checking tika-generated xhtml output from PDFs

slow-check.xq

(: 
 : for each xh:html document in the database, check for 1 occurence of $str1
 : and 2 occurences of $str2, returning the db:path (or name) of the document where 
 : true
 :)
declare namespace xh = "http://www.w3.org/1999/xhtml";

for $html in //xh:html
let $str1 := "(original signatures are on file with official student records"
let $str2 := "to the graduate council:"
let $sc1 := count($html//xh:div[@class='page']//xh:p/text()[contains(lower-case(.),$str1)])
let $sc2 := count($html//xh:div[@class='page']//xh:p/text()[contains(lower-case(.),$str2)])
where $sc1 = 1 and $sc2 = 2
return db:path($html)

not as fast as i would like, but hopefully good enough for govt. work.