:38 |
|
Ilari
| rootbit: (its actually commit). :-) | 19:38 |
| → ustunozgur joined | 19:38 |
| ← przemoc86 left | 19:38 |
| ← przemoc left | 19:38 |
|
Ilari
| rootbit: And pull is fetch and merge. | 19:38 |
| → przemoc joined | 19:38 |
|
rootbit
| semantic.. | 19:38 |
|
| but yeah, it should get pulled | 19:39 |
|
Ilari
| (and patches don't get pulled, they get appiled). :-) | 19:39 |
| ← MetaCosm left | 19:40 |
| → sh1m joined | 19:40 |
| ← bugfux left | 19:40 |
|
rootbit
| OK | 19:41 |
|
| well then | 19:41 |
|
| bath time for me | 19:41 |
|
| thanks for all the help | 19:41 |
| → MetaCosm joined | 19:41 |
| ← przemoc left | 19:43 |
| → przemoc joined | 19:44 |
| ← ablemike left | 19:46 |
| ← sh1mmer left | 19:47 |
| sh1m → sh1mmer | 19:47 |
| VVSiz_ → VVSiz | 19:49 |
| ← magcius left | 19:49 |
| → magcius joined | 19:49 |
| ← __skyl__ left | 19:51 |
| ← stamina_ left | 19:52 |
| ← hyperair left | 19:53 |
| ← przemoc left | 19:55 |
| → przemoc joined | 19:55 |
| ← Sky[x] left | 19:58 |
| → Sky[x] joined | 19:58 |
| → Rolenun joined | 19:59 |
| ← reprore left | 20:00 |
| → foutrelis joined | 20:00 |
| ← przemoc left | 20:00 |
| ← Sky[x] left | 20:01 |
| ← hummingkid left | 20:01 |
|
foutrelis
| Hello. About a year ago I | 20:02 |
|
| -.- | 20:02 |
| → przemoc joined | 20:02 |
|
foutrelis
| Somehow I hit the Return key too early. .-. | 20:02 |
|
offby1
| Blame it on your cat; that's what I do. | 20:02 |
|
Rolenun
| hey, I was filling in all sorts of blanks....you're lucky I didnt finish the sentence for you :D | 20:02 |
|
foutrelis
| offby1: Rolenun :P | 20:03 |
|
| That's what I get for playing around with filter-branch | 20:03 |
|
Pupeno
| How do I force a non-fast-foward push? | 20:08 |
| → Sky[x] joined | 20:08 |
|
foutrelis
| Well, a year ago I thought it would be a good idea to track a directory full of images used on a website along with its code. Now the repo has reached around half a GiB in size, most of which is images (the code history is about 30 MiB in size). To get rid of the images' history, I'm trying this command: http://dpaste.com/137477/ - which alters the commits and removes the images from the index. AFAIK, this operation shouldn't touch the wo | 20:09 |
| → MetaCosm_ joined | 20:10 |
|
foutrelis
| Pupeno: hmmm, push has a --force (-f) option but it's a bit dangerous so consult the man page first (git help push). | 20:11 |
| → przemoc86 joined | 20:15 |
| ngw → ngw|away | 20:16 |
| ← MetaCosm left | 20:16 |
| ngw|away → ngw | 20:16 |
| ← magcius left | 20:17 |
| → magcius joined | 20:18 |
| → icwiener joined | 20:18 |
| ← przemoc86 left | 20:20 |
| ← unreal left | 20:20 |
| → eletuchy joined | 20:20 |
| ← babilen left | 20:20 |
| ← przemoc left | 20:22 |
| → babilen joined | 20:24 |
| → Chillance joined | 20:24 |
| → becomingGuru joined | 20:24 |
| ← babilen_ left | 20:25 |
| → przemoc joined | 20:25 |
| → Gitzilla joined | 20:25 |
| ← Gitzilla left | 20:25 |
| → defectiv joined | 20:26 |
| → loincloth joined | 20:29 |
|
offby1
| Pupeno: I think something like "git push remote +refs/heads/whatever:refs/heads/whatever" -- note the + | 20:29 |
| ← loinclot_ left | 20:30 |
| ← przemoc left | 20:30 |
| → przemoc joined | 20:31 |
| → WinterWeaver joined | 20:31 |
| → kmartiin joined | 20:31 |
|
Ilari
| Can't that be shortened as 'git push remote +whatever'? | 20:32 |
|
offby1
| probably :) | 20:33 |
|
| I can never remember the various defaults | 20:33 |
|
| foutrelis looks angrily at `filter-branch --index-filter' for removing files from the working copy :< | 20:34 |
| → unreal joined | 20:37 |
| ← przemoc left | 20:39 |
| → przemoc joined | 20:39 |
| ← FunkeeMonk left | 20:42 |
| ← WinterWeaver left | 20:44 |
| ← hipe left | 20:47 |
|
cbreak
| wasn't that the point of filter-branch? | 20:47 |
|
foutrelis
| cbreak: But I'm using the index filter :( | 20:48 |
|
| cbreak: specifically this command: http://dpaste.com/137477/ | 20:48 |
|
| I'm probably misunderstanding something :) | 20:49 |
| → |Inthewings| joined | 20:50 |
| ← defectiv left | 20:52 |
|
cbreak
| well, filter branch performs the command on every commit | 20:52 |
|
| it rewrites history | 20:52 |
|
| in your case, "path/to/directory" gets removed from the whole repository | 20:53 |
|
foutrelis
| Oh, so it's like that directory was never introduced and the working tree I'm left with represents that fact? | 20:54 |
| ← becomingGuru left | 20:54 |
|
cbreak
| I don't think it matters what your working directory is | 20:55 |
| ← przemoc left | 20:55 |
|
cbreak
| you can just check out what ever you want | 20:55 |
|
| switch branches or what ever | 20:55 |
| → sensui joined | 20:55 |
| → przemoc joined | 20:55 |
|
cbreak
| yes, it's as if the directory never existed | 20:56 |
|
| well, depending on which branches you filtered of course | 20:56 |
|
foutrelis
| I think I filtered all of them :) | 20:56 |
|
| I still would expect the directory to remain intact though (and just become untracked). | 20:56 |
|
cbreak
| I don't know what it's supposed to do with the working directory | 20:58 |
|
| I wouldn't expect anything | 20:59 |
| ← nsussman left | 20:59 |
|
cbreak
| well, if you want the directory back, that shouldn't be a problem | 20:59 |
|
foutrelis
| No, that's not an issue; I'm testing on clones of the repo. | 20:59 |
|
cbreak
| do something like git checkout HEAD@{1} to get an old version from the reflog and you have what ever you had before youdid what you did | 21:00 |
| ← tehbaut left | 21:00 |
|
cbreak
| (with a detached head) | 21:00 |
| ← przemoc left | 21:00 |
|
foutrelis
| Thanks for the input cbreak. I'm mostly asking to see if this is expected bahavior and whether I'm doing something the wrong way. | 21:01 |
| → kpreid_ joined | 21:05 |
| ← InTheWings left | 21:05 |
| ← chalkbag left | 21:06 |
| → tehbaut joined | 21:06 |
| → przemoc joined | 21:07 |
| ← project2501a left | 21:07 |
| ← sensui left | 21:11 |
| ← incredible left | 21:11 |
| → stewartl42 joined | 21:15 |
|
stewartl42
| question about users... not sure what to search for... I need to set up an internal repository that my team can commit to, but I also want it set up so that at least one other person on the team has to approve the commit. doable? | 21:16 |
| ← kmartiin left | 21:16 |
|
cbreak
| I think I read something about that recently | 21:16 |
|
| I would do it with two repositories | 21:16 |
|
| ... hmm... no, that's something else | 21:17 |
|
stewartl42
| or can I do it with another VCS tool? | 21:17 |
|
cbreak
| http://github.com/sitaramc/gitolite | 21:18 |
|
| with that you can limit access for certain branches | 21:18 |
|
| but I am not sure if it can enforce what you want | 21:18 |
| ← icwiener left | 21:19 |
|
drizzd
| gerrit does review stuff | 21:21 |
|
stewartl42
| we may have to just enforce this with internal process... I manage a team of 4 guys (plus me) who do server management and I want to keep configs in version control. | 21:21 |
|
| if the tool won't do it for us. | 21:22 |
| ← yath left | 21:27 |
| ← foutrelis left | 21:27 |
| ← sumangurung left | 21:28 |
| ← brizly left | 21:28 |
| ← stewartl42 left | 21:28 |
| → psoo joined | 21:28 |
| → foutrelis joined | 21:29 |
|
foutrelis
| Stupid netbook crashed (I hope I didn't miss anything after my last message). | 21:29 |
|
cbreak
| I didn't write anything more. I've never used filter-branch myself :/ | 21:31 |
|
foutrelis
| No worries. :) | 21:31 |
| ← psoo left | 21:32 |
| → psoo joined | 21:32 |
| ← docl left | 21:35 |
| ← __iron left | 21:36 |
| → kmartiin joined | 21:36 |
| → icwiener joined | 21:36 |
| → __iron joined | 21:36 |
| ← psoo left | 21:38 |
| ← tekky left | 21:38 |
| ← erikeldridge left | 21:39 |
|
charon
| foutrelis: g-f-b checks out the rewritten copy of HEAD, in the process of which it notices that your path/to/directory is no longer part of the current commit and thus should be deleted | 21:40 |
|
| foutrelis: i think you can avoid this by detaching HEAD first, but it'll still disappear the second you check out the newly rewritten branch | 21:41 |
|
foutrelis
| That makes sense. | 21:42 |
| ← bambule left | 21:43 |
| → khmarbaise_ joined | 21:43 |
|
foutrelis
| Hum. :) | 21:44 |
| ← kmartiin left | 21:44 |
| ← calin left | 21:48 |
| ← jfxberns left | 21:50 |
| ← GustavoR|aweh left | 21:53 |
| → GustavoRamos joined | 21:53 |
| → jjuran joined | 21:53 |
| → durbin joined | 21:55 |
|
durbin
| hi everyone, are there any good tutorials out their to create a git server using the git:// protocol? | 21:57 |
| ← khmarbaise left | 21:58 |
| khmarbaise_ → khmarbaise | 21:58 |
| ← bauruine left | 21:58 |
| ← patrikf left | 22:00 |
| → tedoc2000 joined | 22:00 |
| → pantsman joined | 22:00 |
| ← icwiener left | 22:01 |
| → r11t joined | 22:03 |
|
wereHamster
| durbin: <HEAD:Documentation/technical/protocol-common.txt> | 22:04 |
|
Gitbot
| [git protocol-common.txt]: http://tinyurl.com/yfttxjd [blob] | 22:04 |
|
wereHamster
| durbin: or take a look at the other files in that directory | 22:04 |
| → DrNick joined | 22:05 |
|
| spearce wonders why durbin wants to write his own git server, aren't there enough libraries already that implement the protocol? | 22:05 |
| → kencur joined | 22:08 |
|
foutrelis
| I think he wants to set up a server, not write one. | 22:10 |
| ← tedoc2000 left | 22:11 |
| ← khmarbaise left | 22:12 |
|
durbin
| thanks guys, I think I found what I was looking for with the git-daemon. happy holidays! | 22:15 |
|
foutrelis
| Happy holidays to you too, durbin ^-^ | 22:15 |
| ← kencur left | 22:17 |
| → whaley joined | 22:23 |
| → Twisted_Mentat joined | 22:26 |
| → godsflaw joined | 22:26 |
| → zackattack joined | 22:27 |
| → tekky joined | 22:29 |
| ← vital left | 22:31 |
| ← rudi_s left | 22:31 |
| → rudi_s joined | 22:32 |
| → vital joined | 22:35 |
| ← godsflaw left | 22:38 |
| ← thiago_home left | 22:38 |
| ← nitrate__ left | 22:39 |
| ← Sigma left | 22:42 |
| → vbgunz__ joined | 22:43 |
| ← snearch left | 22:45 |
| → godsflaw joined | 22:45 |
| → j416 joined | 22:46 |
| ← mtkd left | 22:47 |
| ← foutrelis left | 22:47 |
| → mtkd joined | 22:47 |
| → nanothief joined | 22:49 |
| → vbgunz joined | 22:52 |
| ← vbgunz__ left | 22:53 |
| ← MetaCosm_ left | 22:57 |
| → hummingkid joined | 23:06 |
| ← d0k left | 23:08 |
| → tedoc2000 joined | 23:09 |
| → Turias joined | 23:11 |
| ← Turias left | 23:11 |
| ← unomi1 left | 23:14 |
| → unomi1 joined | 23:14 |
| ← m104 left | 23:16 |
| ← __iron left | 23:16 |
| → __iron joined | 23:16 |
| ← imajes left | 23:16 |
| xmasfu → fujin | 23:20 |
| ← ph^ left | 23:21 |
| ← nanothief left | 23:22 |
| ← x|ixti|x left | 23:22 |
| → Dec1129 joined | 23:23 |
| ← Oejet left | 23:23 |
| ← hummingkid left | 23:23 |
| → kmartiin joined | 23:27 |
| ← Chillance left | 23:29 |
| ← shuffle2 left | 23:30 |
| ← pantsman left | 23:32 |
|
gitster
| can somebody update the /title here? | 23:33 |
| → pantsman joined | 23:33 |
| ← whaley left | 23:34 |
| ← godsflaw left | 23:34 |
| ChanServ set mode: +o | 23:35 |
|
gitster
| Thanks, and happy holidays without hacking ;-) | 23:36 |
| ← maxter left | 23:39 |
| ← giallu left | 23:39 |
| → HG` joined | 23:41 |
| ← pantsman left | 23:42 |
| ← HG` left | 23:43 |
| ← froschi1 left | 23:44 |
| ← kmartiin left | 23:46 |
| ← corni left | 23:48 |
| → ph^ joined | 23:50 |
| ← zackattack left | 23:51 |
| ← mtkd left | 23:54 |
| → sh1m joined | 23:55 |
| → michaels_ joined | 23:55 |
| ← sh1mmer left | 23:59 |
| sh1m → sh1mmer | 23:59 |