RE: How to index correctly a text save with tinyMCE

[Steven A Rowe]

Hi Ariel,

On 6/23/2011 at 12:34 PM, Ariel wrote:
> But it still doesn't convert the code to the correct character, for
> instance: España must be converted to España but it still
> remains as España.

So it looks like your text processing tool(s) escape markup meta-characters 
(e.g. "&" -> "&") after escaping above-ASCII characters to their named 
entity equivalents (e.g. "n" with a tilde to "ñ").  This two-level 
escaping appears to be the problem.

According to the analysis.jsp output you sent, your original text 
"España" was converted to "Espa&ndilde;a" - the first level of 
escaping was reversed.

I suspect you could fix the problem by including HTMLStripCharFilter twice, 
e.g.:

   <charFilter class="solr.HTMLStripCharFilterFactory"/>
   <charFilter class="solr.HTMLStripCharFilterFactory"/>
   <tokenizer class="solr.WhitespaceTokenizerFactory"/>
   ...

Good luck,
Steve