How can I get rid of all script blocks in html files? [duplicate]

Question

How can I remove all script blocks (including multi-line ones) from html files such as:

<script type="text/javascript">
var googletag = googletag || {};
googletag.cmd = googletag.cmd || [];
(function() {
var gads = document.createElement('script');
gads.async = true;
gads.type = 'text/javascript';
var useSSL = 'https:' == document.location.protocol;
gads.src = (useSSL ? 'https:' : 'http:') +
'//www.googletagservices.com/tag/js/gpt.js';
var node = document.getElementsByTagName('script')[0];
node.parentNode.insertBefore(gads, node);
})();
</script>

I tried things like this with no success:

sed -i -e 's/<script.*\n.*<\/script>//g' 'path/to/file.html'

Answer 1

Sed processes the input line by line. It's easier in Perl that can process the whole file at once:

perl -0777 -pe 's=<script>.*?\n.*?</script>==sg'

• -0777 reads the whole file
• ? after * makes it "frugal", i.e. it matches the shortest possible string.
• /s makes . match a newline which it normally doesn't.

Note that it can break if the script contains </script> in a comment or quotes. It would be better to parse the HTML.

Answer 2

With sed you can select ranges and delete them:

sed '/<script/,/<\/script>/d' inputfile