Search Results

Search found 2940 results on 118 pages for 'git'.

Page 35/118 | < Previous Page | 31 32 33 34 35 36 37 38 39 40 41 42 | Next Page >

how to use git rebase to clean up a convoluted history

- by lsiden

After working for several weeks with a half dozen different branches and merges, on both my laptop and work and my desktop at home, my history has gotten a bit convoluted. For example, I just did a fetch, then merged master with origin/master. Now, when I do git show-branches, the output looks like this: ! [login] Changed domain name. ! [master] Merge remote branch 'origin/master' ! [migrate-1.9] Migrating to 1.9.1 on Heroku ! [rebase-master] Merge remote branch 'origin/master' ---- - - [master] Merge remote branch 'origin/master' + + [master^2] A bit of re-arranging and cleanup. - - [master^2^] Merge branch 'rpx-login' + + [master^2^^2] Commented out some debug logging. + + [master^2^^2^] Monkey-patched Rack::Request#ip + + [master^2^^2~2] dump each request to log .... I would like to clean this up with a git rebase. I created a new branch, rebase-master, for this purpose, and on this branch tried git rebase <common-ancestor>. However, I have to resolve many conflicts, and the end result on branch rebase-master no longer matches the corresponding version on master, which has already been tested and works! I thought I saw a solution to this somewhere but can't find it anymore. Does anyone know how to do this? Or will these convoluted ref names go away when I start deleting un-needed branches that I have already merged with? I am the sole developer on this project, so there is no one else who will be affected.

Read the article
Module based web project directory layout with git and symlinks

- by karlthorwald

I am planning my directory structure for a linux/apache/php web project like this: Only www.example.com/webroot/ will be exposed in apache www.example.com/ webroot/ index.php module1/ module2/ modules/ module1/ module1.class.php module1.js module2/ module2.class.php module2.css lib/ lib1/ lib1.class.php the modules/ and lib/ directory will only be in the php path. To make the css and js files visible in the webroot directory I am planning to use symlinks. webroot/ index.php module1/ module1.js (symlinked) module2/ module2.css (symlinked) I tried following these principles: layout by modules and libraries, not by file type and not by "public' or 'non public', index.php is an exception. This is for easier development. symlinking files that need to be public for the modules and libs to a public location, but still mirroring the layout. So the module structure is also visible in the resulting html code in the links, which might help development. How will git handle the symlinking of the single files correctly, is there something to consider? When it comes to images I will need to link directories, how to handle that with git? modules/ module3/ module3.class.php img/ img1.jpg img2.jpg img3.jpg They should be linked here: webroot/ module3/ img/ (symlinked ?) So this is a git and symlink question. But I would be interested to hear comments about the php layout, maybe you want to use the comment function for this.

Read the article
Best support now on windows: Mercurial or Git?

- by mamcx

I want to change my current subversion setup to Mercurial or Git. I read about the two and I have a conflicted view about how well they work on windows. Alot of pages say Git is sub-par on windows, slow and badly integrated. And almost everyone say Mercurial is better. But some say Git now is better and Mercurial is behind. I check the screenshots of TortoiseHG and TortoiseGIT and the mercurial one look "worse"... but maybe is just crappy screenshots? I read about the two, prefer the command-line interface of Mercurial, but seriously, I don't pretend to touch the command line. And if one of the two is a real improvenment to SVN, I don't have to do that (In SVN is necesary go to the metal because something need fix). In SVN I have issues when commit or get code made on OSX (I code on Windows, OSX, Solaris. Mainly windows). So I hope don't get that issues again (I mean, failure to commit to the repo). I have a small repository, doing solo.

Read the article
Please help me with a Git workflow

- by aaron carlino

I'm an SVN user hoping to move to Git. I've been reading documentation and tutorials all day, and I still have unanswered questions. I don't know if this workflow will make sense, but here's my situation, and what I would like to get out of my workflow: Multiple developers, all developing locally on their work stations 3 versions of the website: Dev, Staging, Production Here's my dream: A developer works locally on his own branch, say "developer1", tests on his local machine, and commits his changes. Another developer can pull down those changes into his own branch. Merge developer1 - developer2. When the work is ready to be seen by the public, I'd like to be able to "push" to Dev, Staging, or Production. git push origin staging or maybe.. git merge developer1 staging I'm not sure. Like I said, I'm still new to it. Here are my main questions: -Do my websites (Dev, Staging, Production) have to be repositories? And do they have to be "bare" in order to be the recipients of new changes? -Do I want one repository or many, with several branches? -Does this even make sense, or am I on the wrong path? I've read a lot of tutorials, so I'm really hoping someone can just help me out with my specific situation. Thanks so much!

Read the article
Managing aesthetic code changes in git

- by Ollie Saunders

I find that I make a lot of small changes to my source code, often things that have almost no functional effect. For example: Refining or correcting comments. Moving function definitions within a class for a more natural reading order. Spacing and lining up some declarations for readability. Collapsing something using multiple lines on to one. Removing an old piece of commented-out code. Correcting some inconsistent whitespace. I guess I have a formidable attention to detail in my code. But the problem is I don't know what to do about these changes and they make it difficult to switch between branches etc. in git. I find myself not knowing whether to commit the minor changes, stash them, or put them in a separate branch of little tweaks and merge that in later. None those options seems ideal. The main problem is that these sort of changes are unpredictable. If I was to commit these there would be so many commits with the message "Minor code aesthetic change.", because, the second I make such a commit I notice another similar issue. What should I do when I make a minor change, a significant change, and then another minor change? I'd like to merge the three minor changes into one commit. It's also annoying seeing files as modified in git status when the change barely warrants my attention. I know about git commit --amend but I also know that's bad practice as it makes my repo inconsistent with remotes.

Read the article
Best workflow with Git & Github

- by Tom Schlick

Hey guys, im looking for some advice on how to properly structure the workflow for my team with git & github. we are recent svn converts and its kind of confusing on how we should best setup our day-to-day workflow. Here is a little background, im comfortable with command line and my team is pretty new to it but can follow use commands. We all are working on the same project with 3 environments (development, staging, and production). We are a mix of developers & designers so some use the Git GUI and some command line. Our setup in svn went something like this. We had a branch for development, staging and production. When people were confident with code they would commit and then merge it into the staging. The server would update itself and on a release day (weekly) we would do a diff and push the changes to the production server. Now i setup those branches and got the process with the server running but its the actual workflow that is confusing the hell out of me. It seems like overkill that every time someone makes a change on a file they would create a new branch, commit, merge, and delete that branch... from what i have read they would be able to do it on a specific commit (using the hash), do i have that right? is this an acceptable way to go about things with git? any advice would be greatly appreciated.

Read the article
how to seamlessly integrate subversion and git?

- by mattv

I'm looking for tips on how to seamlessly integrate subversion and git, for deploying web sites by a small team of web developers. We each have our own development versions of our sites on our local machines. We also have dev, staging, and live servers. As our team has grown, we haven't updated our revision control and deployment strategies accordingly. We had all been checking into the trunk of a shared Subversion repository. Both the dev & staging servers ran from a checkout of the trunk, so updating them involved running "svn update" while the live server ran as an export from trunk which required an "svn export" to get the latest code. In either case, we would often update just certain files by updating or exporting just those files or directories. That worked okay when there was just one or two developers. However, a big downside was that we couldn't point to an individual tag that represented what was currently on live at any given time. In keeping with corporate policy, we'd like to continue to use Subversion to store what we're now calling our "production branch," which will be what goes onto staging and live. However, we would like to use Git on our local and development sites. We especially like the idea of easier merges and being able to "cherry pick" updates that need to go live. We had initially planned on using git-svn, but it doesn't seem to work well in a shared environment such as our dev or staging servers. Anyone else doing something like this? What's the best way to make it work? Or are we making it more difficult than it should be?

Read the article
Git + SoA, one repo or many?

- by parsenome

Normally, when I start up a new application, I'd create a new git repository for it. That's well accepted and plays nice with Github when I want to share my code. At work, I'm working in a service oriented architecture. One very common pattern is to add some code to two different applications at the same time - perhaps adding a model with a RESTful interface to one and a web frontend for managing it on another. Using separate git repositories has some warts in this case. Here are what I see as the downsides of doing separate repositories: I have to commit twice I can't correllate related commits very well No single place to go back and trace history - I'd love to be able to bring up all my commits for the day in a single place Forgetting to pull one repo or another is a gotcha On the other hand, I've used perforce a lot and its one giant repository model has lots of warts too. Perforce has features designed to let help you with those, git doesn't. Has anyone else run into this situation? How did you handle it? What worked well, and what didn't?

Read the article
Setting up a git repository on a server

- by lostInTransit

Hi I had posted this question on superuser but didn't get a helpful response. Thought I'd try here since the question does deal with some configurations and settings for using github. I have a central server with SSO installed. All my machines are connected through the lan to this server. I have also setup a remote git repository on this server. Now what I'd like to do is make the server act as a central repository. All my employees can commit their code to the server and the server pushes it to the remote git repository. Can someone please help me out with this process? I am new to git and still learning how to use it effectively. So a step-by-step process or an existing document which I can refer to for this? Also can I integrate it with SSO in any way? The server itself is setup on a Mac and SSO uses Atlassian Crowd. Thanks.

Read the article
How to use git to manage one codebase but have different environments

- by emostar

I'm using git for a personal project at the moment and have run into a problem of having one codebase for two different environments and was wondering what the cleanest way to use git would be. Main Desktop I Use this machine for most of my development. I have a git repository here that I cloned off of an empty repository that I use on my internal server. I do most of my work here and push back to the internal server so I can use that as a master of truth and to ease making backups. Laptop I sometimes want to code on the road, so I did a clone from the internal server and created a new branch called "laptop-branch". Unfortunately some directories MSVC++ version are different than from the Main Desktop environment. I just modified the files in the "laptop-branch" and committed them there. Now I did a lot of changes while on vacation with my laptop, and want to push them to origin, but don't want the changes I made that were related to directories and compiler versions to be pushed back to origin. What would be the best way to get this done?

Read the article
Unable to initialize gitosis-init

- by aunghn

I was tried to setup git and gitosis for our projects. For the gitosis setup, I'm following this article http://scie.nti.st/2007/11/14/hosting-git-repositories-the-easy-and-secure-way as this is first time using git. I got an issue when I run the gitosis-init. I don't know what happen or how to check. Please help me on what I need to check or do. In fact, I don't even know whether this is a problem as I just started using Linux, git and etc. Reinitialized existing Git repository in /home/git/repositories/gitosis-admin.git/ Reinitialized existing Git repository in /home/git/repositories/gitosis-admin.git/ Traceback (most recent call last): File "/usr/bin/gitosis-init", line 8, in <module> load_entry_point('gitosis==0.2', 'console_scripts', 'gitosis-init')() File "/usr/lib/python2.5/site-packages/gitosis-0.2-py2.5.egg/gitosis/app.py", line 24, in run return app.main() File "/usr/lib/python2.5/site-packages/gitosis-0.2-py2.5.egg/gitosis/app.py", line 38, in main self.handle_args(parser, cfg, options, args) File "/usr/lib/python2.5/site-packages/gitosis-0.2-py2.5.egg/gitosis/init.py", line 140, in handle_args run_hook.post_update(cfg=cfg, git_dir=admin_repository) File "/usr/lib/python2.5/site-packages/gitosis-0.2-py2.5.egg/gitosis/run_hook.py", line 40, in post_update path=os.path.join(generated, 'projects.list'), File "/usr/lib/python2.5/site-packages/gitosis-0.2-py2.5.egg/gitosis/gitweb.py", line 109, in generate_project_list f = file(tmp, 'w') IOError: [Errno 13] Permission denied: '/home/git/gitosis/projects.list.30470.tmp' Thanks in advance.

Read the article
Remove file from history completely

- by Iain

A colleague has done a few things I told them not to do: forked the origin repo online cloned the fork, added a file that shouldn't have been added to that local repo pushed this to their fork I've then: merged the changes from the fork and found the file I want to remove this from: my local repo the fork their local repo I have a solution for removing something from the history, taken from Remove file from git repository (history). What I need to know is, should my colleague also go through this, and will a subsequent push remove all info from the fork? (I'd like an alternative to just destroying the fork, as I'm not sure my colleague will do this) SOLUTION: This is the shortest way to get rid of the files: check .git/packed-refs - my problem was that I had there a refs/remotes/origin/master line for a remote repository, delete it, otherwise git won't remove those files (optional) git verify-pack -v .git/objects/pack/#{pack-name}.idx | sort -k 3 -n | tail -5 - to check for the largest files (optional) git rev-list --objects --all | grep a0d770a97ff0fac0be1d777b32cc67fe69eb9a98 - to check what files those are git filter-branch --index-filter 'git rm --cached --ignore-unmatch file_names' - to remove the file from all revisions rm -rf .git/refs/original/ - to remove git's backup git reflog expire --all --expire='0 days' - to expire all the loose objects (optional) git fsck --full --unreachable - to check if there are any loose objects git repack -A -d - repacking the pack git prune - to finally remove those objects

Read the article
What's the best way to replace remote.origin.url in Git?

- by suzukimilanpaak

I'm new to Git. Let's say Alice and Bob had been developing their project by using two Git repositories for each. And, Alice at certain times want to set up a new repository to manage their common progress. Do you think what is the best way to replace remote.origin.url in the configuration of Git? to replace by git config --replace to create new repos by git clone MAIN_REPOS or any?

Read the article
Will git-svn send file permission changes to a SVN repository?

- by theForce

I'm using git-svn in a svn environment. When i check out .sh files they do not get the +x flag. So i change that manually, but now git tells me the file has been modified. My question is: If i'd stage + commit those +x changes, will git push them to the svn server when i "git svn dcommit"? This is not what i want, i just want git to 'memorize' the +x changes locally but not to try to send file permission changes to the svn repository.

Read the article
[zsh] how to clone a local git repository whose name contains a `:'?

- by zshgit

I'm trying to clone a local git repository. The repository's name contains a `:'. This is confusing both me and git. I get the following error: ~/work/c% git clone ../a::b . Initialized empty Git repository in /home/user/work/c/.git/ ssh: Could not resolve hostname ../a: Name or service not known fatal: The remote end hung up unexpectedly How would you escape the `:'? For now I'm just changing the name of the original repository :-) I'm using zshell...

Read the article
How to simplyfy getting the patch for one changeset in git?

- by childno.de

git revision syntax is really powerful, but I can't find how to simplify things like: git diff 1a2e^..1a2e without writing tree'sh1 twice? Is there now "NEXT" syntax, something like: git diff 1a2e^..NEXT OR git changeset 1a2e ?? Another "next" syntax which might be useful if 1a2e is a known revision: git cherry-pick 1a2e+NEXT..origin/featureBranch ^^ "cherry pick anything from 1a2e to master EXCEPT 1a2e itself"

Read the article
I have a global .gitignore but files aren't being ignored, why?

- by Michael Durrant

I have a .gitignore_global in my home directory durrantm.../durrantm$ pwd /home/durrantm durrantm.../durrantm$ ls .git* .gitconfig .gitignore_global The .gitignore_global has: durrantm.../durrantm$ head .gitignore_global # RubyMine # .idea/ # Compiled source # ################### *.dll *.exe # Logs and databases # ###################### but when I git status for a project I still end up getting the .idea files when I start using rubyMine. So my git status still shows this: # modified: .idea/dataSources.xml # modified: .idea/linker.iml # modified: .idea/misc.xml # modified: .idea/workspace.xml I have run git config --global core.excludesfile ~/.gitignore_global bvut it didn't help.

Read the article
Capistrano Error

- by Casey van den Bergh

I'm Running CentOS 5 32 bit version. This is my deploy.rb file on my local computer: #======================== #CONFIG #======================== set :application, "aeripets" set :scm, :git set :git_enable_submodules, 1 set :repository, "[email protected]:aeripets.git" set :branch, "master" set :ssh_options, { :forward_agent => true } set :stage, :production set :user, "root" set :use_sudo, false set :runner, "root" set :deploy_to, "/var/www/#{application}" set :app_server, :passenger set :domain, "aeripets.co.za" #======================== #ROLES #======================== role :app, domain role :web, domain role :db, domain, :primary => true #======================== #CUSTOM #======================== namespace :deploy do task :start, :roles => :app do run "touch #{current_release}/tmp/restart.txt" end task :stop, :roles => :app do # Do nothing. end desc "Restart Application" task :restart, :roles => :app do run "touch #{current_release}/tmp/restart.txt" end end And this the error I get on my local computer when I try to cap deploy. executing deploy' * executingdeploy:update' ** transaction: start * executing deploy:update_code' executing locally: "git ls-remote [email protected]:aeripets.git master" command finished in 1297ms * executing "git clone -q [email protected]:aeripets.git /var/www/seripets/releases/20111126013705 && cd /var/www/seripets/releases/20111126013705 && git checkout -q -b deploy 32ac552f57511b3ae9be1d58aec54d81f78f8376 && git submodule -q init && git submodule -q sync && export GIT_RECURSIVE=$([ ! \"git --version\" \\< \"git version 1.6.5\" ] && echo --recursive) && git submodule -q update --init $GIT_RECURSIVE && (echo 32ac552f57511b3ae9be1d58aec54d81f78f8376 > /var/www/seripets/releases/20111126013705/REVISION)" servers: ["aeripets.co.za"] Password: [aeripets.co.za] executing command ** [aeripets.co.za :: err] sh: git: command not found command finished in 224ms *** [deploy:update_code] rolling back * executing "rm -rf /var/www/seripets/releases/20111126013705; true" servers: ["aeripets.co.za"] [aeripets.co.za] executing command command finished in 238ms failed: "sh -c 'git clone -q [email protected]:aeripets.git /var/www/seripets/releases/20111126013705 && cd /var/www/seripets/releases/20111126013705 && git checkout -q -b deploy 32ac552f57511b3ae9be1d58aec54d81f78f8376 && git submodule -q init && git submodule -q sync && export GIT_RECURSIVE=$([ ! \"git --version`\" \< \"git version 1.6.5\" ] && echo --recursive) && git submodule -q update --init $GIT_RECURSIVE && (echo 32ac552f57511b3ae9be1d58aec54d81f78f8376 /var/www/seripets/releases/20111126013705/REVISION)'" on aeripets.co.za

Read the article
heroku corrupted object, git fsck fails in rails

- by Ryan Max

Hello. I am trying to push an app to heroku and I am getting the error detailed here. So I am trying to determine the corrupt objects using git fsck -full but it isn't returning anything. Nothing happens: Ryan@Ryan-PC ~ $ git fsck --full Ryan@Ryan-PC But I get the object error when I try to push the object to heroku. Is there anyway I can go about repairing the corrupt repository, or can I just delete it and start over? How do I go about doing this?

Read the article
To get a prompt which indicates Git-branch in Zsh

- by Masi

I run the following codes separately as my prompt unsuccessfully in .zshrc. This suggests me that apparently I do not have a program called __git_ps1. It is not in MacPorts. #1 PROMPT="$(__git_ps1 " \[\033[1;32m\] (%s)\[\033[0m\]")\$"$ #2 PROMPT="$(__git_ps1 " (%s)")\$"$ #3 # Get the name of the branch we are on git_prompt_info() { branch_prompt=$(__git_ps1) if [ -n "$branch_prompt" ]; then status_icon=$(git_status) echo $branch_prompt $status_icon fi } # Show character if changes are pending git_status() { if current_git_status=$(git status | grep 'added to commit' 2> /dev/null); then echo "?" fi } autoload -U colors colors setopt prompt_subst PROMPT=' %~%{$fg_bold[black]%}$(git_prompt_info) ? %{$reset_color%}' How can you get a prompt which shows the name of a Git-branch?

Read the article
Using GIT Smart HTTP via IIS

- by Andrew Matthews

I recently read Scott Chacon's post "Smart HTTP Transport", and I was hoping that it might have become possible via IIS (windows 7) since that post was written. I haven't been able to find anything showing how it can be done, and Apache is not an option in my IIS 7 based environment. So, I'm at a loss (git daemon was foiled for me by a combination of AVG anti-virus and AD). I want to provide LDAP authenticated read/write access for selected users. So this question seems not to be relevant. Do you know of a way to provide access to GIT via IIS?

Read the article
Howto add a changed file to an older (not last) commit in Git

- by David Klein

Hey, I changed several things over the last hour and committed them step by step. But I just realized I've forgot to add a changed file some commits ago. The Log looks like this: GIT TidyUpRequests u:1 d:0> git log commit fc6734b6351f6c36a587dba6dbd9d5efa30c09ce Author: David Klein <> Date: Tue Apr 27 09:43:55 2010 +0200 The Main program now tests both Webservices at once commit 8a2c6014c2b035e37aebd310a6393a1ecb39f463 Author: David Klein <> Date: Tue Apr 27 09:43:27 2010 +0200 ISBNDBQueryHandler now uses the XPath functions from XPath.fs too commit 06a504e277fd98d97eed4dad22dfa5933d81451f Author: David Klein <> Date: Tue Apr 27 09:30:34 2010 +0200 AmazonQueryHandler now uses the XPath Helper functions defined in XPath.fs commit a0865e28be35a3011d0b6091819ec32922dd2dd8 <--- changed file should go here Author: David Klein <> Date: Tue Apr 27 09:29:53 2010 +0200 Factored out some common XPath Operations Any ideas? :)

Read the article
WebDAV auto-versioning in Git or Hg or any modern VCS

- by Marcus P S

I just recently learned of SVN's auto-versioning feature for WebDAV. Although I understand this is not replacement for proper versioning, with messages documenting change sets, it strikes me as a solid and safe replacement to Dropbox (minus nice GUIs and web pages). However, since commits in auto-versioning are frequent, I'd imagine that Git or Hg would be better suited for this, just because of their more compact databases (although I wonder if the distributed nature of things could make the automation ugly for resolving conflicts). Is this a feature that has been implemented using Git or Hg, as far as anyone knows?

Read the article
Git checkout <SHA> and Heroku

- by Bob

I created a local Git repo on my laptop and then pushed the source to Heroku creating a remote branch. After a few days of commits and pushes, I need to rollback to an earlier commit. Here's what I did. cd <app root> git checkout 35fbd894eef3e114c814cc3c7ac7bb50b28f6b73 Someone told me that doing the checkout created a new working tree(?) and not the branch itself, so when I pushed the rollback changes to Heroku, it said everything is up to date and nothing was pushed. How do I fix this situation? Thanks for your help in advance.

Read the article
What type of git server do you use? or how do you use git?

- by Johan

Hi Let's say we have a small team, 1-5 persons. What type of "git setup" would you use? Would you use gitweb and apache so you could run over http? Or would you use the user-accounts and ssh in some way? Today I'm familiar to use SubVersion thou apache (http), but I'm not sure it is right to setup to use git the same way... Thanks Johan Update: It feels like if we combine the answer that Dietrich Epp and the one hallidave gave I could get a quick and good solution. A common dir in the servers filesystem where all can write, and that dir is also exposed with apache. That way everybody can always get the latest, but only trusted people can write to it...

Read the article

< Previous Page | 31 32 33 34 35 36 37 38 39 40 41 42 | Next Page >