How To Find And Copy Files To A Single Directory Recursively in Linux?

Using the 'find' Command

The 'find' command in Linux helps locate files and directories. Its basic syntax is:

find [path] [options] [expression]

Where:

• [path] is the directory to search in
• [options] change the search behavior
• [expression] sets the search criteria

You can search for files based on:

• Name: Use '-name' with the filename pattern
• Type: Use '-type' with 'f' for files or 'd' for directories
• Size: Use '-size' with the file size
• Modification time: Use '-mtime' with the number of days

For example, to find all .txt files in the current directory and its subdirectories:

find . -name "*.txt"

find . -name "*.txt" -not -path "./temp/*"

Combining 'find' with 'cp' for Copying

Piping commands lets you use the output of one command as input for another. By combining 'find' with 'cp', you can locate and copy files in one step.

To use 'find' and 'cp' together, use the '-exec' option with 'find':

find [path] [criteria] -exec cp {} [destination] \;

For example, to find all .jpg files and copy them to a 'pictures' folder:

find . -name "*.jpg" -exec cp {} ~/pictures \;

This command searches for .jpg files in the current directory and its subdirectories, then copies each file to the 'pictures' folder in the home directory.

Setting Up the Search Parameters

To start searching for files, define the search directory. This can be a specific path or the current directory ('.').

To search in the /home/user/documents directory:

find /home/user/documents

Next, set file matching criteria. Common criteria include:

• Name: Use '-name' with the filename pattern in quotes
• Type: Use '-type' with 'f' for files or 'd' for directories
• Size: Use '-size' with '+' for larger than or '-' for smaller than, followed by the size ('c' for bytes, 'k' for kilobytes, 'M' for megabytes)

To find all PDF files larger than 1MB:

find /home/user/documents -name "*.pdf" -type f -size +1M

Executing the Copy Operation

To copy files found by the 'find' command, use the '-exec' option with the 'cp' command. The basic structure is:

find [search_directory] [criteria] -exec cp {} [destination_directory] \;

'{}' represents each file found by 'find'.

To copy all .jpg files from /home/user/pictures to /backup/images while keeping the original directory structure, use:

find /home/user/pictures -name "*.jpg" -exec cp --parents {} /backup/images \;

The '--parents' option with 'cp' keeps the directory structure. This helps organize files and avoid name conflicts in the destination directory.

find /home/user/pictures -name "*.jpg" -print0 | xargs -0 cp -t /backup/images

find /home/user/documents -name "*.txt" -not -path "*/tmp/*" -exec cp {} /backup/texts \;

Handling Permissions and Ownership

When copying files, you may face permission issues. These can occur if you don't have access to read the source files or write to the destination directory. To avoid these problems, you might need to run the copy command with sudo privileges.

To keep the original file attributes, use the '-p' option with the 'cp' command. Here's an example:

find /source/directory -name "*.txt" -exec cp -p {} /destination/directory \;

This command copies all .txt files from the source directory to the destination directory, keeping their original permissions and ownership.

For more control over permissions, use the '--preserve' option with 'cp':

find /source/directory -name "*.txt" -exec cp --preserve=mode,ownership,timestamps {} /destination/directory \;

This command keeps the file mode, ownership, and timestamps of the original files.

Dealing with Duplicate File Names

When copying files to a new location, you might find files with the same names. By default, the 'cp' command will overwrite existing files in the destination directory.

To avoid overwriting files, use the '-n' option with 'cp':

find /source/directory -name "*.txt" -exec cp -n {} /destination/directory \;

This command will not overwrite existing files in the destination directory.

Another method is to add a suffix to copied files. Use the '--backup' option with 'cp':

find /source/directory -name "*.txt" -exec cp --backup=numbered {} /destination/directory \;

This command adds a number to the end of any duplicate filenames in the destination directory.

You can also use the 'rsync' command instead of 'cp' for more control over file copying:

find /source/directory -name "*.txt" -print0 | rsync -av --files-from=- --from0 /source/directory /destination/directory

This command uses 'rsync' to copy files, which can handle duplicates more flexibly and provides options for updating only changed files.

find /source/directory -name "*.txt" -print0 | cpio -pmd0 /destination/directory