Feature Request - api to return character offsets of non-boilerplate text about boilerpipe HOT 3 CLOSED

Sorry, in step (1) above, it should read:
1) pass HTML to boilerpipe's HTML highlighter

Matching at term-level is out of scope for boilerpipe.

See Lucene's Highlighter class for a starting point:
http://lucene.apache.org/core/old_versioned_docs/versions/3_5_0/api/all/org/apac
he/lucene/search/highlight/Highlighter.html

Hi, if this is not a complete reject... :-)

I am not asking for matching at term level for boilerpipe. I am asking for 
character offsets (wrt the input text) in the non-boilerplate output returned 
by boilerpipe (ie step 2 in the original post).

So assuming an input text:
# 0                  1                  2                  3                  4 
                 5
# 012345678901234567890123456789012345678901234567890123456789
   BOILERPLATEsome good textMORE BOILERPLATE....

the output of boilerpipe is:
some good text

I am asking for a way to say that "some good text"  starts at position 10 and 
ends at position 21 in the original text. The rest I can do in my application.

I ask because I think this is already being done by the boilerpipe highlighter, 
so the information exists, but I couldn't figure out a way to get to it.

Thanks again,
Sujit