[SOLVED] Place braces around multiples lines of <p> tags using sed, and stop when it reaches <h1>, <h2>, <h3>

Issue

I have an HTML file with <p>, <h1>, <h2>, and <h3> tags. I need to make the <p> tags to be in braces, and if they’re subsequent <p> tags on each line, the braces should begin at the first instance, and the closing brace at the closing </p> tag, even if it’s a few lines down.

Example content:

<p>Paragraph 1</p>
<p>Paragraph 2</p>
<p>Paragraph 3</p>
<h1>Heading 1</h1>
<p>Paragraph 4</p>
<h3>Heading 2</h3>

Desired output:

{<p>Paragraph 1</p>
<p>Paragraph 2</p>
<p>Paragraph 3</p>}
<h1>Heading 1</h1>
{<p>Paragraph 4</p>}
<h3>Heading 2</h3>

Note in the desired output that the closing brace is when the lines swap to an <h1>. The number of subsequent <p> tags can be anywhere from 2, to 20.

My current sed solution is just replacing opening <p> tags with an open brace, and closing </p> tags with a closing brace.

sed 's|<p>|{|g' | sed 's|</p>|}|g'

Unfortunately this does it line by line, and what I need is to match multiple lines, and ignore closing/opening tags if they're followed by another <p> tag so that they're lumped together.

I've been unable to find a solution for this yet. I'm happy to use perl, awk, sed, whatever gets the job done. It just seems like I need a way to recognise this particular pattern.

Edit: Ed Morton's solution below worked perfectly for me.

Solution

$ cat tst.awk
/^<p/ {
    ps = (ps == "" ? "" : ps ORS) $0
    next
}
ps != "" { print "{" ps "}"; ps="" }
{ print }
END { if (ps != "") print "{" ps "}" }

$ awk -f tst.awk file
{<p>Paragraph 1</p>
<p>Paragraph 2</p>
<p>Paragraph 3</p>}
<h1>Heading 1</h1>
{<p>Paragraph 4</p>}
<h3>Heading 2</h3>

Answered By - Ed Morton