How can I "grep" for a filename instead of the contents of a file?

Question

grep is used to search within a file to see if any line matches a given regular expression. However, I have this situation - I want to write a regular expression that will match the filename itself (and not the contents of the file). I will run this from the system's root directory, to find all those files that match the regular expression.

For example, if I want to find all Visual Basic form files that start with an "f" and end with .frm, I'll use the regular expression -

   "f[[:alnum:]]*\.frm"

Can grep do this? If not, is there a utility that would let me do this?

The must be a lot of duplicates from the first nearly four years. — Peter Mortensen, Apr 24 '21 at 10:34
***Not*** a duplicate (controlling the output from grep (independent of the matching, etc.)): *[How can I use grep to show just filenames on Linux?](https://stackoverflow.com/questions/6637882/how-can-i-use-grep-to-show-just-filenames-on-linux)* — Peter Mortensen, Apr 24 '21 at 10:38
This question is a duplicate of [How can I recursively find all files in current and subfolders based on wildcard matching?](https://stackoverflow.com/q/5905054/11725753) — EvgenKo423, Jul 06 '21 at 17:43
@PeterMortensen the irony is by closing this and commenting you successfully put this into the first page of google results for this kind of question. — IT Alex, Aug 08 '22 at 12:03

score 158 · Accepted Answer · answered Apr 18 '12 at 15:02

158

You need to use find instead of grep in this case.

You can also use find in combination with grep or egrep:

$ find | grep "f[[:alnum:]]\.frm"

answered Apr 18 '12 at 15:02

Pablo Santa Cruz

176,835
32
241
292

24

This is unnecessary. Find has that functionality built-in with the `-regex` option. – whereswalden Aug 11 '14 at 21:05
9

it is unnecessary here, but grep supports perl regex whereas find does not. – Dan Oct 28 '17 at 20:39
@alpha_989, I said grep supports it, which it does: https://www.gnu.org/software/grep/manual/html_node/grep-Programs.html#grep-Programs point is piping to grep is only needed if you need pcre. whereswalden has the right answer – Dan Mar 29 '18 at 19:19
2

@RM: No, that still searches file contents. It just *prints* file names instead of printing the matching lines. – user2357112 Mar 31 '21 at 22:45

score 122 · Answer 2 · edited Mar 31 '21 at 22:46

Example

find <path> -name '*FileName*'

From manual:

find -name pattern

Base of file name (the path with the leading directories removed) matches shell pattern pattern. Because the leading directories are removed, the file names considered for a match with -name will never include a slash, so "-name a/b" will never match anything (you probably need to use -path instead). The metacharacters ("*", "?", and "[]") match a "." at the start of the base name (this is a change in find‐ utils-4.2.2; see section STANDARDS CONFORMANCE below). To ignore a directory and the files under it, use -prune; see an example in the description of -path. Braces are not recognised as being special, despite the fact that some shells including Bash imbue braces with a special meaning in shell patterns. The filename matching is performed with the use of the fnmatch(3) library function. Don't forget to enclose the pattern in quotes in order to protect it from expansion by the shell.

I found this answer to be the most useful. I added a couple of additional options: `find . -maxdepth 1 -name "*filename*" -print` — JSON C11, Apr 01 '19 at 07:14

score 58 · Answer 3 · answered Aug 11 '14 at 21:10

58

As Pablo said, you need to use find instead of grep, but there's no need to pipe find to grep. find has that functionality built in:

find . -regex 'f[[:alnum:]]\.frm'

find is a very powerful program for searching for files by name and supports searching by file type, depth limiting, combining different search terms with boolean operations, and executing arbitrary commands on found files. See the find man page for more information.

answered Aug 11 '14 at 21:10

whereswalden

4,819
3
27
41

1

I would add the very handy feature of find to search by date with `-mtime` and `-mmin` – Johnride Nov 03 '14 at 20:38
1

Also, -regex switch performs matches with the whole path. So to work the example needs to add .* at the regex beginning. – tumasgiu Nov 29 '16 at 12:28

score 23 · Answer 4 · edited Mar 31 '21 at 22:40

23

You can find the relative path of a file using tree. Just pipe the output to grep to filter down:

tree -f | grep filename

Here is a function you can put into your .bash_profile, .bashrc, .zshrc or other...

findfile(){ tree -f | grep $1; } # $1 = filename, -f is full path

edited Mar 31 '21 at 22:40

Peter Mortensen

30,738
21
105
131

answered Oct 26 '16 at 02:48

jasonleonhard

12,047
89
66

-f is optional. – jasonleonhard Feb 16 '19 at 23:22

score 22 · Answer 5 · edited Mar 31 '21 at 22:42

22

The easiest way is

find . | grep test

Here find will list all the files in the (.), i.e., the current directory, recursively.

And then it is just a simple grep. All the files which name has "test" will appear.

You can play with grep as per your requirement. Note: As the grep is a generic string classification. It can result in giving you not only file names. But if a path has a directory ('/xyz_test_123/other.txt') it would also be part of the result set.

edited Mar 31 '21 at 22:42

Peter Mortensen

30,738
21
105
131

answered Jun 02 '20 at 04:24

sachyy

532
4
8

1

This will find anything with "test" in the path, e.g. `./foo/bar/latest/fred.txt`. – Paul R Jun 05 '20 at 07:35
oh. That one i didn't think of. Thank you for mentioning. – sachyy Jun 06 '20 at 12:31

score 13 · Answer 6 · edited Apr 24 '21 at 10:29

13

find -iname "file_name"

Syntax:
find -type type_descriptor file_name_here

type_descriptor types:

f: regular file

d: directory

l: symbolic link

c: character devices

b: block devices

edited Apr 24 '21 at 10:29

Peter Mortensen

30,738
21
105
131

answered Nov 23 '15 at 16:50

Sireesh Yarlagadda

12,978
3
74
76

1

I cannot seem to use a filename following the `-type typedescriptor` alone. Is this a mistake? – Allan Nørgaard Jun 14 '16 at 06:15

score 6 · Answer 7 · edited Apr 24 '21 at 11:29

6

find . | grep KeywordToSearch

Here . means the current directory which is the value for the path parameter for the find command. It is piped to grep to search the keyword which should return all matching results.

Note: This is case sensitive. So for example fileName and FileName are not same.

edited Apr 24 '21 at 11:29

Peter Mortensen

30,738
21
105
131

answered Jan 28 '21 at 08:07

darth vader

501
8
11

score 4 · Answer 8 · answered Sep 12 '16 at 12:58

4

You can also do:

tree | grep filename

This pipes the output of the tree command to grep for a search. This will only tell you whether the file exists though.

answered Sep 12 '16 at 12:58

caylus

470
4
11

score -1 · Answer 9 · answered Feb 14 '19 at 15:28

-1

Also for multiple files.

tree /path/to/directory/ | grep -i "file1 \| file2 \| file3"

answered Feb 14 '19 at 15:28

DimiDak

4,820
2
26
32

score -2 · Answer 10 · answered Feb 06 '18 at 00:21

-2

No, grep works just fine for this:

 grep -rl "filename" [starting point]

 grep -rL "not in filename"

answered Feb 06 '18 at 00:21

R M

542
4
11

3

This searches file contents for "filename", then prints the names of files with matches. OP wants match file names, not file contents. – Matthew Jun 25 '18 at 21:07
3

Agreed, this doesn't search filenames, it just prints the names of the filenames for which it's found *content* matches. You've misunderstood both the `-l` and `-L` options considerably. – Hashim Aziz Sep 16 '18 at 18:11

How can I "grep" for a filename instead of the contents of a file?

10 Answers10