CODEWITHSUNDEEP
linuxubuntugrep
Raymosrunerx
Raymosrunerx

Reputation: 179

Grepping characters that don't fit a certain set of given characters

I was wondering how I could grep all the characters that don't fit a set of characters you want to give it.

For example, say I have a file that has the following text

XXXXX0000000HelloXXXXXX
MyXXXName0000Is00000000
XXXXXJamesXXX0X0X0X0XXX

I want it to print out "HelloMyNameIsJames" or those characters in some other fashion (if it comes out as one character per line, I would be fine with that).

Any help would be greatly appreciated! Raymosrunerx

Upvotes: 1

Views: 103

Answers (5)

Steve
Steve

Reputation: 54452

If you'd prefer one word per line, here's one way using GNU grep:

grep -oP '(?<=[0X]|\b)[^0X]+(?=[0X]|\b)' file.txt

Results:

Hello
My
Name
Is
James

You can use sed to neaten up your output:

< file.txt grep -oP '(?<=[0X]|\b)[^0X]+(?=[0X]|\b)' | sed ':loop; N; $!b loop; s/\n/ /g'

Results:

Hello My Name Is James

Upvotes: 1

Anand V
Anand V

Reputation: 124

If the lines are contained in file testdata, the following command will print a line each as shown:

sed -e 's/[0X]//g' testdata | xargs
Hello MyNameIs James

Hope this satisfies your needs!

Upvotes: 1

choroba
choroba

Reputation: 241938

Another solution, one character per line, using grep (if your grep supports the -o):

grep -o '[^X0]'

Upvotes: 1

Jon Lin
Jon Lin

Reputation: 143906

try:

sed -e 's/[X0]//g' input

grep isn't going to remove stuff for you.

Upvotes: 0

sehe
sehe

Reputation: 393239

tr -d X0 < input

Output:

Hello
MyNameIs
James

To remove line ends too:

tr -d 'X0\n' < input

Ouput

HelloMyNameIsJames

(no trailing line end)

Upvotes: 1

Related Questions