The solution is to place a leading slash on the `.gitignore` entries:

    /foo/
    /bar.txt


(I thought I tried this before posting on StackOverflow, but clearly I hadn&#39;t tried it properly, as this works perfectly.)

From the gitignore manpage:

&gt; An optional prefix ! which negates the pattern; any matching file excluded by a previous pattern will become included again. If a negated pattern matches, this will override lower precedence patterns sources.

So `!*` as the first line in your .gitignore will clear all previous patterns.

I&#39;m performance tuning interactive games in Java for the Android platform.  Once in a while there is a hiccup in drawing and interaction for garbage collection.  Usually it&#39;s less than one tenth of a second, but sometimes it can be as large as 200ms on very slow devices.

I am using the ddms profiler (part of the Android SDK) to search out where my memory allocations come from and excise them from my inner drawing and logic loops.  

The worst offender had been short loops done like,

    for(GameObject gob : interactiveObjects)
        gob.onDraw(canvas);

where every single time the loop was executed there was an `iterator` allocated.  I&#39;m using arrays (`ArrayList`) for my objects now.  If I ever want trees or hashes in an inner loop I know that I need to be careful or even reimplement them instead of using the Java Collections framework since I can&#39;t afford the extra garbage collection.  That may come up when I&#39;m looking at priority queues.

I also have trouble where I want to display scores and progress using `Canvas.drawText`.  This is bad,

    canvas.drawText(&quot;Your score is: &quot; + Score.points, x, y, paint);

because `Strings`, `char` arrays and `StringBuffers` will be allocated all over to make it work.  If you have a few text display items and run the frame 60 times a second that begins to add up and will increase your garbage collection hiccups.  I think the best choice here is to keep `char[]` arrays and decode your `int` or `double` manually into it and concatenate strings onto the beginning and end.  I&#39;d like to hear if there&#39;s something cleaner.

I know there must be others out there dealing with this.  How do you handle it and what are the pitfalls and best practices you&#39;ve discovered to run interactively on Java or Android?  These gc issues are enough to make me miss manual memory management, but not very much.

How can I avoid garbage collection delays in Java games? (Best Practices)

I have attempted to implement a producer/consumer pattern in c#.  I have a consumer thread that monitors a shared queue, and a producer thread that places items onto the shared queue.  The producer thread is subscribed to receive data...that is, it has an event handler, and just sits around and waits for an OnData event to fire (the data is being sent from a 3rd party api).  When it gets the data, it sticks it on the queue so the consumer can handle it. 

When the OnData event does fire in the producer, I had expected it to be handled by my producer thread.  But that doesn&#39;t seem to be what is happening.  The OnData event seems as if it&#39;s being handled on a new thread instead!  Is this how .net always works...events are handled on their own thread?  Can I control what thread will handle events when they&#39;re raised?  What if hundreds of events are raised near-simultaneously...would each have its own thread?


In .NET, what thread will Events be handled in?

How to I add something to the `.gitignore` so that the match is not recursive?

For example, I wish to ignore the directory `foo` and the file `bar.txt` in the current directory, but not any that exist in subdirectories.

I have tried this for my `.gitignore` file:

    foo/
    bar.txt

But unfortunately git applies this recursively, so that `otherdir/bar.txt` and `otherdir/foo/` also get ignored, which is not what I want.

(Is there a command in git that shows me all ignored files, and reference the `.gitignore` file that is responsible for the file being ignored?  This would be useful for debugging.)


How to I add something to the .gitignore so that the match is not recursive?

How to I add something to the <code>.gitignore</code> so that the match is not recursive?
For example, I wish to ignore the directory <code>foo</code> and the file <code>bar.txt</code> in the current directory, but not any that exist in subdirectories.
I have tried this for my <code>.gitignore</code> file:
<pre><code class="hljs">foo/
bar.txt
</code></pre>
But unfortunately git applies this recursively, so that <code>otherdir/bar.txt</code> and <code>otherdir/foo/</code> also get ignored, which is not what I want.
(Is there a command in git that shows me all ignored files, and reference the <code>.gitignore</code> file that is responsible for the file being ignored? This would be useful for debugging.)

After cloning an SVN repository using git-svn with the `-s` option (`git svn clone http://server/repo -s`), how does one create a branch or tag and have pushed to the relevant branch/tag directory in the repository when `dcommit`ing?

For instance; if I were to use git to create a `foobar` branch locally (`git checkout -b foobar`) how can I have git-svn **create** the branch on the server (`http://server/repo/branches/foobar`)?

**I&#39;m using Git 1.5.5.6.**

---

**Please Note:**

The accepted method below **does not work with Git 1.5.5.6** as there is no `git svn branch` method. I&#39;m still looking for a solution to this that doesn&#39;t involve resolving to working with svn directly.

Git-svn: create &amp; push a new branch/tag?

This morning we pulled from our repo, and git put us on (no branch).

I don&#39;t understand this, why did this happen? And how to get out of it without losing our changes?

Why did git set us on (no branch)?

I&#39;ve went through the [excellent guide provided by Tim Davis][1] which is about configuring Git to work with SSH under Windows in order to produce a Git Server in order to have a main place for my DVCS. 

I am in the process of creating a clone for my project. I’ve went through all the steps till this point, but I keep getting this from TortoiseGit:

    git.exe clone -v “ssh://Administrator@127.0.0.1:22/SSH/Home/administrator/myapp.git” “E:\GitTest\myapp”
    
    bash: Administrator@127.0.0.1: command not found
    Initialized empty Git repository in E:/GitTest/myapp/.git/
    fatal: The remote end hung up unexpectedly
    Success

and nothing gets cloned.

BTW: The `TortoisePLink` comes up just before this message appears and asks me: “`login as:`” ( I thought that this info is given in the command, i.e: `Administrator@blahblah`.

My home variable is set to the correct place:
From a Git Bash shell:

    echo $HOME
    /c/SSH/home/Administrator

I’ve also tried using Putty’s plink instead of `TortoisePLink` (in both Git’s and `TortoiseGit`’s installation). This time the error was narrowed down to:

    git.exe clone -v “ssh://Administrator@127.0.0.1:22/c:/SSH/Home/administrator/myapp.git” “E:\GitTest\myapp”
    
    Initialized empty Git repository in E:/GitTest/myapp/.git/
    fatal: The remote end hung up unexpectedly


  [1]: https://www.timdavis.com/posts/setting-up-a-msysgit-server-with-copssh-on-windows






Git with SSH on Windows

I can&#39;t really ever think of a time when I would use `git merge` rather than `git rebase` and *not* want to have a commit show up.  Is there any way to configure git to have fast forwarding off by default?  The fact that there&#39;s an `--ff` option would seem to imply that there&#39;s a way, but I can&#39;t seem to find it in the documentation.

Can I make fast forwarding be off by default in git?

I want to know if there is a way to set a flag by default for git command. Specifically, I want to set the `--abbrev-commit` flag so that when executing `git log `, I want to execute `git log --abbrev-commit`.

Unlike the question &quot;[is there any way to set a flag by default for a git command?][1]&quot;, there is apparently not a configuration flag for adding --abbrev-commit to git log. Furthermore, the git manual states that I cannot create an alias: [&quot;To avoid confusion and troubles with script usage, aliases that hide existing git commands are ignored&quot;][2]

My third option is to invent a new alias like `glog=log --abbrev-commit` in my .gitconfig file. But I&#39;d rather not invent my own DSL with new commands. 

Is there another way to achieve it so that the `abbrev-commit` flag is set by default??

   [1]: https://stackoverflow.com/questions/1278296/is-there-any-way-to-set-a-flag-by-default-for-a-git-command
   [2]: https://git-scm.com/docs/git-config#Documentation/git-config.txt-alias



Setting git default flags on commands

After installing and using Visual Studio 2010, I&#39;m seeing some newer file types (at least with C++ projects ... don&#39;t know about the other types) as compared to 2008.  e.g. .sdf, .opensdf, which I guess are the replacement for ncb files with Intellisense info stored in SQL Server Compact files?  I also notice .log files are generated, which appear to be build logs.

Given this, what&#39;s safe to add to my global ignore pattern?  Off the bat, I&#39;d assume .sdf, .opensdf, but what else?


What should be contained in a global source code control ignore pattern for Visual Studio 2010?

I have a lot of projects in my **.Net** solution. I would like to exclude all &quot;bin/Debug&quot; and &quot;bin/Release&quot; folders (and their contents), but still include the &quot;bin&quot; folder itself and any dll&#39;s contained therein.

`.gitignore` with &quot;bin/&quot; ignores &quot;**Debug**&quot; and &quot;**Release**&quot; folders, but also any dll&#39;s contained in the &quot;bin&quot; folder.

`bin/Debug` or `bin/Release` in the `.gitignore` file does not exclude the directories unless I fully qualify the ignore pattern as `Solution/Project/bin/Debug` - which I don&#39;t want to do as I will need to include this full pattern for each project in my solution, as well as add it for any new projects added.

Any suggestions?

How to git ignore subfolders / subdirectories?

If I have a working copy of a Subversion repository, is there a way to delete all unversioned or ignored files in that working copy with a single command or tool?  Essentially, I&#39;m looking for the SVN analogue to [`git clean`][1].

Either a command line or GUI solution (for TortoiseSVN) would be acceptable.


  [1]: http://www.kernel.org/pub/software/scm/git/docs/git-clean.html

How can I delete all unversioned/ignored files/folders in my working copy?

I love to use `git diff -w` to ignore whitespace differences. But, I just noticed that it ignores even whitespace differences in the middle of lines. How could I only ignore whitespace differences that come at the start (^) or end ($) of lines?

Git diff -w ignore whitespace only at start &amp; end of lines

What is the correct way to make git ignore temporary files produced by vim in all directories (either globally across the system or locally for a single project)?

git ignore vim temporary files

What should be the content of the .gitignore file for a java project in netbeans?


what should be in .gitignore file for a netbeans java project?

When you use

    git rm --cached myfile

it doesn&#39;t delete from the local filesystem, which is the goal. But if you&#39;ve already versioned and committed the file, pushed it to a central repository, and pulled it into yet another repository before using the command, it will delete the file from that system.

Is there a way to just remove the file from versioning without deleting it from any filesystem?

Edit: Clarified, I hope.

Git: How to remove file from index without deleting files from any repository

I am attempting to use a .gitignore file with negated patterns (lines starting with !), but it&#39;s not working the way I expect.

As a minimal example, I have the folllowing directory structure:

    C:/gittest
     -- .gitignore
     -- aaa/
       -- bbb/
         -- file.txt
       -- ccc/
         -- otherfile.txt

and in my gitignore file, I have this:

    aaa/
    !aaa/ccc/

My understanding (based on [this doc page][1]) is that the file aaa/ccc/otherfile.txt should not be ignored, but in fact git is ignoring everything under aaa.

Am I misunderstanding this sentence: &quot;An optional prefix ! which negates the pattern; any matching file excluded by a previous pattern will become included again.&quot;?

BTW, this is on Windows with msysgit 1.7.0.2.

   


  [1]: https://git-scm.com/docs/gitignore

How do negated patterns work in .gitignore?

I have configured numerous `.gitignore` files to filter out many different unwanted files from a set of about 6,000 untracked files. I want to do `git add .` when I&#39;ve got my filtered list looking the way I want it.

But, then I want to disable the `.gitignore` filters temporarily to see what got left behind, and make sure there was nothing important accidentally filtered.

I know that `git-clean` includes an option to ignore `.gitignore files`. Is there a similar option for `git-status`?

I could go through and delete all the `.gitignore` files, do the check, then restore them, but it seems there should be an easier way?

Content Type	Original Author	Original Content on Stackoverflow
Question	pauldoo	View Question on Stackoverflow
Solution 1 - Git	pauldoo	View Answer on Stackoverflow
Solution 2 - Git	hermannloose	View Answer on Stackoverflow

How to I add something to the .gitignore so that the match is not recursive?

Git Problem Overview

Git Solutions

Solution 1 - Git

Solution 2 - Git

In .NET, what thread will Events be handled in?

How can I avoid garbage collection delays in Java games? (Best Practices)

Attributions