cp | |
---|---|
![]() Example usage of cp command | |
Original author(s) | AT&T Bell Laboratories |
Developer(s) | Various open-source and commercial developers |
Initial release | November 3, 1971 |
Written in | Plan 9: C |
Operating system | Unix, Unix-like, Plan 9, Inferno, KolibriOS |
Platform | Cross-platform |
Type | Command |
License | coreutils: GPLv3 Plan 9: MIT License |
cp
is a shell command for copying files and directories.
If the user has write access to a target file, the command copies the content by opening it in update mode. This preserves the file's inode instead of creating a new file with default permissions.
The command was part of Version 1 Unix, [1] and is specified by POSIX. The implementation from GNU has many additional options beyond the POSIX specification. [2] The command is bundled in GNU Core Utilities [3] and is available in the EFI shell. [4]
-f
(force) – specifies removal of each target file if it cannot be opened for write operations; removal precedes any copying-H
(dereference) – follows symbolic links so that the destination has the target file rather than a link to the target-i
(interactive) – prompts user to overwrite each target file that clashes with a source file-n
(no clobbering) – prevents overwriting files-p
(preserve) – preserves metadata of each source file in the destination; including: time of last modification and last access, ownership, and file permissions-R
or -r
(recursive) – copy directories recursivelyThe command has three principal modes of operation as inferred from command-line arguments. [5]
For a path to an existing file followed by a path that does not refer to an existing directory, the file at the first path is copied to the second path.
cp [-fHip][--] sourcefile targetfile
For one or more paths to existing files followed by a path to an existing directory, the files are copied to the directory.
cp [-fHip] [--] sourcefile... targetdirectory
With the recurse command-line option, typically -r
, a path to an existing directory and a second path, the files of the directory are copied to the second path. If the second path refers to nothing, the source directory is copied to that path. If the second path refers to an existing directory, the source directory is copied into the destination directory as a subdirectory.
cp -r|-R [-fHip] [--] sourcedirectory... targetdirectory
This copies file prog.c to file prog.bak. If prog.bak does not already exist, this creates it. If it does exist, its content will be replaced.
cp prog.c prog.bak
This copies the files jones and smith into the pre-existing directory clients.
cp jones smith clients
This copies file smith to a file named smith.jr. Instead of creating a file with the current date and time stamp, the command copies the date and time from the original. The copy also receives other metadata from the original including access control protection.
cp -p smith smith.jr
This recursively copies the directory clients, including its files, subdirectories, and the files in those subdirectories, to a new directory customers/clients.
cp -R clients customers
Some implementations behave differently in recursive mode, depending on the termination of the directory path. Using cp -R clients/ customers
in the GNU implementation behaves as above. However, with a BSD implementation, it copies the contents of the clients directory, instead of the directory clients itself. The same happens in both GNU and BSD implementations if the path of the source directory ends in . or .. (with or without trailing slash).