Hi all,

What is the prevailing wisdom on how to detect the beginning of a document within a document class in this stateless world.  This seems like a basic concept, but I haven't been able to find an adequate answer so far.  I have a large file where each page has an ID on it, read by Zone OCR.  When the value changes that should indicate the first page of a new document.  So, the first page of the file is a new document and every page thereafter where the value changes is a new document.  I've experimented with the whole Token Collector concept, but haven't yet found a solution.  How should this work?