0

I had a regular file named ScriptA.sh and renamed it to ScriptB.sh.

I created a symlink ScriptA.sh pointing to ScriptB.sh.

When I commit/push it and go check at the remote repository, there has ScriptA.sh and ScriptB.sh with the full script content... I was expecting ScriptA.sh to have only the path to ScriptB.sh as its data...

Worse is: if I change the data at ScriptB.sh, only it is detected as having changed, therefore I can only commit that one... and cannot commit ScriptA.sh!

I thought if I changed the symlink contentet of ScriptA.sh to be ../bin/ScriptB.sh, commit and then change it back, and commit again, it may fix? or may I mess something? EDIT: I just tested and the symlink change was not detected...

EDIT: I tried to remove the symlink ScriptA.sh and commit, and add it back and commit again, but the uploaded file had not the symlink data but the regular file full data :(

PS.: I use Ubuntu/Nautilus with RabbitVCS Git.

Aquarius Power
  • 3,729
  • 5
  • 32
  • 67

1 Answers1

0

Use git gui (it is the normal git command with a gui parameter).
Its detection for symlinks was flawless!
You may need to install the package git-gui.

You have to click on Stage to Commit after selecting the files, then Commit and finally Push; and you will have to type the remote repository password on the terminal from where you run git gui, when asked.

Obs.: just a tip to make it have the same diff compare found at RabbitVCS. At "Add New Tool Command" on tools menu, put this simple command: meld $FILENAME

Ref.:
https://stackoverflow.com/questions/1516720/git-gui-client-for-linux
What's the difference between Git add (ctrl+A) and Stage to Commit (ctrl+T) in the git GUI?

Community
  • 1
  • 1
Aquarius Power
  • 3,729
  • 5
  • 32
  • 67