Monthly Archives: September 2016

Find and Replace All Occurrences of a File on Linux Server from Command Line

Today I was given the challenge of providing a client with a command line command that would allow my client to replace all occurrences of any file named this-file.php with a new file with the same name that resides in the server's root directory. Here is how I accomplished the challenge with a single line command from the Linux command prompt:

find . -name "this-file.php" -print -exec cp this-file.php {} \;

What we have there is basically two commands in one. First is the command to find all occurrences of this-file.php on the server and display the results. I have it displaying results so you can verify what files were replaced in the end. So, the command to just find all files named this-file.php and list them is:

find . -name "this-file.php" -print

Type the above command at your Linux command prompt and you'll get a list of paths found to files with the name this-file.php.  However, it should be noted that the above command is the command to use if you are in the base directory of your server, otherwise you will want to specify the root directory. In the above command, the '.' (period) means to search in current directory. So if you're not in the curent directory, use something more appropriate such as a / instead of a . so it forces it to search all directories above and including /. Try this:

find / -iname "this-file.php"

That is a safer command than the first because it doesn't matter what directory you call it from. Notice we also didn't use the -print parameter at the end this time. It is actually moot because it will print the results anyway. It was included in the first example to explain it's use in the find and replace command, but is not needed when find is used by itself.

The second part of the challenge was to replace all those files listed with a new version of the file stored in the root directory. Therefore, before we go any further, we need to change into the root directory or whatever directory you will use to store the new file that will be used to replace all of the other files with. So change directories with the cd command like this:

cd /

That changes you into the root directory where you can go ahead and upload your new file if you haven't done so already.

Next is the command to copy this-file.php and overwrite the old version of the file. As long as you use the same file name, this works out-of-the-box with Linux. The simple command to copy a file from the root directory to a website's sub-directory might look something like this:

cp this-file.php /var/www/amykukuck.com/public_html/wp-content/themes/oily-sites-wide/templates/this-file.php

Combining two Linux Commands into One Line

The final part of our challenge is to combine the two commands we have learned above into one simple command to find and replace all occurrence of the file named this-file.php with a new version of it stored in the root directory that has the same file name. Here are the two commands combined:

find . -name "this-file.php" -print -exec cp this-file.php {} \;

Notice the find command which ends with -print looks the same. Then the find command is followed by -exec and then the cp command. Finally the combined command ends with {} \;. That last part has two functions. The brackets, {}, are used to replace what is found by the find command each time it finds something. The \; part lets Linux know the command is done and it can stop. Please note that for this to work the way it is written here, you must change directories into the directory that contains the new file in which will replace all instances of the file with the same name elsewhere on your server. So, be sure to execute the combined command only after changing into the correct directory which contains your new file!