1. The indexer scans the input directory and builds an
representing the directory structure and all the file
names it finds there. It also adds information about
newness -- i.e. which files that are new to the index,
which files appear to be updated, which files appear to
2. It passes this document to a new stylesheet, maybe
called "docSelector.xsl" or something like that.
3. That stylesheet picks and chooses from the files in
the directories, and based on their newness and/or
meta-data, outputs a flat list of which files to index.
For each, it specifies a prefilter and a display
4. Indexer works from that list to build the index.
Log in to post a comment.