Use find exec with sed to replace part of file name and copy it

Question

I have some settings file which need to go to appropriate location. eg. settings.location1, settings.location2, settings.location3 which need to go to the appropriate folders namely location1, location2 and location3 and be renamed as simple settings.

eg.

/some/path/settings.location1 -> /other/path/location1/settings

/some/path/settings.location2 -> /other/path/location2/settings

I came up with the following comand:

find /some/path/settings.* -exec cp {} /other/path/$(echo {} | sed 's/.*\.//')/settings \;

But for some reason the sed does not get executed. And the result is

cp: cannot create regular file `/other/path//some/path/settings.location1/settings': No such file or director

It seems that if I run them separately all commands get executed well, but not if I put in exec. What am I doing wrong?

Answer 1

to make this work you need to create the intermediate folder when they are not present in the file system.

You can try:

find /some/path/settings.* -exec mkdir -p /other/path/$(ls -m1 {})/ && cp {} /other/path/$(ls -m1 {})/settings \;

Answer 2

I solved this by writing a very simple shell script which modified its parameters and did the operation, then having find do an exec on that script, passing {} multiple times.

In terms of your action, I think it would be something like:

find /some/path/settings.* -exec /bin/bash -c 'cp "$0"              \
  /other/path/"${1:s^/some/path/settings.^^}"/settings' {} {} \;

N.B. I inserted a line break after "$1" for ease of reading, but this should all be on one line. And I added double quotes around paths in case there are spaces in your paths. You might not need that precaution.

The Bash man page says,

-c If the -c option is present, then commands are read from the first non-option argument command_string. If there are arguments after the command_string, the first argument is assigned to $0 and any remaining arguments are assigned to the positional parameters. The assignment to $0 sets the name of the shell, which is used in warning and error messages.

The contents of the -exec script are: the bash interpreter path, the -c option, a quoted argument, find's path {}, then find's path {} again.

The quoted argument contains the shell script. Bash sets the next two arguments to $0 and $1, then runs the shell script. The script uses history substitution ${1:s^/some/path/settings.^^} to modify the path from find to keep only the part you need.