consider this string:
file
This is some text
<#
Love this or that
#>
He is running like a rabbit
In order to, for example, grep for newline strings, id use:
cat file | grep -Pzo '(?s)<#.*?#>'
Which will output:
<#
Love this or that
#>
But what if i want to REMOVE the match? how can I do it?
You can do it with sed
sed '/<#/,/#>/d' file
It removing text between two specific strings.
<# up to the next one after it that contains #> (or the end of the file if not found), so it would also delete foo in foo <# and wouldn't work for <# foo #>
That -P (a GNU extension) is for Perl regexps actually, so:
<file perl -0777 -pe 's/<#.*?#>//sg'
If it doesn't need to be grep, you can do it with awk:
awk 'BEGIN {p=1}; $1 == "<#" {p=0}; $1 == "#>" {p=1; next}; p == 1 {print}' file
This will set an internal flag p to 1 at the beginning, set it to 0 once your "exclusion start" pattern is found, and sets it back to 1 when encountering the "exclusion end". It will then only print lines where p equals 1. The next statement in the rule for matching the end is necessary to prevent the end pattern from being printed.
file, then the command-line I posted prints out all lines except those between <# and #>. Could you specify if you get any error messages?
$1 == "<#" {p=1} !p; $1 == "#>" {p=0}
I don't know if this is elegant, I put it under consideration:
tr "\n" "%" < file | sed -e 's/<#%.*#>%//' | tr "%" "\n"
Replace new lines with %
tr "\n" "%"
Replace everthing between <#%...#>%:
sed -e 's/<#%.*#>%//'
Replace % with new lines:
tr "$" "\n"
Ouput:
This is some text
He is running like a rabbit
pcregrepavailable on your system?grep -P ...ispcregrep(afaik).sedandawkas well?