On Sun, Dec 15, 2002 at 02:32:32PM +0100, [EMAIL PROTECTED] wrote: > The same search results for mittel without cs parameters does give > good results.
Yes, but my problems searching for words with umlauts. As I've written in my original post searching e.g. for "F�rdermitel" without cs= only looks for "F": http://www2.jugendpolitik.net/cgi-bin/jugendpolitik.net-search.cgi?q=F%F6rdermittel If I specify cs=iso-8859-1 it searches for the complete word, but displays other umlauts the wrong way e.g. "vielf�ltigen" in the first search-result (those are only "?" in the browser): http://www2.jugendpolitik.net/cgi-bin/jugendpolitik.net-search.cgi?q=F%F6rdermittel&cs=iso-8859-1 Searching with utf-8 instead does not give any results. Even worse, the search-field contains weird chars then: http://www2.jugendpolitik.net/cgi-bin/jugendpolitik.net-search.cgi?q=F%F6rdermittel&cs=utf-8 Please note, that the search-form itself adds cs=iso-8859-1... Also if I call the indexed pages manually using telnet to port 80, I get them with charset=iso-8859-1, so aspseek should index them as such, shouldn't it? Balu
