Linux grep command examples

Linux grep command FAQ: Can you share some Linux/Unix grep command examples? Sure. Think of the Linux/Unix grep command as a "search" command for Unix and Linux system. (Many people wish it was named "search"). It's used to search for text strings and regular expressions within one or more files. I think it's easiest to learn something like this by showing a collection of examples, so let's take a look at some grep command examples.

1) A typical grep example - searching for a text string in one file

This first grep command example searches for all occurrences of the text string fred within the "/etc/passwd" file. It will find and print (on the screen) all of the lines in this file that contain the text string fred, including lines that contain usernames like "fred" - and also "alfred".

grep 'fred' /etc/passwd

In a simple grep example like this, the quotes around the string fred aren't necessary, but they are needed when you get into using regular expressions (search patterns).

[akamtikar@draco9 active]$ grep ORDERBOOK outorderbook.log | wc -l

280

[akamtikar@draco9 active]$ grep "ORDERBOOK" outorderbook.log | wc -l

280

[akamtikar@draco9 active]$ grep 'ORDERBOOK' outorderbook.log | wc -l

280

2) Linux grep command - searching for a string in multiple files

Our next Linux grep command example searches for all occurrences of the text string joe within all files of the current directory:

grep 'joe' *

The * wildcard matches all files in the current directory. (More accurately, it matches all files in the current directory that don't begin with a decimal character, and it will also match any subdirectories in the current directory, but that usually isn't much of a problem.) As another example, you can also use grep to search all files in the current directory that end in the file extension ".txt", as shown here:

grep 'joe' *.txt

[akamtikar@draco9 active]$ grep ORDERBOOK ../../../logs/ |wc -l

0

[akamtikar@draco9 active]$ grep ORDERBOOK ../../../../Nischal/ |wc -l

0

[akamtikar@draco9 active]$ grep ORDERBOOK ../../execproc/ |wc -l

0

[akamtikar@draco9 active]$ grep ORDERBOOK ../active/ |wc -l

0

[akamtikar@draco9 active]$ grep ORDERBOOK |wc -l

[akamtikar@draco9 active]$ grep ORDERBOOK |wc -l

[akamtikar@draco9 active]$ grep ORDERBOOK ./ |wc -l

0

[akamtikar@draco9 active]$ grep ORDERBOOK * |wc -l

302

[akamtikar@draco9 active]$ grep O.*.RBO* outorderbook.log | wc -l

280

[akamtikar@draco9 active]$ grep ORDERBO* outorderbook.log | wc -l

280

3) Case-insensitive file searching with the Unix grep command

To perform a case-insensitive search with the grep command, just add the -i option, like this:

grep -i score gettysburg-address.txt

This grep search example matches the string "score", whether it is uppercase, lowercase, or some mix of the two.

[akamtikar@draco9 active]$ grep oRDERBOOK outorderbook.log | wc -l

0

[akamtikar@draco9 active]$ grep -i oRDERBOOK outorderbook.log | wc -l

280

4) Reversing the meaning of a grep search

You can reverse the meaning of a Linux grep search with the -v option. For instance, to show all the lines of my /etc/passwd file that don't contain the string fred, I'd issue this grep command:

grep -v fred /etc/passwd

[akamtikar@draco9 active]$ grep -i -v oRDERBOOK outorderbook.log | wc -l

15843

[akamtikar@draco9 active]$ grep -i oRDERBOOK outorderbook.log | wc -l

280

[akamtikar@draco9 active]$ wc outorderbook.log -l

16123 outorderbook.log

5) Using grep in a Unix/Linux command pipeline

The grep command is often used in a Unix/Linux pipeline. For instance, to show all the Apache httpd processes running on my Linux system, I can use the grep command in a pipeline with the 'ps' command:

ps auxwww | grep httpd

This returns the following output:

root nobody nobody nobody nobody nobody nobody nobody nobody nobody root grep http

17937 21538 24481 26089 27842 27843 27911 28280 30404 31895 31939

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

14760

6880 ?

Ss S S S S S S S S S R+

Apr01 Apr03 Apr03 Apr03 Apr03 Apr03 Apr03 Apr03 Apr03 Apr03 Apr03

0:39 0:01 0:00 0:01 0:00 0:00 0:01 0:00 0:00 0:00 0:00

/usr/local/apache/bin/httpd -k start 24372 17108 ? 14760 6396 ? /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start 24144 16876 ? 24896 17636 ? 24192 16936 ? 23888 16648 ? 24664 17256 ? 24360 17112 ? 14760 1848 6296 ? 548 pts/0 /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start /usr/local/apache/bin/httpd -k start

(I deleted about half of the "httpd -k start" lines from that output manually to save a little space.) Similarly, here's how you can find all the Java processes running on your system using the ps and grep commands in a Unix pipeline:

ps auxwww | grep -i java

In this example I've piped the output of the ps auxwww command into my grep command. The grep command only prints the lines that have the string "java" in them; all other lines from the ps command are not printed. One way to find all the sub-directories in the current directory is to mix the Linux ls and grep commands together in a pipe, like this:

ls -al | grep '^d'

Here I'm using grep to list only those lines where the first character in the line is the letter d.

[akamtikar@draco9 Nischal]$ ls -la

total 104 drwxrwxrwx 20 akamtikar nse 4096 Nov 8 17:46 . drwxrwxrwx 7 root root 4096 Nov 15 18:34 ..

drwxr-x--- 14 akamtikar nse 4096 Oct 21 16:04 Batch_Data drwxr-x--- 17 akamtikar nse 4096 Nov 8 17:34 batches drwxrwxrwx 18 akamtikar nse 4096 Nov 3 19:05 Binary_Data drwxrwxrwx 2 akamtikar nse 4096 Oct 22 17:39 Build drwxr-x--- 7 akamtikar nse 4096 Nov 8 17:47 config drwxrwxrwx 6 akamtikar nse 4096 Oct 21 16:04 data drwxrwxrwx 4 akamtikar nse 4096 Nov 1 12:29 etc drwxrwxrwx 4 akamtikar nse 4096 Nov 16 17:48 FOTS -rwxrwxrwx 1 akamtikar nse 516 Oct 21 13:11 FOTS_CHANGE_TRADING_DATE_IN drwxr-x--- 2 akamtikar nse 4096 Oct 21 16:03 Login_Topicwise drwxr-x--- 11 akamtikar nse 4096 Nov 12 19:16 logs drwxr-x--- 7 akamtikar nse 4096 Oct 21 16:04 Publisher_Utility drwxr-x--- 7 akamtikar nse 4096 Oct 21 16:04 Regression_Utility_Data -rwxr-x--- 1 akamtikar nse 464 Oct 21 16:03 removeagentlogs.sh -rwxrwxrwx 1 akamtikar nse 369 Oct 21 18:41 removelogs.sh -rwxrwxrwx 1 akamtikar nse 91 Oct 11 15:26 removeshm.sh

drwxr-x--- 10 akamtikar nse 4096 Oct 21 18:43 safestores drwxr-x--- 4 akamtikar nse 4096 Nov 8 17:40 scripts -rwxr-x--- 1 akamtikar nse 516 Oct 21 16:03 Shutdown_BE -rwxr-x--- 1 akamtikar nse 516 Oct 21 16:03 STARTUP_IN_BE

drwxr-x--- 2 akamtikar nse 4096 Oct 21 16:09 UTIL_config drwxr-x--- 3 akamtikar nse 4096 Oct 21 16:03 VULCAN-DEV drwxr-x--- 4 akamtikar nse 4096 Oct 21 16:04 vulcanproc drwxr-x--- 8 akamtikar nse 4096 Nov 4 21:05 Working_config

[akamtikar@draco9 Nischal]$ ls -la | wc -l

27

[akamtikar@draco9 Nischal]$ ls -la | grep '^-'

-rwxrwxrwx 1 akamtikar nse 516 Oct 21 13:11 FOTS_CHANGE_TRADING_DATE_IN -rwxr-x--- 1 akamtikar nse 464 Oct 21 16:03 removeagentlogs.sh -rwxrwxrwx 1 akamtikar nse 369 Oct 21 18:41 removelogs.sh -rwxrwxrwx 1 akamtikar nse 91 Oct 11 15:26 removeshm.sh

-rwxr-x--- 1 akamtikar nse 516 Oct 21 16:03 Shutdown_BE -rwxr-x--- 1 akamtikar nse 516 Oct 21 16:03 STARTUP_IN_BE

[akamtikar@draco9 Nischal]$ ls -la | grep '^-' | wc -l

6

[akamtikar@draco9 Nischal]$ ls -la | grep '^d'

drwxrwxrwx 20 akamtikar nse 4096 Nov 8 17:46 . drwxrwxrwx 7 root root 4096 Nov 15 18:34 ..

drwxr-x--- 14 akamtikar nse 4096 Oct 21 16:04 Batch_Data drwxr-x--- 17 akamtikar nse 4096 Nov 8 17:34 batches drwxrwxrwx 18 akamtikar nse 4096 Nov 3 19:05 Binary_Data drwxrwxrwx 2 akamtikar nse 4096 Oct 22 17:39 Build drwxr-x--- 7 akamtikar nse 4096 Nov 8 17:47 config drwxrwxrwx 6 akamtikar nse 4096 Oct 21 16:04 data drwxrwxrwx 4 akamtikar nse 4096 Nov 1 12:29 etc drwxrwxrwx 4 akamtikar nse 4096 Nov 16 17:48 FOTS

drwxr-x--- 2 akamtikar nse 4096 Oct 21 16:03 Login_Topicwise drwxr-x--- 11 akamtikar nse 4096 Nov 12 19:16 logs drwxr-x--- 7 akamtikar nse 4096 Oct 21 16:04 Publisher_Utility drwxr-x--- 7 akamtikar nse 4096 Oct 21 16:04 Regression_Utility_Data drwxr-x--- 10 akamtikar nse 4096 Oct 21 18:43 safestores drwxr-x--- 4 akamtikar nse 4096 Nov 8 17:40 scripts drwxr-x--- 2 akamtikar nse 4096 Oct 21 16:09 UTIL_config drwxr-x--- 3 akamtikar nse 4096 Oct 21 16:03 VULCAN-DEV drwxr-x--- 4 akamtikar nse 4096 Oct 21 16:04 vulcanproc drwxr-x--- 8 akamtikar nse 4096 Nov 4 21:05 Working_config

[akamtikar@draco9 Nischal]$ ls -la | grep '^d'| wc -l

20

6) Using the Linux grep command to search for multiple patterns at one time (egrep)
You can use a different version of the grep command to search for multiple patterns at one time. To do this, just use the "egrep" command instead of grep, like this:

egrep 'score|nation|liberty|equal' gettysburg-address.txt

This Unix egrep command searches the file named gettysburg-address.txt for the four strings shown (score, nation, liberty, and equal). It returns any lines from the file that contain any of those words. I should also note that "egrep" stands for "extended grep", and as you can see, it lets you do things like searching for multiple patterns at one time.

[akamtikar@draco9 active]$ egrep '999.00|2010111266666686|1002.00' * |wc

-l 554

[akamtikar@draco9 active]$ egrep '999.00|2010111266666686|1002.00' * |

wc -l 554

[akamtikar@draco9 active]$ egrep ' 999.00 | 2010111266666686 | 1002.00 ' *

| wc -l 552

[akamtikar@draco9 active]$ egrep ' 999.00 | 2010111266666686 |1002.00 ' * |

wc -l 552

[akamtikar@draco9 active]$ egrep ' 999.00 | 2010111266666686| 1002.00 ' * |

wc -l 552

[akamtikar@draco9 active]$ egrep ' 999.00 |2010111266666686 | 1002.00 ' * |

wc -l 552

[akamtikar@draco9 active]$ egrep ' 999.00| 2010111266666686 | 1002.00 ' * |

wc -l 552

[akamtikar@draco9 active]$ egrep ' 999.00 | 2010111266666686 | 1002.00 ' *

| wc -l 552

[akamtikar@draco9 active]$ egrep '999.00 |2010111266666686 |1002.00 ' * |

wc -l 552

[akamtikar@draco9 active]$ egrep '999.00|2010111266666686|1002.00' * |

wc -l 554

[akamtikar@draco9 active]$ egrep '999.00|2010111266666686|1002.00' *|wc

-l 554
Some of the other options of grep are:

-v display those lines that do NOT match -n precede each matching line with the line number -c print only the total count of matched lines

[akamtikar@draco9 active]$ egrep -c '999.00|2010111266666686|1002.00' *

batchsrvc_0_1.log:0 batchsrvc_0.lfh:0 batchsrvc.metrics:0 closeoutsrvc_0_1.log:0 closeoutsrvc_0.lfh:0 closeoutsrvc.metrics:0 egrep.txt:277 execprocmain_1.log:0 execprocmain.lfh:0 outorderbook.log:276 procexecsrvc_0_1.log:0 procexecsrvc_0.lfh:0 procexecsrvc.metrics:0 procslsrvc_0_1.log:0 procslsrvc_0.lfh:0 procslsrvc.metrics:0 sysevents.log:1

[akamtikar@draco9 active]$ egrep -c '999.00|2010111266666686|1002.00' * |

wc -l 17

[akamtikar@draco9 publisher]$ egrep -a 'transCode = 2075|transCode =

9002|transCode = 9004|transCode = 2130' net1pubsrvc_0_1.log | wc -l 107

[akamtikar@draco9 publisher]$ egrep -a -n 'transCode = 2075|transCode =

9002|transCode = 9004|transCode = 2130' net1pubsrvc_0_1.log > net1_cancel.txt

7) Linux grep pattern matching and regular expressions (regex patterns)

Of course the Linux grep command is much more powerful than this, and can handle very powerful regular expressions (regex patterns). In a simple example, suppose you want to search for the strings "Foo" or "Goo" in all files in the current directory. That grep command would be:

grep '[FG]oo' * [akamtikar@draco9 publisher]$ cat names good food mood sood tood hood cood bood zood tood wood nood pood [akamtikar@draco9 publisher]$ grep names:food names:mood [akamtikar@draco9 publisher]$ grep names:food names:mood names:tood names:tood names:pood [akamtikar@draco9 publisher]$ grep -a '[gmpsf]ood' * '[mptf]ood' * '[mf]ood' *

pt2pubsrvc_0_1.log:Nov 12, 2010; 19:27:27:: DEBUG:: s.c[ 100]:: goodTillDate = 0

36::

26::printGeneralFields::src/fots/structures/orderentry/generalfield pt2pubsrvc_0_1.log:Nov 12, 2010; 19:27:27:: DEBUG:: s.c[ 100]:: goodTillDate = 0 36:: 36::

26::printGeneralFields::src/fots/structures/orderentry/generalfield pt2pubsrvc_0_1.log:Nov 12, 2010; 19:27:27:: DEBUG:: s.c[ 100]:: goodTillDate = 0

26::printGeneralFields::src/fots/structures/orderentry/generalfield names:food names:mood names:sood names:pood

If you want to search for a sequence of three integers with grep you might use a command like this:

grep '[0-9][0-9][0-9]' *
This next grep command searches for all occurrences of the text string fred within the "/etc/passwd" file, but also requires that the "f" in the name "fred" be in the first column of each record (that's what the caret character tells grep). Using this more-advanced search, a user named "alfred" would not be matched, because the letter "a" will be in the first column.

grep '^fred' /etc/passwd [akamtikar@draco9 publisher]$ grep '^cood' * names:cood [akamtikar@draco9 publisher]$ grep '^tood' * names:tood names:tood
Regular expressions can get much, much more complicated (and powerful) than this, so I'll just leave it here for now.

[akamtikar@draco9 FOTS]$ egrep '[0-9][0-9][0-9]*' egrep

123nischal 231nischal 331nischal 431nischal 531nischal 631nischal 731nischal 132nischal 123nischal 123nischal 431nischal 555nischal

[akamtikar@draco9 FOTS]$ egrep '[1][0-9][0-9]*' egrep

123nischal 132nischal 123nischal 123nischal

[akamtikar@draco9 FOTS]$ egrep '[1][0-9][0-9]' egrep

123nischal 132nischal 123nischal 123nischal

[akamtikar@draco9 publisher]$ grep -ai 'transcode = 9002'

net1pubsrvc_0_1.log | wc -l 25

[akamtikar@draco9 publisher]$ grep -ai 'transcode = 9002'

passivemkt1pubsrvc_0_1.log | wc -l 30

[akamtikar@draco9 publisher]$ grep -ai 'transcode = 9002' * | wc -l

55 [akamtikar@draco9 publisher]$ grep -a -i 'transcode = 9002' * | wc -l 55 [akamtikar@draco9 publisher]$ grep -a -i -c 'transcode = 9002' * cosoutmsg.log:0 cospubsrvc_0_1.log:0 cospubsrvc_0.lfh:0 cospubsrvc.metrics:0 dsnormaloutmsg.log:0 dsnormalpubsrvc_0_1.log:0 dsnormalpubsrvc_0.lfh:0 dsnormalpubsrvc.metrics:0 dspriorityoutmsg.log:0 dsprioritypubsrvc_0_1.log:0 dsprioritypubsrvc_0.lfh:0 dsprioritypubsrvc.metrics:0 net1outmsg.log:0 net1pubsrvc_0_1.log:25 net1pubsrvc_0.lfh:0 net1pubsrvc.metrics:0

net2outmsg.log:0 net2pubsrvc_0_1.log:0 net2pubsrvc_0.lfh:0 net2pubsrvc.metrics:0 passivemkt1pubsrvc_0_1.log:30 passivemkt1pubsrvc_0.lfh:0 passivemkt1pubsrvc.metrics:0 postprocpubhbsrvc_0_1.log:0 postprocpubhbsrvc_0.lfh:0 postprocpubhbsrvc.metrics:0 pt2outmsg.log:0 pt2pubsrvc_0_1.log:0 pt2pubsrvc_0.lfh:0 pt2pubsrvc.metrics:0 [akamtikar@draco9 publisher]$ grep -aic 'transcode = 9002' * cosoutmsg.log:0 cospubsrvc_0_1.log:0 cospubsrvc_0.lfh:0 cospubsrvc.metrics:0 dsnormaloutmsg.log:0 dsnormalpubsrvc_0_1.log:0 dsnormalpubsrvc_0.lfh:0 dsnormalpubsrvc.metrics:0 dspriorityoutmsg.log:0 dsprioritypubsrvc_0_1.log:0 dsprioritypubsrvc_0.lfh:0

dsprioritypubsrvc.metrics:0 net1outmsg.log:0 net1pubsrvc_0_1.log:25 net1pubsrvc_0.lfh:0 net1pubsrvc.metrics:0 net2outmsg.log:0 net2pubsrvc_0_1.log:0 net2pubsrvc_0.lfh:0 net2pubsrvc.metrics:0 passivemkt1pubsrvc_0_1.log:30 passivemkt1pubsrvc_0.lfh:0 passivemkt1pubsrvc.metrics:0 postprocpubhbsrvc_0_1.log:0 postprocpubhbsrvc_0.lfh:0 postprocpubhbsrvc.metrics:0 pt2outmsg.log:0 pt2pubsrvc_0_1.log:0 pt2pubsrvc_0.lfh:0 pt2pubsrvc.metrics:0 [akamtikar@draco9 publisher]$ egrep -ic 'transcode = 9002|transcode = 9004| transcode = 2070|transcode = 2106' * cosoutmsg.log:0 cospubsrvc_0_1.log:0 cospubsrvc_0.lfh:0 cospubsrvc.metrics:0 dsnormaloutmsg.log:0 dsnormalpubsrvc_0_1.log:0

dsnormalpubsrvc_0.lfh:0 dsnormalpubsrvc.metrics:0 dspriorityoutmsg.log:0 dsprioritypubsrvc_0_1.log:0 dsprioritypubsrvc_0.lfh:0 dsprioritypubsrvc.metrics:0 net1outmsg.log:0 net1pubsrvc_0_1.log:55 net1pubsrvc_0.lfh:0 net1pubsrvc.metrics:0 net2outmsg.log:0 net2pubsrvc_0_1.log:0 net2pubsrvc_0.lfh:0 net2pubsrvc.metrics:0 passivemkt1pubsrvc_0_1.log:117 passivemkt1pubsrvc_0.lfh:0 passivemkt1pubsrvc.metrics:0 postprocpubhbsrvc_0_1.log:0 postprocpubhbsrvc_0.lfh:0 postprocpubhbsrvc.metrics:0 pt2outmsg.log:0 pt2pubsrvc_0_1.log:0 pt2pubsrvc_0.lfh:0 pt2pubsrvc.metrics:0

8) Display only filenames with a grep search

If you're looking through a lot of files for a pattern, and you just want to find the names of the files that contain your pattern (or "patterns", as shown with egrep) -- but don't want to see each individual grep pattern match -just add the "-l" (lowercase letter L) to your grep command, like this:

grep -l StartInterval *.plist

This command doesn't show every line in every file that contains the string "StartInterval"; it just shows the names of all the files that contain this string, like this:

com.apple.atrun.plist com.apple.backupd-auto.plist com.apple.dashboard.advisory.fetch.plist com.apple.locationd.plist org.amavis.amavisd_cleanup.plist

Of course you can also combine grep command arguments, so if you didn't happen to know how to capitalize StartInterval in that previous example, you could just add the "-i" argument to ignore case, like this:

grep -il startinterval *.plist

and that would have worked just fine as well, returning the same results as the previous grep command example.

[akamtikar@draco9 publisher]$ egrep -il 'transcode = 9002|transcode = 9004| transcode = 2070|transcode = 2106' * net1pubsrvc_0_1.log passivemkt1pubsrvc_0_1.log

[akamtikar@draco9 publisher]$ grep 'let' ngrepfilename Linux grep command FAQ: Can you share let's some Linux/Unix grep command examples? Think of the Linux/Unix grep let's command as a "search" command for Unix and Linux system Many people wish it was named "search" let's

It's used to search for text strings let's and regular expressions within one or more files I think it's easiest to learn let's something like this by showing a collection of examples so let's take a look at some grep command examples.

[akamtikar@draco9 publisher]$ grep -n 'let' ngrepfilename 1:Linux grep command FAQ: Can you share let's some Linux/Unix grep command examples? 2:Think of the Linux/Unix grep let's command as a "search" command for Unix and Linux system 3:Many people wish it was named "search" let's 4:It's used to search for text strings let's and regular expressions within one or more files 5:I think it's easiest to learn let's something like this by showing a collection of examples 6:so let's take a look at some grep command examples.

10) grep before/after - Showing lines before or after your grep pattern match
After a recent comment, I just learned that you can display lines before or after your grep pattern match, which is also very cool. To display five lines before the phrase "the living" in my sample document, use the -B argument, like this:

grep -B 5 "the living" gettysburg-address.txt

This grep command example returns this output:

The world will little note, nor long remember, what we say here, but can never forget what they did here.

It is for us, the living,

Similarly, to show the five lines after that same search phrase, use the -A argument with your Unix grep command, like this:

grep -A 5 "the living" gettysburg-address.txt

This grep "after" command returns the following output:

It is for us, the living, rather to be dedicated here to the unfinished work which they have, thus far, so nobly carried on. It is rather for us to be here dedicated to the great task remaining before us Of course you can use any number after the -A and -B options, I'm just using the number five here as an example. [akamtikar@draco9 publisher]$ grep -B 5 "look" ngrep Linux grep command FAQ: Can you share let's some Linux/Unix grep command examples? Think of the Linux/Unix grep let's command as a "search" command for Unix and Linux system Many people wish it was named "search" let's It's used to search for text strings let's and regular expressions within one or more files I think it's easiest to learn let's something like this by showing a collection of examples so let's take a look at some grep command examples. [akamtikar@draco9 publisher]$ grep -A 5 "FAQ" ngrep Linux grep command FAQ: Can you share let's some Linux/Unix grep command examples? Think of the Linux/Unix grep let's command as a "search" command for Unix and Linux system Many people wish it was named "search" let's

It's used to search for text strings let's and regular expressions within one or more files I think it's easiest to learn let's something like this by showing a collection of examples so let's take a look at some grep command examples.

11) Power file searching with find and grep

A lot of times I know that the string "foo" exists in a file somewhere in my directory tree, but I can't remember where. In those cases I roll out a power command, a Linux find command that uses grep to search what it finds:

find . -type f -exec grep -il 'foo' {} \;

This is a special way of mixing the Linux find and grep commands together to search every file in every subdirectory of my current location. It searches for the string "foo" in every file below the current directory, in a case-insensitive manner. This find/grep command can be broken down like this:

"." means "look in the current directory" "-type f" means "look in files only" "-exec grep -il foo" means "search for the string 'foo' in a case-insensitive manner, and return the matching line and filename when a match is found "{} \;" is some bizarre find syntax that you need to add to the end of your find command whenever you add the -exec option. (Sorry for my opinion there ... but you have to agree, that syntax is a little unusual.)

Note that on Mac OS X systems you may be able to use the mdfind command instead of this find/grep combination command. The mdfind command is a command-line equivalent of the Spotlight search functionality.

Related Unix/Linux grep commands and tutorials

We hope you enjoyed this Linux grep command tutorial and our grep examples. There are at least two other commands related to grep that you should at least be aware of. The fgrep command stands for "fast grep", or "fixed strings", depending on who you talk to. The egrep command stands for "extended grep", and lets you use even more powerful regular expressions. The strings command is good at finding printable strings in a binary file.

The locate command is more related to the find command, but I thought I would note that it is good at finding files in the entire filesystem when you know the filename, or part of the filename. And as I mentioned in the previous section Mac OS X systems have the mdfind command. As a practical matter I use plain old grep 99% of the time.

11) Power file searching with find and grep

A lot of times I know that the string "foo" exists in a file somewhere in my directory tree, but I can't remember where. In those cases I roll out a power command, a Linux find command that uses grep to search what it finds:

find . -type f -exec grep -il 'foo' {} \;

This is a special way of mixing the Linux find and grep commands together to search every file in every subdirectory of my current location. It searches for the string "foo" in every file below the current directory, in a case-insensitive manner. This find/grep command can be broken down like this:

"." means "look in the current directory" "-type f" means "look in files only" "-exec grep -il foo" means "search for the string 'foo' in a case-insensitive manner, and return the matching line and filename when a match is found "{} \;" is some bizarre find syntax that you need to add to the end of your find command whenever you add the -exec option. (Sorry for my opinion there ... but you have to agree, that syntax is a little unusual.)

Note that on Mac OS X systems you may be able to use the mdfind command instead of this find/grep combination command. The mdfind command is a command-line equivalent of the Spotlight search functionality.

Related Unix/Linux grep commands and tutorials

We hope you enjoyed this Linux grep command tutorial and our grep examples. There are at least two other commands related to grep that you should at least be aware of. The fgrep command stands for "fast grep", or "fixed strings", depending on who you talk to. The egrep command stands for "extended grep", and lets you use even more powerful regular expressions. The strings command is good at finding printable strings in a binary file.

The locate command is more related to the find command, but I thought I would note that it is good at finding files in the entire filesystem when you know the filename, or part of the filename. And as I mentioned in the previous section Mac OS X systems have the mdfind command. As a practical matter I use plain old grep 99% of the time.

[akamtikar@draco9 publisher]$ find /safestore3/Nischal/ -type f -exec grep -il "NavinNischal" {} \; /safestore3/Nischal/logs/postproc/active/publisher/ngrep