Hi,
I've played around with Zend_Search_Lucene and it's really great. I've implemented my own result highlighting (since the default is not as customizable as I'd like) by tokenizing the document content (as the built-in highlighter does) and matching those tokens (of which I know the position in the document) against the query terms. This works great except for phrase queries obviously. It seems that the phrase is still split up into individual tokens and there doesn't seem to be a way to determine if a query contained phrases etc. (of course I could've missed it, please correct me if I did). So if I search for the phrase "Zend Framework" all the individual words will be highlighted when it should actually be only the one phrase. Is there some way to achieve this or is this at least planned? I'm using the QueryParser to build the query and the TextNum Analyzer if that's relevant. Btw. If I use the builtin highlighting it only works if I convert the document content to ASCII/TRANSLIT, obviously because the Tokenizer uses this encoding, but it would be really nice to mention that in the highlighting section in the manual, it took me quite some time to figure that out. thx in advance for any help Stefan Oestreicher
