UNIX & Linux Grep

gistfile1.txt

Grep searches strings/patterns inside other strings/text.

Basic usage:
grep some-string some-file

-> This will return all lines of the file named 'some-file' which contains the string 'some-string'.

ls | grep some-string

-> This will return all files which contains the string 'some-string' in their names

Wildcards:

The . wildcard can be used to specify that any character (just one) will match the searched string if everything else match.

grep 'function .e' some-file

-> This will return all lines from the file 'some-file' which contains a string with the form 'function .e'. Match examples:
function revisions {
function delete {

The * wildcard can be used to specify that the character before it can be 0 or more times repeated. Examples:
grep 'h.*h' 'some-file'

-> This will match any string which contains a substring which has an 'h' followed by 0 or more character of any kind, and then and 'h'. Match examples:
hash
the boy is going to the town (Match on: he boy ... to th)

a* matches "", "a", "aa", ...
aa* matches "a", "aa", "aaa", ...

To escape a special character (grep won't interpret its special meaning) the \ character must be used before the target character.

To find all lines which contains the string 'image.gif', the dot character must be escaped:
grep 'image\.gif' 'some-file'

If we have the following list of files:
bower.json
dummyjsfile
gulpfile.js
karma.conf.js
package.json
protractor.conf.js

Executing ls | grep '.js' will return all files, because in this case, the dot is a wildcard that will match any character. If we want to match the dot in the file name, we must escape the dot in the grep statement in this way:

ls | grep '\.js' <- this will return all files which contains the '.js' string as part of their name.

grep 'somefile.ext' will match files like these:
somefile-ext
somefile.ext
somefileaext
somefilebext

The ? character match any string in which the character before an escaped ? is present one or zero times. For example:
grep 'abc\?' <- will match 'abc' and 'ab'

ls | grep '\.jso\?' <- will match files which contains '.js' or '.jso'

An expression surrounded by escaped parentheses is treated by as a single character. So, in combination with the ? character, it is possible to specify optional substring.

grep 'abcd\(mm\)\?xyz' <- will match 'abcdmmxyz' and 'abcdxyz'

The [] can be used to match a character from a list.
grep [hH] <- will match the strings "hello" and "Hello"

grep [0-9] <- will match the strings "1", "2", "a3bc", but it won't match "a" or "house"

There are some alternative forms:
[[:alpha:]] is the same as [a-zA-Z]
[[:upper:]] is the same as [A-Z]
[[:lower:]] is the same as [a-z]
[[:digit:]] is the same as [0-9]
[[:alnum:]] is the same as [0-9a-zA-Z]
[[:space:]] matches any white space including tabs

The [] can be used to find not matches too. This can be done putting a ^ in the first position inside the [].
grep ([^x]*)a <- will match '()a', 'x(b)ax', '(dsakjheqw123123)a' but it won't match something like '(x)a' (anything with an x between the parentheses).

To match a specific number of repetitions of a pattern, the pattern must be followed by \{n\}, in which n is the number of repetitions.

grep '[[:digit:]]\{3\}' <- will match any string with at least 3 consecutive numbers
grep '\+[[:digit:]]\{3\}-[[:digit:]]\{5\}' <- will match strings like '... +562-12356 ...'

The $ character matches the end of the line. The ^ character matches the beginning of the line.

grep '^[[:alnum:]]\+@[[:alnum:]]\+\.[[:alpha:]]\+$' <- will match any string that has 1 or more characters before an @, and then 1 or more characters after the @, and that is ended by a . and any number of characters.

Match examples: 
  [email protected]
  [email protected]
  [email protected]
Don't match examples:
  @xac.com (empty string before @)
  [email protected] (empty string after @ and before .)
  as@1as. (empty string after .)
  [email protected] (not [[:alpha:]] character after .)
  
ll | grep '^[d].*[[:space:]]\+\.' <- files that start with . (hidden) and in which their permission string starts with a 'd' 
drwxr-xr-x   30 rcanepa  staff   1.0K Aug  5 17:41 .
drwxr-xr-x    7 rcanepa  staff   238B Feb 10 14:04 ..
drwxr-xr-x   15 rcanepa  staff   510B Aug  5 18:41 .git
drwxr-xr-x   10 rcanepa  staff   340B Feb 10 14:04 .idea
drwxr-xr-x    4 rcanepa  staff   136B Aug  3 16:22 .tmp

To match a string between two patterns, the escaped | character must be used.

ll | grep '\(\.js$\)\|\(\.sh$\)' <- will match files terminated by .js or .sh
-rwxr-xr-x    1 rcanepa  staff   512B Mar 10 19:22 deploy.sh
-rw-r--r--    1 rcanepa  staff   232B Feb 10 14:04 gulpfile.js
-rw-r--r--    1 rcanepa  staff   243B Feb 10 14:04 karma.conf.js
-rw-r--r--    1 rcanepa  staff   708B Feb 10 14:04 protractor.conf.js
-rwxr-xr-x    1 rcanepa  staff   769B Aug  5 17:00 stats.sh

ls | grep 'a.*\(\.js\|\.sh$\)$' <- match files names that contains an 'a' and end by '.js' or '.sh'
karma.conf.js
protractor.conf.js
stats.sh