Regular expression for finding class names in HTML

Question

I'd like to use grep to find out if/where an html class is used across a bunch of files. The regex pattern should find not only <p class="foo"> but also <p class="foo bar foo-bar">.

So far I'm able to find class="foo" with this example below, can't make it work with multiple classnames:

grep -Ern "class=\"result+(\"| )" *

Any suggestions? Thanks! Mike

Answer 1

Depends what metacharacters your grep supprts, try:

'class=\"([a-z]+ ?)+\"'

Answer 2

How about something like this:

grep -Erno 'class[ \t]*=[ \t]*"[^"]+"' *

That will also allow for more whitespace and should give you output similar to:

1:class="foo bar baz"
3:class = "haha"

To see all classes used, you can pipe output from the above into the following:

cut -f2 -d'"' | xargs | sort | uniq

Answer 3

Regular expressions are a pretty poor tool for parsing HTML. Try looking into simpleXML ( http://php.net/manual/en/book.simplexml.php ). Roll-your-own regEx on HTML is begging for trouble.

Answer 4

Don't do it. It will drive you insane: http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454

Instead, use a HTML parser. It's not hard.

EDIT: Here's an example in PowerShell

Get-ChildItem -Recurse *.html | where { 
    ([xml](Get-Content $_)).SelectNodes( '//*' ) | where { $_.GetAttribute( "class" ).Contains( "foo" ) } 
}