Filter only displayed HTML content

[Matthew Lynch]

I have found e2guardian to be a very useful tool.  However, it tends to overblock.  One of the reasons may be because it is filtering all HTML content, including javascript.  For example: 

A benign piano website was blocked because of the search terms "xxx" and "p0rn", which turned out to be contained within an image specified in base64 encoding.

Is there a way to make e2guardian scan only displayed content, and overlook content within HTML tags?  For example, it would detect 'xxx' in

<p>xxx</p>

but not 

<p id="xxx">Displayed content</p>

[Matthew Lynch]

I'm sorry.  I may have answered my own question.  It appears if I set "phrasefiltermode" to 1 it might do what I am describing above.