Recursive text search with grep and file patterns

Question

Given this example folder structure:

/folder1/file1.txt
/folder1/file2.djd
/folder2/file3.txt
/folder2/file2.fha

How do I do a recursive text search on all *.txt files with grep from "/"?

("grep -r <pattern> *.txt" fails when run from "/", since there are no .txt files in that folder.)

score 63 · Accepted Answer · edited Feb 20 '18 at 13:35

63

My version of GNU Grep has a switch for this:

grep -R --include='*.txt' $Pattern

Described as follows:

--include=GLOB
Search only files whose base name matches GLOB (using wildcard matching as described under --exclude).

edited Feb 20 '18 at 13:35

Andrew Schulman

8,561
21
31
47

answered May 19 '09 at 12:01

Kyle Brandt

82,107
71
302
444

score 20 · Answer 2 · answered May 19 '09 at 12:02

20

If you have a large number of files it would be useful to incorporate xargs into the command to avoid an 'Argument list too long' error.

find . -name '*.txt' -print | xargs grep <pattern>

answered May 19 '09 at 12:02

Mark

754
1
7
12

6

If there are spaces in any of the file or directory names, use this form: find . -name '*.txt' -print0 | xargs -0 grep – Jason Luther May 19 '09 at 13:37
2

And of course there's the issue of filenames that start with `-`. – T.J. Crowder Feb 07 '11 at 12:08
`grep` is faster of `find`. – Nabi K.A.Z. Aug 16 '18 at 05:18

score 2 · Answer 3 · answered May 13 '16 at 07:22

2

you might be able to make use of your zsh's EXTENDED_GLOB option (docs)

grep <pattern> **/*.txt

answered May 13 '16 at 07:22

santa

121
2

1

Only if OP is using zsh, but interesting nonetheless. – Law29 May 14 '16 at 13:01

score 1 · Answer 4 · answered May 03 '10 at 16:54

1

You may want to take a look at ack at http://betterthangrep.com, which has facilities for selecting files to search by filetype.

answered May 03 '10 at 16:54

Andy Lester

740
5
16

score 0 · Answer 5 · answered Mar 26 '19 at 07:53

0

It's 2019 and there's no way I would still use grep for recursive text searching.

IMHO todays answers should include ripgrep:

rg <pattern> -ttxt

answered Mar 26 '19 at 07:53

santa

121
2

score 0 · Answer 6 · answered May 19 '09 at 11:43

0

find . -name '*.txt' -type f -exec grep <pattern> {} \;

answered May 19 '09 at 11:43

innaM

1,428
9
9

you might want to use "find . -name '*.txt' -type f -exec grep {} +" instead so that it rather behaves similiar the verision with from Mark Robinson - works only with GNU find to my knowledge – Martin M. Jun 10 '09 at 08:49

score 0 · Answer 7 · answered May 19 '09 at 11:46

Mannis answer would fork a new grep-process for every textfile. If you have lots of textfiles there, you might consider grepping every file first and pick the .txt-files when thats done:

grep -r <pattern> * | grep \.txt:

That's more disk-intensive, but might be faster anyway.

Recursive text search with grep and file patterns

7 Answers7

Linked