Maintain zip file structure

Question

I have a directory

main_folder
  first_folder
    second_folder
      empty.txt
      another_folder
        another_empty.txt

I want to zip second_folder, and I'm in main_folder, so I zipped with the command

zip -jr first_folder/zip_file first_folder/second_folder

The zip file contains

empty.txt
another_empty.txt

But I want

empty.txt
another_folder
  another_empty.txt

How can I have that?

Answer 1

The -j (junk paths) flag you set is killing what you want since it strips out the folder paths. That being said, however, I don't really see a way to just do second+. From the Man Page:

You may want to make a zip archive that contains the files in foo, without recording the directory name, foo. You can use the -j option to leave off the paths, as in:
zip -j foo foo/*

From what I can tell, that is just how it works in "zip" (nothing in the man suggests otherwise).

Edit: I was a bit after the other guy as I was trying to find out how to do what you want.

Why don't you just navigate to the directory and zip it? If you want to end up back where you were, this will do it (although a waste of time imo).

cd first_folder/second_folder/ && zip -r name ./ && mv name.zip ${here} && cd -

(Thanks to @KarthikT for the "cd -" shortcut).

Answer 2

From man page for zip

-j --junk-paths Store just the name of a saved file (junk the path), and do not store directory names. By default, zip will store the full path (relative to the current directory).

Just removing the -j should fix it.