Filtering URLs in CrawlDB

[Dennis Kubes]

If I wrote a new normalizer and added some regex filters to filter out 
urls in crawldb and then I ran mergedb with a single db to filter and 
then ran mergesegs with a single segment to filter does anyone know if I 
would then be required to run through a re-parse?

Reason I am asking is because I went through this process without a 
re-parse and upon indexing I get blank index files.  So what I was 
thinking is that urls weren't matching up because they were now normalized.

Dennis