CODEWITHSUNDEEP
regexbashsed
Qpa
Qpa

Reputation: 33

replace a certain string in a file - sed

The problem is there is a file with many strings like $VARx$ and x stands for a number from 1 to 9: $VAR1$, $VAR2$, $VAR3$, etc.

I want to quote these strings like $VARx$ → "$VARx$". I read or know that this could be possible with sed out of the command line but I have no idea how. Something like that but don't work:

sed -i "s/\\\$VAR1\$/\"\$VAR1\"\$/g" file.txt

The special part would be to include the variable x.

Upvotes: 0

Views: 76

Answers (2)

Idriss Neumann
Idriss Neumann

Reputation: 3838

Another alternative using sed :

sed -i 's/\(\$VAR[1-9]\$\)/"\1"/g' file.txt

This solution seems a bit closer to the PO's specification :

a file with many strings like $VARx$ and x stands for a number from 1 to 9.

=> [1-9] matches only one digit between 1 and 9.

In this example, I also wanted to show the use of capturing parentheses : \1 corresponds to the first expression captured by the capturing parentheses (\(...\)). Of course, & work as well for this case :

sed -i 's/\$VAR[1-9]\$/"&"/g' file.txt

Be careful: -i will do the substitution directly into the file instead of standard output.

Upvotes: 2

ooga
ooga

Reputation: 15501

Try

sed -i 's/\$VAR[[:digit:]]\+\$/"&"/g' file.txt

[[:digit:]]\+ matches 1 or more digits. & represents the matching string.

GNU sed manual: http://www.gnu.org/software/sed/manual/html_node/index.html

Upvotes: 4

Related Questions