CODEWITHSUNDEEP
bashawksed
sundrys1
sundrys1

Reputation: 305

I need delete two " " with sed command

I need to delete "" in file 

"CITFFUSKD-E0"

I have tried sed 's/\"//.

Result is:

CITFFUSKD-E0"

How I can delete both ?

Also I need to delete everything behind first word but input can be this one:

"CITFFUSKD-E0"
"CITFFUSKD_E0"
"CITFFUSKD E0"

Result I want it:

CITFFUSKD

Upvotes: 0

Views: 115

Answers (7)

Claes Wikner
Claes Wikner

Reputation: 1517

awk '{gsub(/^.|....$/,"")}NR==1' file

CITFFUSKD

Upvotes: 0

Kyle Banerjee
Kyle Banerjee

Reputation: 2794

Many legitimate ways to solve this.

I favor using what you know about your data to simplify solutions -- this is usually an option. If everything in your file follows the same pattern, you can simply extract the first set of capitalized letters encountered:

sed 's/"\([A-Z]\+\).*$/\1/' file

Upvotes: 0

Walter A
Walter A

Reputation: 20022

When you have 1 line, you can use

grep -Eo "(\w)*" file | head -1

For normal files (starting with a double quote on each line) , try this

tr -c [^[:alnum:]] '"' < file | cut -d'"' -f2

Upvotes: 0

Wiktor Stribiżew
Wiktor Stribiżew

Reputation: 627190

You may use

sed 's/"//g' file | sed 's/[^[:alnum:]].*//' > newfile

Or, contract the two sed commands into one sed call as @Wiimm suggests:

sed 's/"//g;s/[^[:alnum:]].*//' file > newfile

If you want to replace inline, see sed edit file in place.

Explanation:

  • sed 's/"//g' file - removes all " chars from the file
  • sed 's/[^[:alnum:]].*//' > newfile - also removes all chars from a line starting from the first non-alphanumeric char and saves the result into a newfile.

Upvotes: 1

Jotne
Jotne

Reputation: 41460

This should do all in one go, remove the ", print the first part:

awk -F\" '{split($2,a,"-| |_");print a[1]}' file
CITFFUSKD
CITFFUSKD
CITFFUSKD

Upvotes: 0

KamilCuk
KamilCuk

Reputation: 141583

delete everything behind first word

sed 's/^"\([[:alpha:]]*\)[^[:alpha:]]*.*/\1/'

Match the first ". Then match a sequence of alphabetic characters. Match until you find non-alphabetic character ^[:alpha:]. Then match the rest. Substitute it all for \1 - it is a backreference for the part inside \( ... \), ie. the first word.

I need delete two “ ” with sed command

Remove all possible ":

sed 's/"//g'

Extract the string between ":

sed 's/"\([^"]*\)"/\1/'

Remove everything except alphanumeric characters (numbers + a-z + a-Z, ie. [0-9a-zA-z]):

sed 's/[^[:alnum:]]//g'

Upvotes: 0

RavinderSingh13
RavinderSingh13

Reputation: 133680

Could you please try following.

awk 'match($0,/[a-zA-Z]+[^a-zA-Z]*/){val=substr($0,RSTART,RLENGTH);gsub(/[^a-zA-Z]+/,"",val);print val}' Input_file

Upvotes: 0

Related Questions