Will use of Web Property Extractors be a good solution for this scenario?

Question

Hi all
We have a scenario where we would like to use our existing SAP Portal
environment along with the TREX to search in a a database with 150000+
cases much similar to SAP Customer Messages.
We are thinking of implementing the following scenario:
1) We develop a custom application that extracts cases from
our "customer message application", and converts them into html-files.
2) These 150000+ html-files are published on a web server that can be
seen by the SAP Portal (and corresponding TREX).
3) We integrate this "artificial" web site into our portal using a Web
Repository Manager with configured Web Property Extractors (content is published in meta tags).
4) We set up an index and configure search iviews etc, so the
supporters are able to search in different parts of the "customer messages" (=different properties).
My questions are now:
1) Can the Web Property Extractors be used in other repository managers
(like the CM repository manager, eg) or can you set up a Web Repository
Manager to look into a file server instead of a web server or will the above mentioned scenario be the only feasible solution?
2) What considerations should be done in relation to the above scenario in order to assure good search
performance and a well performing database? We have a bunch of text
snippets (that can be very long) that we wants to extract into
properties - thus being stored in the database. Will this be a problem?
Or will it not effect performance if the database is running on a SAN?
Best regards,
Martin Søgaard

Answer 1

Maybe something along the lines of this A Form to Design a Form sample would suit?