I maintain a private Git repository with all of my config and dotfiles (.bashrc, profile.ps1, .emacs etc.).
On Windows this repository is stored under C:\git\config. Most applications expect the files to be elsewhere, so I added hard links between the repository and the expected locations.
Example
On Linux .emacs is located in ~/git/config/.emacs but emacs expects it to be at ~/.emacs. I run:
$ sudo ln -s ~/git/config/.emacs ~/.emacs
On Windows my .emacs is located in C:\git\config\.emacs, but emacs expects it to be in C:\users\ayrton\.emacs. I run:
PS> cmd /c mklink /H C:\users\ayrton\.emacs C:\git\config\.emacs
Issue
On Linux this seems to work fine: when I update the original file, the contents of the link update and everything stays in sync.
On Windows, the links break after a period of time and the files become out of sync (the file contents are different).
Why do the links break on Windows? Is there an alternative solution?
I've seen this StackOverflow post: Can't Hard Link the gitconfig File
So I’ve finally found a solution that takes the best of both: put the repo in a subdirectory, and instead of symlinks, add a configuration option for “core.worktree” to be your home directory. Now when you’re in your home directory you’re not in a git repo (so the first problem is gone), and you don’t need to deal with fragile symlinks as in the second case. You still have the minor hassle of excluding paths that you don’t want versioned (eg, the “*” in “.git/info/exclude” trick), but that’s not new.
The problem here is that the expected locations are different on Windows vs. Linux. For example, VSCode expects the user settings to be in:
Linux: $HOME/.config/Code/User/settings.json
Windows: %APPDATA%\Code\User\settings.json
Ideally I would like my repository to be platform independent. If take the core.worktree approach (e.g. make core.worktree be / or C:\, then exclude everything except specific files) I would have to maintain two copies of some configuration files when their absolute paths differ across operating systems.
Hardlinks can break if a editor opens/creates the file as a new blank file each time you save. It would not surprise me if Notepad did this because it reads the entire file into memory and has no need for the original file after it has loaded the file.
You can try to create a file symlink instead of hardlink on Windows.
Related
We're sharing SYMLINKD files on our git project. It almost works, except git modifies our SYMLINKD files to SYMLINK files when pulled on another machine.
To be clear, on the original machine, symlink is created using the command:
mklink /D Annotations ..\..\submodules\Annotations\Assets
On the original machine, the dir cmd displays:
25/04/2018 09:52 <SYMLINKD> Annotations [..\..\submodules\Annotations\Assets]
After cloning, on the receiving machine, we get
27/04/2018 10:52 <SYMLINK> Annotations [..\..\submodules\Annotations\Assets]
As you might guess, a file target type pointing at a a directory [....\submodules\Annotations\Assets] does not work correctly.
To fix this problem we either need to:
Prevent git from modifying our symlink types.
Fix our symlinks with batch script triggered on a githook
We're going we 2, since we do not want to require all users to use a modified version of git.
My limited knowledge of batch scripting is impeding me. So far, I have looked into simply modifying the attrib of the file, using the info here:
How to get attributes of a file using batch file and https://superuser.com/questions/653951/how-to-remove-read-only-attribute-recursively-on-windows-7.
Can anyone suggest what attrib commands I need to modify the symlink?
Alternatively, I realise I can delete and recreate the symlink, but how do I get the target directory for the existing symlink short of using the dir command and parsing the path from the output?
I think it's https://github.com/git-for-windows/git/issues/1646.
To be more clear: your question appears to be a manifestation of the XY problem: the Git instance used to clone/fetch the project appears to incorrectly treat symbolic links to directories—creating symbolic links pointing to files instead. So it appears to be a bug in GfW, so instead of digging it up you've invented a workaround and ask how to make it work.
So, I'd better try help GfW maintainer and whoever reported #1646 to fix the problem. If you need a stop-gap solution, I'd say a proper way would be to go another route and script several calls to git ls-tree to figure out what the directory symlinks are (they'd have a special set of permission bits;
you may start here).
So you would traverse all the tree objects of the HEAD commit, recursively,
figuring out what the symlinks pointing at directories are and then
fixup the matching entries in the work tree by deleting them
and recreating with mklink /D or whatever creates a correct sort of
symlink.
Unfortunately, I'm afraid trying to script this using lame possibilities
of cmd.exe-s scripting facilities would be an exercise in futility.
I'd take some more "real" programming language (PowerShell as an example,
and—since you're probably a Windows shop—even a .NET would be OK).
I'm working a programming project from school and decided to give Bash for windows a try. I downloaded the skeleton files for the project and pasted them inside the folder where my bash home folder should be. When I open the terminal and go inside the folder it shows as if there was nothing there even though I'm at the right folder. How can I get the files to be displayed?
EDIT: By now, there is actually a supported way to access these files. You have to use the \\wsl$\Legacy "network" share. For example, you could access your WSL home folder in Windows using \\wsl$\Legacy\home\<username>. Still, you should not directly access the lxss directory.
You are not supposed to touch the LXSS folder ever. The files there are use special attributes which are not understood by normal Windows applications to provide the features of a full Linux file system which NTFS on its own can't provide.
There is one hard-and-fast rule when it comes to Bash on Windows:
DO NOT, under ANY circumstances, create and/or modify Linux files
using Windows apps, tools, scripts, consoles, etc. Creating/changing
Linux files from Windows will likely result in data corruption and/or
damage your Linux environment requiring you to uninstall & reinstall
your distro! Note: Your “Linux files” are any of the files and folders
under %localappdata%\lxss – which is where the Linux filesystem –
distro and your own files – are stored on your drive
If you want to copy a file into a WSL directory which is not under /mnt, then use WSL to copy the file, i.e. use cp /mnt/c/original/folder/of/the/file.txt ~/ for example.
A more advanced alternative is starting a local SSH server in WSL and using Win-SSHFS to mount the WSL root as a drive in Windows. This way, file access will go through WSL.
Its unclear what you're asking in your question. What did you paste? Whole files? Text? Where did you paste it? Did you do the pasting into the bash window? If so was it inside a text file or did you just paste to the prompt? I'm going to assume that you pasted files to the directory that you believe that your bash home opens in i.e ~/. You may not be pasting the files into the correct directory because the default directory for bash on windows is not where you would assume it to be. you should be able to reach your desktop by typing cd /mnt/c/Users/yourUserNameHere/Desktop
I used to develop on linux environment, but now I need to write a git project on "windows" os. I need to ignore a directory from where the git is initiated.
like folder/.git/, now I need to ignore /folder/project/bin/*. So I have added .gitignore file on folder/.gitignore using both text pad and Visual studio.
.gitignore file contents are
/project/bin/*
Now, I expect git status should not show /project/bin folder.
Note, I have tried other ways too, but it seems that git is considering the .gitignore file as a general text file or so.
Just a guess, but I suspect that the leading forward-slash is throwing something off. I'm not sure about the trailing star either (as I reference my own .gitignore files).
Your .gitignore should be this:
project/bin/
I frustratingly found the same issue. I couldn't seem to ignore folders, but files were OK. My issue started with folders that contain spaces. All the combinations I could find after a searching (full directory path, trying to ignore case, escaping the spaces, etc) didn't resolve it.
I had to remove spaces in directory names, then I was able to ignore files.
I tried it on WSL (Windows sub-system for Linux - git was already installed), no problems.
Then reading more, I realised that if files are previously tracked, then the .gitignore won't work. I tested it by deleting all the files in the directory, then adding the directory to .gitignore path/to/my_files/*.csv (in my case I'm ignoring data in .csv files) and doing a commit then push (good job this is just dev).
After that worked, I was able to use git rm -r --cached ./path/to/my_files/ to un track all the files in another directory rather than cutting and pasting them back (Thanks to this answer)
So:
Un-track all the files
Add the directory in .gitignore (In windows it wasn't case sensitive)
add/commit/push as normal.
My organization has a group policy in effect that will only allow executable programs to reside under the Program Files (or Program Files X86) directory. Obviously, these directories are not writable by normal users. I have access to local admin rights, so I can install things there if I want. But of course it doesn't make sense to put the whole cygwin tree there, since users need to be able to write to /home, /tmp, etc. I am thinking I might be able to do something with links, maybe install to c:\cygwin, then move just the /bin directory under Program Files and create a hard link to it?
Has anyone run into a similar situation and come up with an elegant solution? This is Win 7 Enterprise.
Prior to Cygwin 1.7.34, solving such problems required a fair bit of hoop-jumping,¹ but now it's easy:
If you haven't installed Cygwin yet, do so.²
If you have Cygwin installed already and you started with a version of Cygwin prior to 1.7.34, move /etc/passwd and /etc/group out of the way,³ then upgrade to the current version.
Start the Cygwin Terminal.
Open Cygwin's /etc/nsswitch.conf in your favorite text editor.⁴ Add a line like this:
db_home: /%H
That's it! When you next re-start Cygwin, it will treat your Windows profile directory as your Cygwin home directory.⁵ This means you will have useful sub-folders like Desktop and Downloads as sub-folders, which matches the way OS X and a lot of desktop Linuxes work.
Some people might not want these two directories to be treated as equivalents. You can choose any path scheme you like. For example, you could change it to /cygdrive/c/Users/%U/cygwin to put your Cygwin home folder into a cygwin subdirectory of your Windows profile directory.
This new feature of Cygwin is very powerful and can do a lot more than I show here. For example, you can change a Cygwin user home directory via AD instead, if you like. See that documentation for details.
You might also want to rearrange a few other elements of the Cygwin path scheme. You can do so by editing Cygwin's /etc/fstab file. You probably want to move /tmp, /usr/tmp, and /var/tmp to a directory that non-admin users can write to, for one thing:
c:/tmp /tmp ntfs auto 0 0
c:/tmp /usr/tmp ntfs auto 0 0
c:/tmp /var/tmp ntfs auto 0 0
This will let you install (and later update) Cygwin as an Administrator while still letting unprivileged users run Cygwin. This makes Cygwin behave more like Linux or Unix. Since most software in the Cygwin package repository comes from that world, you can count on it to behave correctly under such a scheme.
Footnotes:
The first version of this answer tells you how to do an equivalent thing with older versions of Cygwin.
Cygwin doesn't care where you install it, so if the default doesn't work for you, feel free to change it. Some ideas:
C:\Program Files\Cygwin
C:\Users\jeremy\Cygwin
D:\cygwin
Cygwin will remember your choice on subsequent updates.
If you don't move these files out of the way, they interfere with the solution we build above. This part of Cygwin is complex enough to deserve a whole section in the Cygwin user manual. Hint. :)
Cygwin installs a stripped-down version of the Vim text editor by default.
If you don't like vi, there are many other text editors in the Cygwin package repository.
If you do like vi, I suggest installing the full version of Vim, then adding alias vi=vim to your ~/.bashrc.
You can also use a native Windows GUI text editor. Cygwin's /etc/nsswitch.conf parser appears to cope with DOS line endings.
The Cygwin DLL is building this path from the %HOMEDRIVE% and %HOMEPATH% environment variables, then converting it to POSIX form.
Is there some sort of general pathname-rewriting/normalizing facility I can exploit in Emacs to allow source files in compile-mode to be translated to local paths based on remote paths? I was hoping to use symbolic links, but that's not working for me in Windows.
I'm running GNU Emacs 23.1.1, and have tried cygwin-mount and w32-symlinks.el (with (customize-option 'w32-symlinks-handle-shortcuts)); I can't get open-file/dired/completions OR compile-mode next-error to follow either windows shortcuts or cygwin symlinks.
I'm remotely compiling (over ssh) and the warnings/errors give filenames where a prefix of the absolute pathnames (e.g. /nfs/topaz) on the remote host needs to be replaced with a different path (z:/, in fact), which I was hoping to do with symbolic links.
/ does go to (gets tab-expanded to in emacs) c:/cygwin and ~ to z:/ (my $HOME).
What I'm hoping for is either: a fix for symlink resolution (tab completion or opening a windows shortcut gets me the .lnk file and not what it points to), or the ability to supply a list of rewrites for prefixes of pathnames. Junctions aren't an option because the replacement path I want is on a different drive (it's also a Samba mount).
Here's a partial solution:
(custom-set-variables '(compilation-search-path (quote (nil "z:/dir1" "~/dir2"))
etc. Of course, I have to add dozens of paths, but it's a huge improvement (next-error actually works now!)
This is also useful in other complicated build environments: I use OpenWRT, and the build process copies (or extracts) source code to a separate build directory in which compilation occurs. So naively following file names in the emacs compilation buffer leads to edit the temporary copies, not the originals. Dangerous.
After re-reading a few times, I still am not quite sure what your situation is -
Are you compiling remotely and editing locally? If so, it seems like you should try doing both tasks on the same machine, or maybe try tramp mode in emacs. I just don't see an easy way to translate a remote path to a local one.