1

I can't understand how git lfs works. Do I need to maintain LFS store ? Do I need to do "git lfs track *.dat" every time ? Do I need also do "git add *.dat" every time ?

  • I've never used GIT Large File Storage, but it seems you have to add `git lfs track *.dat` in `.gitattributes` – Ron van der Heijden Mar 08 '18 at 08:19
  • I know. My question is: Do you I need also do git add file.dat ? git lfs need to do only once time ? @RonvanderHeijden –  Mar 08 '18 at 08:50

1 Answers1

1

  1. Do I need to maintain LFS store?

    No, the LFS store is maintained by the Git server.

  2. Do I need to do git lfs track *.dat every time?

    No, when you execute git lfs track *.dat, a .gitattributes file with the following content is created:

     *.dat filter=lfs diff=lfs merge=lfs -text

    From now on every .dat file you create will be stored in the Git LFS. Do not forget to also add/commit/push the .gitattributes file so you never will need to recreate it again, in a case you (or someone else) needs to create the local repository.

  3. Do I need also do git add *.dat every time?

    Yes, the git add command continues to work the same way.

To learn more about Git LFS and understand the issue it solves, see the What is Git LFS? video.

Gabriel Staples
  • 36,492
  • 15
  • 194
  • 265
Marcelo Ávila de Oliveira
  • 19,950
  • 3
  • 39
  • 50
  • Can you explain how it works ? I can't understand why need to separate repository to LFS store and bitbucket ? what is goal of idea ? @Marcelo Ávila de Oliveira –  Mar 08 '18 at 12:44
  • I updated my answer with a nice video to understand the Git LFS idea. – Marcelo Ávila de Oliveira Mar 08 '18 at 13:34
  • I saw it but i don't understand the main idea. How does it save place ? @Marcelo Ávila de Oliveira –  Mar 08 '18 at 13:55
  • When you clone a Git repository you receive all versions of all files. This can be too much time consuming for binary files. The Git LFS enables to keep binary files in a different place, so when you clone a repository Git LFS will manage to you receive just one version of the binary files saving you a lot of time (and resources). – Marcelo Ávila de Oliveira Mar 08 '18 at 14:46
  • The goal of lfs save time of clone only ? what is about of push and pull ? Do I only get pointer to binary file ? @Marcelo Ávila de Oliveira –  Mar 08 '18 at 14:50
  • Just in the clone. No, if you have the Git LFS correctly installed you'll receive the real file not the pointer. But again: just the version needed (accordingly with your Git HEAD) not all versions. – Marcelo Ávila de Oliveira Mar 08 '18 at 15:03
  • In simple git I also get needed verstion if I wrote: git clone -b @Marcelo Ávila de Oliveira –  Mar 08 '18 at 15:57
  • You get the needed version AND ALL OTHER VERSIONS! – Marcelo Ávila de Oliveira Mar 08 '18 at 16:48
  • Very useful video link! I added it to my references in my answer here: [How to use `git lfs` as a basic user: What is the difference between `git lfs fetch`, `git lfs fetch --all`, `git lfs pull`, and `git lfs checkout`?](https://stackoverflow.com/a/72610495/4561887) – Gabriel Staples Nov 21 '22 at 22:33