Hi, Special URL tokenization if often a good thing to do with URLs. For example, over on Simpy.com you can do searches like this:
http://www.simpy.com/links/site/techcrunch.com http://www.simpy.com/links/site/www.techcrunch.com If you were to examine results, you'd see that the two result sets mostly overlap. That is because a "site" field contains both tokens. Often people reverse the domain or hostname, so then can do searches such as com* and they do more sophisticated URL tokenizing. Anyhow, for your needs you could also try something simple as http://domain/articles/2008/* to find docs with that URL prefix. Otis -- Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch ----- Original Message ---- > From: JLIST <[EMAIL PROTECTED]> > To: solr-user@lucene.apache.org > Sent: Saturday, May 3, 2008 9:59:47 PM > Subject: startsWith? > > Hi, I wonder it's possible search for text/string fields that starts > with a substring, similar to Java's startsWith function? For example, > if I have a URL indexed as text or string field, can I find URLs that > starts with "http://domain/articles/2008/"; ? > > If not, what's the best way to implement a query like this? By > splitting up the URL into sections and index them incrementally > like below? > > http://domain/ > http://domain/articles/ > http://domain/articles/2008/ > ... > >
