You&#39;ve got what `rebase` does backwards. `git rebase master` does what you&#39;re asking for — takes the changes on the current branch (since its divergence from master) and replays them on top of `master`, then sets the head of the current branch to be the head of that new history. It *doesn&#39;t* replay the changes from `master` on top of the current branch.

Another way to look at it is to consider `git rebase master` as:

&gt; Rebase the current branch **on top of** `master`

Here , &#39;`master`&#39; is the **upstream** branch, and that explain why, during a rebase, [`ours` and `theirs` are reversed][1].


  [1]: https://stackoverflow.com/questions/2959443/why-is-the-meaning-of-ours-and-theirs-reversed-with-git-svn/2960751#2960751

when rebasing current changes on top of master, you can: 
1. pull down latest master: `git pull &lt;remote_name&gt; master`
2. checkout the branch you want to rebase changes into: `git checkout &lt;branch_name&gt;`
3. perform rebase: `git rebase master`

an alternative path that lets you rebase changes from a remote branch (origin/master, for example)into a local branch without updating the remote branch locally is: `git rebase origin/master` 

Classes such as `Stream`, `StreamReader`, `StreamWriter` etc implements `IDisposable` interface. That means, we can call `Dispose()` method on objects of these classes. They&#39;ve also defined a `public` method called `Close()`. Now that confuses me, as to what should I call once I&#39;m done with objects? What if I call both? 

My current code is this:


    using (Stream responseStream = response.GetResponseStream())
    {
       using (StreamReader reader = new StreamReader(responseStream))
       {
          using (StreamWriter writer = new StreamWriter(filename))
          {
             int chunkSize = 1024;
             while (!reader.EndOfStream)
             {
                char[] buffer = new char[chunkSize];
                int count = reader.Read(buffer, 0, chunkSize);
                if (count != 0)
                {
                   writer.Write(buffer, 0, count);
                }
             }
             writer.Close();
          }
          reader.Close();
       }
    }

As you see, I&#39;ve written `using()` constructs, which automatically call `Dispose()` method on each object. But I also call `Close()` methods. Is it right?

Please suggest me the best practices when using stream objects. :-)

MSDN example doesn&#39;t use `using()` constructs, and call `Close()` method:

  - [How to: Download Files with FTP][1]

Is it good?

[1]:http://msdn.microsoft.com/en-us/library/ms229711.aspx

Should I call Close() or Dispose() for stream objects?

In virtualenvwrapper, is there a simple way to list all virtualenv on my machine? 

(like what yolk -l does to list all python packages in the current virtual environment?)

**CLARIFICATION:** &quot;ls -la&quot; in my env directory does not count. I am looking for a virtualenv or virtualenvwrapper specific command.

List all virtualenv

Okay. If I&#39;m on a branch (say `working`), and I want to merge in the changes from another branch (say `master`), then I run the command `git-merge master` while on the `working` branch, and the changes get merged in without rebasing the history at all. If I run `git-rebase master`, then the changes in `master` are rebased to be put on the top of my `working` branch. But what if I want to merge in the changes from `master` but rebase my changes in `working` to be on top? How do I do that? Can it be done?

I could run `git-rebase working` on my `master` branch to put my changes on top in the `master` branch, but I&#39;d like to be able to do that in my `working` branch, and I have no idea how. The closest that I can think of doing is creating a new branch from `master` and then rebase `working`&#39;s changes on top of that, but then I&#39;d have a new branch instead of altering the `working` branch.

How do you rebase the current branch&#39;s changes on top of changes being merged in?

<p>Okay. If I'm on a branch (say <code>working</code>), and I want to merge in the changes from another branch (say <code>master</code>), then I run the command <code>git-merge master</code> while on the <code>working</code> branch, and the changes get merged in without rebasing the history at all. If I run <code>git-rebase master</code>, then the changes in <code>master</code> are rebased to be put on the top of my <code>working</code> branch. But what if I want to merge in the changes from <code>master</code> but rebase my changes in <code>working</code> to be on top? How do I do that? Can it be done?</p>
<p>I could run <code>git-rebase working</code> on my <code>master</code> branch to put my changes on top in the <code>master</code> branch, but I'd like to be able to do that in my <code>working</code> branch, and I have no idea how. The closest that I can think of doing is creating a new branch from <code>master</code> and then rebase <code>working</code>'s changes on top of that, but then I'd have a new branch instead of altering the <code>working</code> branch.</p>


What is the difference between `git clone` and `git checkout`?

What is the difference between git clone and checkout?

Why does Git use

    git push &lt;remote&gt; :&lt;branch&gt;

as in 


    git push origin :featureA

to delete the branch `featureA` from the remote server?

I am interested in why the colon was used as the delete flag. 

It&#39;s so different from `git branch -d &lt;localbranch&gt;`.

Why don&#39;t we do something like

    git branch -d --remote origin &lt;branchname&gt;

or is there a deeper meaning behind the colon symbol that I didn&#39;t know?


Why does Git use a colon (:&lt;branch&gt;) to delete remote branches?

How do I add SSH keys for &#39;apache&#39; user in Linux? 

**BACKGROUND** 

I am trying to add a service hook to github to notify a URL once I push to my repo. I have the following php page set up:

    &lt;?php `git pull origin master`;  

However I get the following output:

    sh: git: Permission denied

This is because the keys I generated for github access were generated by my &#39;root&#39; user. However when I exectue a command from php it is the &#39;apache&#39; user that runs it.

The keys therefore do not correspond and permission is denied to pull.

As I cannot switch user from the terminal to generate keys as &#39;apache&#39;, I am not too sure what to do. Can anyone suggest a solution?


Generating SSH keys for &#39;apache&#39; user

I could probably setup an alias, but it seems like I should be able to set this as an option in the config file, only I don&#39;t see anyway to do it.

I only want the `--ignore-space-change` when I&#39;m doing diff, not when I&#39;m doing apply or anything else.  I&#39;m trying to make the diff easier to understand by not cluttering it with with extraneous +/- lines that have no real changes on them.

How to make git diff --ignore-space-change the default

In Git, I understand that a branch is a pointer to a commit.

How do I make a specific branch point to a specific commit? Say I want to make `master` point at `1258f0d0aae...`, how do I do that?

How do I make a branch point at a specific commit?

Is there any way of simulating a `git merge` between two branches, the current working branch and the master, but without making any changes?

I often have conflicts when I have to make a `git merge`. Is there any way of simulating the merge first?

How to test a merge without actually merging first

I&#39;m facing a peculiar problem with SVN `merge`. I want to merge from a dev branch to trunk.
We have multiple dev branches cut off the trunk at the same time.

I&#39;m merging one of those branches to trunk with this command:

    svn merge trunk branch_1

I see changes that are not part of this branch, getting merged into trunk.
What am I doing wrong ?

SVN Version : 

&gt; Subversion command-line client, version
&gt; 1.6.16-SlikSvn-tag-1.6.16@1076804-WIN32.

Merge a Branch into Trunk

How can I merge the columns of two data frames, containing a distinct set of columns but *some* rows with the same names? The fields for rows that don&#39;t occur in both data frames should be filled with zeros:

    &gt; d
        a   b   c   d   e   f   g   h   i  j
    1 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10
    2 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9  1
    &gt; e
       k  l  m  n  o  p  q  r  s  t
    1 11 12 13 14 15 16 17 18 19 20
    3 21 22 23 24 25 26 27 28 29 30
    &gt; de
        a   b   c   d   e   f   g   h   i  j  k  l  m  n  o  p  q  r  s  t
    1 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10 11 12 13 14 15 16 17 18 19 20
    2 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9  1  0  0  0  0  0  0  0  0  0  0
    3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  0 21 22 23 24 25 26 27 28 29 30

Merge data frames based on rownames in R

I have a list of many data.frames that I want to merge. The issue here is that each data.frame differs in terms of the number of rows and columns, but they all share the key variables (which I&#39;ve called `&quot;var1&quot;` and `&quot;var2&quot;` in the code below). If the data.frames were identical in terms of columns, I could merely `rbind`, for which plyr&#39;s [rbind.fill](https://stackoverflow.com/questions/2851327/r-converting-a-list-of-data-frames-into-one-data-frame) would do the job, but that&#39;s not the case with these data. 

Because the `merge` command only works on 2 data.frames, I turned to the Internet for ideas. I got this one from [here](http://www.mail-archive.com/r-help@r-project.org/msg46572.html), which worked perfectly in R 2.7.2, which is what I had at the time:

    merge.rec &lt;- function(.list, ...){
        if(length(.list)==1) return(.list[[1]])
        Recall(c(list(merge(.list[[1]], .list[[2]], ...)), .list[-(1:2)]), ...)
    }

And I would call the function like so:

    df &lt;- merge.rec(my.list, by.x = c(&quot;var1&quot;, &quot;var2&quot;), 
                    by.y = c(&quot;var1&quot;, &quot;var2&quot;), all = T, suffixes=c(&quot;&quot;, &quot;&quot;))

But in any R version after 2.7.2, including 2.11 and 2.12, this code fails with the following error:

    Error in match.names(clabs, names(xi)) : 
      names do not match previous names

(Incidently, I see other references to this error [elsewhere](https://stackoverflow.com/questions/2504543/merge-multiple-data-frames-error-in-match-namesclabs-namesxi-names-do-n) with no resolution).

Is there any way to solve this? 

Simultaneously merge multiple data.frames in a list

Is there a way to merge XCode 4.2 Storyboard files? 

I&#39;m working with another developer on an iPhone project and all my attempts at merging changes into a storyboard file have met with failure and after which Xcode is no longer able to load the storyboard. Consequently I have to do an svn revert and then redo my modifications.

XCode Storyboard merging

I&#39;d like to rebase to a specific commit, not to a HEAD of the other branch:

    A --- B --- C          master
     \
      \-- D                topic

to

    A --- B --- C          master
           \
            \-- D          topic

instead of 

    A --- B --- C          master
                 \
                  \-- D    topic


How can I achieve that?

Git: How to rebase to a specific commit?

I have conflicting branches, `feature_x` branched from `main`.

Let&#39;s say when rebasing `feature_x` on current `main`, while resolving conflicts, I decide to take **some** (not all) of &quot;their&quot; (i.e. `main`) files as-is.
How do I do that?

I tried:
```bash
git checkout main:foo/bar.java
fatal: reference is not a tree: TS-modules-tmp:foo/bar.java
  
git checkout refs/heads/main:foo/bar.java
fatal: reference is not a tree: refs/heads/TS-modules-tmp:foo/bar.java
```
    


How to get &quot;their&quot; changes in the middle of conflicting Git rebase?

I am facing an issue that I am not sure how to resolve.

I did a rebase against master from my branch:

    git rebase master

and got the following error

     First, rewinding head to replay your work on top of it...
     Applying: checkstyled.
     Using index info to reconstruct a base tree...
     Falling back to patching base and 3-way merge...
     Auto-merging AssetsLoader.java
     CONFLICT (content): Merge conflict in AssetsLoader.java
     Failed to merge in the changes.
     Patch failed at 0001 checkstyled.

So I went to my favourite editor, fixed the 1 line conflict, saved the file and did a git status and got the following output:

     # Not currently on any branch.
     # Changes to be committed:
     #   (use &quot;git reset HEAD &lt;file&gt;...&quot; to unstage)
     #
     #	modified:   PassengerContactHandler.java
     #
     # Unmerged paths:
     #   (use &quot;git reset HEAD &lt;file&gt;...&quot; to unstage)
     #   (use &quot;git add/rm &lt;file&gt;...&quot; as appropriate to mark resolution)
     #
     #	both modified:      AssetsLoader.java
     #
     
I did a git add AssetsLoader.java and a git status and got the following:

     # Not currently on any branch.
     # Changes to be committed:
     #   (use &quot;git reset HEAD &lt;file&gt;...&quot; to unstage)
     #
     #	modified:   AssetsLoader.java
     #	modified:   PassengerContactHandler.java
     #

and when I did git rebase --continue I get:

    git rebase --continue
    You must edit all merge conflicts and then
    mark them as resolved using git add


I know I can skip the patch and continue the rebase, but I am not sure if the changes in PassengerContactHandler.java will be rebased into my branch or not.

so I am not sure, How should I proceed?

Edit: Could it be that the file with the resolved conflict is exactly like the original version?

Thanks a lot,
Lucas

Edit, it just happened to me again:

It just happened to me again,

    (307ac0d...)|REBASE)$ git status
    # Not currently on any branch.
    # Changes to be committed:
    #   (use &quot;git reset HEAD &lt;file&gt;...&quot; to unstage)
    #
    #	modified:   assets/world/level1/Level-1.xml
    #	modified:   George.java
    #	modified:   DefaultPassenger.java
    #
    # Untracked files:
    #   (use &quot;git add &lt;file&gt;...&quot; to include in what will be committed)
    #
    #	mb-art/originalAssets/27dec/

((307ac0d...)|REBASE)$ git rebase --continue

    You must edit all merge conflicts and then
    mark them as resolved using git add


git --version

    git version 1.7.1


          

Git rebase --continue complains even when all merge conflicts have been resolved

I was reading on:
https://wiki.diasporafoundation.org/Git_workflow#Rebase_your_development_branch_on_the_latest_upstream

Here is an extract:

&gt; Your Repository Up to Date
&gt; 
&gt; In order to get the latest updates from the development trunk do a
&gt; one-time setup to establish the main GitHub repo as a remote by
&gt; entering:
&gt; 
&gt; 
&gt;     $ git remote add upstream git://github.com/diaspora/diaspora.git
&gt; 
&gt; Rebase Your Development Branch on the Latest Upstream
&gt; 
&gt; To keep your development branch up to date, rebase your changes on top
&gt; of the current state of the upstream master. See the What’s
&gt; git-rebase? section below to learn more about rebasing.
&gt; 
&gt; If you’ve set up an upstream branch as detailed above, and a
&gt; development branch called 100-retweet-bugfix, you’d update upstream,
&gt; update your local master, and rebase your branch from it like so:
&gt; 
&gt;     $ git fetch upstream
&gt; 
&gt;     $ git checkout master
&gt; 
&gt;     $ git rebase upstream/master
&gt; 
&gt;     $ git checkout 100-retweet-bugfix
&gt; 
&gt; [make sure all is committed as necessary in branch]
&gt; 
&gt;     $ git rebase master

Why is adding a &#39;remote upstream&#39; needed in this case? Coudn&#39;t I have just done:

    $ git checkout master

    $ git pull origin master

    $ git checkout 100-retweet-bugfix

[make sure all is committed as necessary in branch]

    $ git rebase master

What does &#39;git remote add upstream&#39; help achieve?

My boss decided recently to try out an outsourcing group &quot;for increased capacity&quot; as we modify an existing application to add new features. Despite my concern that the group he chose didn&#39;t seem to communicate well and wasn&#39;t asking enough questions to truly understand and be effective, we went with them anyway.

We set them up to collaborate on our Git repository (hosted on GitHub). We created a branch just for them (TheOutsourcedBranch, we&#39;ll call it) and requested that they to do all of their work within this branch.

Additionally, we asked that they push their commits at the end of each day so that we can get a decent idea of how fast they work and (more importantly) how good their code is.

Today they pushed for the first time. Five commits, all of them in master. I&#39;m trying to figure out how to move their commits from master into TheOutsourcedBranch, without losing the work they&#39;ve done (although from the looks of it, we&#39;re going to have to throw it away anyway).

![Visual Reference -- Orange dots are their commits, gray are mine.][1]

Orange dots are their commits, gray dots are mine.

I know there are a zillion questions about rebasing (and that might be what I need to do here) but I&#39;m having a hard time figuring out which of them apply to my specific case. And in case it matters, the outsourcing group and myself are the only people that are really using the repo right now.

Thanks in advance!


----------


**Edit:**
I may have found what I&#39;m looking for: https://stackoverflow.com/questions/1628563/git-move-recent-commit-to-a-new-branch

I&#39;d like to check my understanding, though. Locally, I create a branch off of master (TheOutsourcedBranch), which would contain everything A-H. Then I reset master back to C. Master would now contain A-C.

Assuming that is correct, that would be fine for me (locally). But then what do I need to do to &quot;force&quot; the remote repository (GitHub) to accept my local view of how things played out and discard its version of the history?

Once this has played out I think I can re-work this question to make it more generic/informative and cut out some fluff--otherwise it&#39;s probably a good candidate for deletion, I would guess.

  [1]: http://i.stack.imgur.com/2bkyX.png

Git - Moving Pushed Commits to a Different Branch

I have a cloned project from a master branch from remote repository `remote_repo`. I create a new branch and I commit to that branch. Other programmers pushed to `remote_repo` to the master branch. 

I now need to rebase my local branch `RB` onto `remote_repo`&#39;s `master` branch. 

How to do this? What commands to type to a terminal?

How to rebase local branch onto remote master

I&#39;m having a problem with my git repo. For the last couple of days whenever I do a push to the server I get this message: &quot;Auto packing the repository for optimum performance&quot;, and it does not seem to go away and return the shell.

I also tried checking out to a new branch and then doing a rebase on my previous branch and then did `git gc` to remove the unused history objects and then did a push but still this message appears. Please let me know what&#39;s going on with my repo.

Content Type	Original Author	Original Content on Stackoverflow
Question	Jonathan M Davis	View Question on Stackoverflow
Solution 1 - Git	hobbs	View Answer on Stackoverflow
Solution 2 - Git	VonC	View Answer on Stackoverflow
Solution 3 - Git	Ben Perlmutter	View Answer on Stackoverflow

How do you rebase the current branch's changes on top of changes being merged in?

Git Problem Overview

Git Solutions

Solution 1 - Git

Solution 2 - Git

Solution 3 - Git

List all virtualenv

Should I call Close() or Dispose() for stream objects?

Attributions