= Find files with find

The `find` command is one of the most useful Linux commands, especially when faced with the hundreds and thousands of files and folders a modern computer contains.
As its name implies, `find` helps you find things, and not just by filename.
Whether you're on you're own computer or you're trying to support someone on an unfamiliar system, here are X things you can do with `find` to help you locate important data.

== 1. Find a single file by name

When you know the name of a file, but you can't remember where you saved it, use `find` to search your home directory.
Use `2>/dev/null` to silence permission errors (or use `sudo` to gain all permissions).

[source,bash]
----
$ find / -name "Foo.txt" 2>/dev/null
/home/seth/Documents/Foo.txt
----

== 2. Find a single file by approximate name

If you can't remember the exact name of the file, or you're not sure whether something was capitalized or not, you can do a partial and case-insensitive search:

[source,bash]
----
$ find / -iname "*foo*txt" 2>/dev/null
/home/seth/Documents/Foo.txt
/home/seth/Documents/foo.txt
/home/seth/Documents/foobar.txt
----

== 3. Find everything

An interesting alternative to `ls -R`:

[source,bash]
----
$ find ~/Documents -ls
3554235 0 drwxr-xr-x 2 seth seth 54 Sep 14 05:36 /home/seth/Documents/
3554224 0 -rw-rw-r-- 1 seth seth  0 Sep 14 05:36 /home/seth/Documents/Foo.txt
3766411 0 -rw-rw-r-- 1 seth seth  0 Sep 14 05:36 /home/seth/Documents/foo.txt
3766416 0 -rw-rw-r-- 1 seth seth  0 Sep 14 05:36 /home/seth/Documents/foobar.txt
----

Notice that I don't use `2>/dev/null` in this instance, because I'm only listing the contents of a https://redhat.com/sysadmin/absolute-path-and-relative-path-linux[file path] within my home directory, so I don't anticipate permission errors.

== 4. Find by content

A `find` command doesn't have to perform just one task.
In fact, there's an option in `find` that enables you to execute a different command on whatever results `find` returns.
This can be especially useful when you need to search for a file by *content* rather than by name, or you need to search by both.

[source,bash]
----
$ find ~/Documents/ -name "*txt" -exec grep -Hi penguin {} \;
/home/seth/Documents/Foo.txt:I like penguins.
/home/seth/Documents/foo.txt:Penguins are fun.
----

== 5. Find files by type

Using the `-type` option, you can find files, directories, symlinks, named pipes, sockets, and more.

[source,bash]
----
$ find ~ -type f
/home/seth/.bash_logout
/home/seth/.bash_profile
/home/seth/.bashrc
/home/seth/.emacs
/home/seth/.local/share/keyrings/login.keyring
/home/seth/.local/share/keyrings/user.keystore
/home/seth/.local/share/gnome-shell/gnome-overrides-migrated
[...]
----

As long as you're using the GNU version of `find`, you can include multiple file types in your search results:

[source,bash]
----
$ find ~ -type f,l -name "notebook*"
/home/seth/notebook.org
/home/seth/Documents/notebook-alias.org
----

== 6. List just directories

A shortcoming of the `ls` command is that you can't filter its results by file type, so `ls` can be noisey if you only want a listing of directories in a path.
The `find` command combined with the `-type d` option is a better choice:

[source,bash]
----
$ find ~/Public -type d
find ~/Public/ -type d
/home/seth/Public/
/home/seth/Public/example.com
/home/seth/Public/example.com/www
/home/seth/Public/example.com/www/img
/home/seth/Public/example.com/www/font
/home/seth/Public/example.com/www/style
----

== 7. Limit listing results

With hundreds of files in even a default user directory, and thousands more outside of that, sometimes you get more results from `find` than you want.
You can limit the depth of searches with the `-maxdepth` option, followed by the number of directories you want `find` to descend into after the starting point:

[source,bash]
----
$ find ~/Public/ -maxdepth 1 -type d
/home/seth/Public/
/home/seth/Public/example.com
----

== 8. Find empty files

Sometimes it's helpful to find empty files as a way to declutter:

[source,bash]
----
$ find ~ -type f -empty
random.idea.txt
----

Technically, you can use `find` to remove empty files, but programmatic removal of files is dangerous.
For instance, forget to include `-type f` in a search for empty _files_ and you get directories in your results, so by adding a delete flag you would remove potentially important directory structures.
It's vital to compose your `find` command and then verify the results before deleting.
Furthermore, a misplaced delete flag in `find` can delete results before qualifying them (in other words, you can delete directories in a command intended to delete only files by placing the delete flag before the type flag.)
I prefer to use `xargs` or http://LINK-TO-PARALLEL-ARTICLE[parallel] and a https://www.redhat.com/sysadmin/recover-file-deletion-linux#trashy[trash command] on the rare occasion that I remove files with `find`.

== 9. Find files by age

The `-mtime` option allows you to limit a search to files older than, but also files newer than, _some value_ times 24.

[source,bash]
----
$ find /var/log -iname "*~" -o -iname "*log*" -mtime +30
----

The `+` before the `-mtime` number doesn't mean to add that number to the time, it's a conditional statement that matches (in this example) a value _greater than_ 24 times 30.
In other words, the sample code finds log files that haven't been modified in a month or more.

To find log files modified within the past week, you can use the `-` conditional:

[source,bash]
----
$ find /var/log -iname "*~" -o -iname "*log*" -mtime -7
/var/log/tallylog
/var/log/cups/error_log
/var/log/cups/access_log
/var/log/cups/page_log
/var/log/anaconda/anaconda.log
/var/log/anaconda/syslog
/var/log/anaconda/X.log
[...]
----

You already know about the `-ls` flag, so you can combine that with these commands for clarity:

[source,bash]
----
$ find /var/log -iname "*~" -o -iname "*log*" -mtime -7 -ls
-rw-------  1 root root            0 Jun  9 18:20 /var/log/tallylog
-rw-------  1 root lp      332 Aug 11 15:05 /var/log/cups/error_log
-rw-------  1 root lp      332 Aug 11 15:05 /var/log/cups/access_log
-rw-------  1 root lp      332 Aug 11 15:05 /var/log/cups/page_log
-rw-------  1 root root  53733 Jun  9 18:24 /var/log/anaconda/anaconda.log
-rw-------  1 root root 835513 Jun  9 18:24 /var/log/anaconda/syslog
-rw-------  1 root root  21131 Jun  9 18:24 /var/log/anaconda/X.log
[...]
----

== 10. Searching a path

Sometimes you know the directory structure leading up to a file you need, you just don't know where the directory structure is located within the system.
To search within a path string, you can use the `-ipath` option, which treats dots and slashes not as regex characters, but as dots and slashes.

[source,bash]
----
$ find / -type d -name 'img' -ipath "*public_html/example.com*" 2>/dev/null
/home/tux/Public/public_html/example.com/font
----

== Found it

The `find` command is an essential tool for a sysadmin.
It's useful when investigating or getting to know a new system, for finding misplaced data, and for https://www.redhat.com/sysadmin/find-best-friend[troubleshooting] everyday problems.
But it's also just a convenience tool.
You don't need a "good" reason to use `find`.
If it makes a task easier to just search for something instead of traversing your system, then use `find`.
It's an understated but infinitely useful tool that embodies the sublime pleasure of everyday Linux.
Start using it today, and _find_ out what makes it great.

You are viewing proxied material from sdf.org. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.