unixsolaris-10

convert multiple lines into single line in a file only after a specific pattern

I am trying to convert multiple lines into single line only after a specific pattern, in a text file on Solaris 10 machine.

file be like:-

<open>
   <bas>ABC1</bas>
   <stat>1</stat>
</open>
<open>
   <bas>ABC10</bas>
   <stat>1</stat>
</open>
<open>
   <bas>ABC11</bas>
   <stat>4</stat>
   <line>1234</line>
</open>
<new>
</new>

expected output:-

<open><bas>ABC1</bas><stat>1</stat></open>
<open><bas>ABC10</bas><stat>1</stat></open>
<open><bas>ABC11</bas><stat>4</stat><line>1234</line></open>
<new>
</new>

I tried with but didn't succeed

perl -pne 'chomp unless(/^<\/open>/);' file_name
Solution

  • Assuming the file original.txt contains:

    <open>
<bas>ABC1</bas>
<stat>1</stat>
</open>
<open>
<bas>ABC10</bas>
<stat>1</stat>
</open>
<open>
<bas>ABC11</bas>
<stat>4</stat>
<line>1234</line>
</open>
<new>
</new>

    And so.bash contains:

    #!/bin/bash

awk -f so.awk original.txt

    and file so.awk contains:

    BEGIN {
    inopen=0
}
/<open>/ {
    inopen=1
    printf "%s",$0
    next
}
/<\/open>/ {
    inopen=0
    print
    next
}
inopen==1 {
    gsub(" ", "")
    printf "%s",$0
    next
}
{ print }

    The output will be:

    <open><bas>ABC1</bas><stat>1</stat></open>
<open><bas>ABC10</bas><stat>1</stat></open>
<open><bas>ABC11</bas><stat>4</stat><line>1234</line></open>
<new>
</new>