 |
 |
On May 31, 2006, at 12:00 PM, Kevin Kierans wrote: > Re: ISBN going to 13 characters > Anyone willing to share a (preferably) perl 13 to 10 character > and a 10 to 13 character conversion routine? > Or point me to one? There's some java code for 10 -> 13 listed on the front page of oss4lib.org right now, you could probably borrow from that. (I'm not really sure why the following is connected to this message, but as long as it is...) > -----Original Message----- > From: Code for Libraries [mailto:[log in to unmask]]On Behalf Of > Ross Singer > Sent: Tuesday, May 30, 2006 8:04 PM > To: [log in to unmask] > Subject: Re: [CODE4LIB] fun with kinosearch > > Pylucene seems to be left out of the party (I guess because instead of > being a port -- it's actually the Java classes used by Python? Is > that right?). PyLucene is indeed java lucene, compiled down to native code using gcj, and wrapped back up into python. Which makes it pretty much exactly like regular lucene, but faster, and with a little pythonic friendliness added for good measure. It has already been updated to match the release of lucene 2.0. > I'm not sure where stemming comes in (does Lucene do this?), You can apply the Porter stemming algorithm (or any other) as an additional Analyzer at index- and query-parsing time. It's straightforward to do and I think there are examples of exactly this in Lucene in Action (a highly recommended read for any lucene implementers). -Dan