[SOLVED] Changing matched output of grep using sed/awk

Issue

I have a file with the following

           <li data-row="1" data-col="2" data-sizex="1" data-sizey="1" class="red">
            <article>
                                                            <h1>Some unique text</h1>
          
           <li data-row="1" data-col="3" data-sizex="1" data-sizey="1" class="red">
            <article>
                                                            <h1>More unique text</h1>

           <li data-row="1" data-col="3" data-sizex="1" data-sizey="1" class="red">
            <article>
                                                            <h1>Even more unique text</h1>

I need to be able change the colour value before a specific string i.e

search for "More unique text" and replace the colour red to orange so the output is

            <li data-row="1" data-col="3" data-sizex="1" data-sizey="1" class="orange">
             <article>
                                                            <h1>More unique text</h1>

and the other colour values are left as red.

I can get the colour in the line i want to change using

grep -B 2 "More unique text" file | sed -e '2,4d'

which returns

<li data-row="1" data-col="3" data-sizex="1" data-sizey="1" class="red">

But how do i change the value red to orange?

Solution

xml parser like xmlstarlet is better suited to parse xml/html. If your input strictly follows the rules mentioned in your question, you can use

perl -i -0777 -pe 's/class="\Kred(?=".*\n.*\n.*More unique text)/orange/'

• -i for inplace editing
• -0777 to slurp entire input as single string, so this isn't suited for large input files
• class="\K match class=" but don't include it as part of matched text
• red text to match
• (?=".*\n.*\n.*More unique text) match " and check if More unique text is present two lines ahead
  • (?=regexp) is a lookahead feature, won't be part of matched text

Answered By - Sundeep