We all have files on our computers — documents, photos, source code, you name it. So many of them. Definitely more than I can remember. And if not challenging, it might be time consuming to find the right one you’re looking for. In this post, we’ll have a look at how to make sense of your files on the command line, and especially how to quickly find the ones you’re looking for.
Good news is there are few quite useful utilities in the Linux commandline designed specifically to look for files on your computer. We’ll have a look at three of those: ls, tree, and find.
ls
If you know where your files are, and you just need to list them or see information about them, ls is here for you.
Just running ls lists all visible files and directories in the current directory:
$ ls Documents Music Pictures Videos notes.txt
Adding the -l option shows basic information about the files. And together with the -h option you’ll see file sizes in a human-readable format:
$ ls -lh total 60K drwxr-xr-x 2 adam adam 4.0K Nov 2 13:07 Documents drwxr-xr-x 2 adam adam 4.0K Nov 2 13:07 Music drwxr-xr-x 2 adam adam 4.0K Nov 2 13:13 Pictures drwxr-xr-x 2 adam adam 4.0K Nov 2 13:07 Videos -rw-r--r-- 1 adam adam 43K Nov 2 13:12 notes.txt
Is can also search a specific place:
$ ls Pictures/ trees.png wallpaper.png
Or a specific file — even with just a part of the name:
$ ls *.txt notes.txt
Something missing? Looking for a hidden file? No problem, use the -a option:
$ ls -a . .bash_logout .bashrc Documents Pictures notes.txt .. .bash_profile .vimrc Music Videos
There are many other useful options for ls, and you can combine them together to achieve what you need. Learn about them by running:
$ man ls
tree
If you want to see, well, a tree structure of your files, tree is a good choice. It’s probably not installed by default which you can do yourself using the package manager DNF:
$ sudo dnf install tree
Running tree without any options or parameters shows the whole tree starting at the current directory. Just a warning, this output might be huge, because it will include all files and directories:
$ tree . |-- Documents | |-- notes.txt | |-- secret | | `-- christmas-presents.txt | `-- work | |-- project-abc | | |-- README.md | | |-- do-things.sh | | `-- project-notes.txt | `-- status-reports.txt |-- Music |-- Pictures | |-- trees.png | `-- wallpaper.png |-- Videos `-- notes.txt
If that’s too much, I can limit the number of levels it goes using the -L option followed by a number specifying the number of levels I want to see:
$ tree -L 2 . |-- Documents | |-- notes.txt | |-- secret | `-- work |-- Music |-- Pictures | |-- trees.png | `-- wallpaper.png |-- Videos `-- notes.txt
You can also display a tree of a specific path:
$ tree Documents/work/ Documents/work/ |-- project-abc | |-- README.md | |-- do-things.sh | `-- project-notes.txt `-- status-reports.txt
To browse and search a huge tree, you can use it together with less:
$ tree | less
Again, there are other options you can use with three, and you can combine them together for even more power. The manual page has them all:
$ man tree
find
And what about files that live somewhere in the unknown? Let’s find them!
In case you don’t have find on your system, you can install it using DNF:
$ sudo dnf install findutils
Running find without any options or parameters recursively lists all files and directories in the current directory.
$ find . ./Documents ./Documents/secret ./Documents/secret/christmas-presents.txt ./Documents/notes.txt ./Documents/work ./Documents/work/status-reports.txt ./Documents/work/project-abc ./Documents/work/project-abc/README.md ./Documents/work/project-abc/do-things.sh ./Documents/work/project-abc/project-notes.txt ./.bash_logout ./.bashrc ./Videos ./.bash_profile ./.vimrc ./Pictures ./Pictures/trees.png ./Pictures/wallpaper.png ./notes.txt ./Music
But the true power of find is that you can search by name:
$ find -name do-things.sh ./Documents/work/project-abc/do-things.sh
Or just a part of a name — like the file extension. Let’s find all .txt files:
$ find -name "*.txt" ./Documents/secret/christmas-presents.txt ./Documents/notes.txt ./Documents/work/status-reports.txt ./Documents/work/project-abc/project-notes.txt ./notes.txt
You can also look for files by size. That might be especially useful if you’re running out of space. Let’s list all files larger than 1 MB:
$ find -size +1M ./Pictures/trees.png ./Pictures/wallpaper.png
Searching a specific directory is also possible. Let’s say I want to find a file in my Documents directory, and I know it has the word “project” in its name:
$ find Documents -name "*project*" Documents/work/project-abc Documents/work/project-abc/project-notes.txt
Ah! That also showed the directory. One thing I can do is to limit the search query to files only:
$ find Documents -name "*project*" -type f Documents/work/project-abc/project-notes.txt
And again, find have many more options you can use, the man page might definitely help you:
$ man find
Ellis
If you want to find files by content:
Lukas
You forgot the most important tool in my opinion: locate. I quote from the Arch wiki:
locate is a common Unix tool for quickly finding files by name. It offers speed improvements over the find tool by searching a pre-constructed database file, rather than the filesystem directly. The downside of this approach is that changes made since the construction of the database file cannot be detected by locate. This problem can be minimised by scheduled database updates.
WLBI
That’s right. My most used tool is
locate
In combination with
updatedb
nolty
+1
Robert Brady
$ locate -b ‘\cat’
Florian
If you find yourself often in the position to find specific files by their name, index based tools might give you a significant edge over the usage of ‘find -name’. Take a look at locate/slocate.
Fred Weigel
How about the venerable locate? (updatedb/locate)
sudo dnf install locate
queries like
locate –all yellow ball
which returns all filenames that contain yellow and ball.
RM
Thanks for your comment. I knew about “locate” but did not know about the “all” switch!
Mark Orchard
Very useful, thanks for the information. I keep forgetting the “Tree” function.
Is there an option with find to automatically cd to a file or directory once located?
Fred Weigel
Look at the -exec action in the man page
Spike
I don’t think so. How could it know which to cd if there are multiple matches?!
You could write a function that uses the first result and cd into it, if it’s a directory,
Elliott S
You might want to try installing
. In Fedora by default, only the tab completion is enabled, so you would type cd ** then press Tab, and it will open a list of directories from which you can start typing to pick one by pressing Enter.
If you follow the instructions in the
in the package, you can enable keyboard shortcuts as well. Then you can just type cd and press Ctrl+T to open the completion list directly.
Nathan D
Find all text files in all subdirectories where the phrase foo is used:
[Editor’s note: fixed an unclosed double-quote-pair]
Ray McCaffity
There is also “which”
which bash
/usr/bin/bash
Bartosz
I recommend “silver searcher” (ag) for searching content of text files (code mainly), “fasd” for fast finding already used files/folders and completing other commands with it and fzf for great fuzzy searching.
Julian
When mentioning fzf one could also mention rip grep (rg) as an very fast grep alternative and fd instead of find. With these tools one then can also navigate to files and directories.
JdeH
For file content, I find searchodt.sh and searchods.sh very useful for Open/LibreOffice document and spreadsheet files, respectively.
For pdf files, pdfgrep is really good.
Szymon
Possible to find files by photo exif ? i.e. camera model ?
EraZor
Definately +1 for Julian’s tip of ripgrep. Ultra high performant, multithreaded search tool for grepping or finding files. Totally upped my searching experiance!
If you have not: Take a look at it right now!