gistfile1.py

def indexFilesFromPath(dir, idxPath):
    initVM(CLASSPATH)
    writer = IndexWriter(idxPath, StandardAnalyzer(), True)
    count = 1
    for root, dirs, files in os.walk(dir):
        for file in files:
            #print "- "+str(root)+" "+file
            sfx = file[-3:]
            if sfx.lower() not in [ "mp3","mp4", "ogg", "flac"  ]:
                continue
            filepath = os.path.join(root, file)
            try:
                doc = Document()
                f = tagpy.FileRef(filepath)
                title = unicode(f.tag().title)
                artist = unicode(f.tag().artist)
                album = unicode(f.tag().album)
                genre = unicode(f.tag().genre)
                path = unicode(filepath)
                doc.add(Field("title", title, Field.Store.YES, Field.Index.TOKENIZED))
                doc.add(Field("artist", artist, Field.Store.YES, Field.Index.TOKENIZED))
                doc.add(Field("album",album, Field.Store.YES, Field.Index.TOKENIZED))
                doc.add(Field("genre",genre, Field.Store.YES, Field.Index.TOKENIZED))
                doc.add(Field("all", album + u" " + artist + u" " + title, Field.Store.YES, Field.Index.TOKENIZED))
                doc.add(Field("path", path, Field.Store.YES, Field.Index.UN_TOKENIZED))
                print "\r %5d files indexed"%count,  # (%s/%s/%s)..."%(count, f.tag().artist, f.tag().album. f.tag().title) 
                count+=1
                writer.addDocument(doc)
            except Exception, e:
                
                #print str(e)
                # print "+"
                continue
    
    print "Done %d"%count

    writer.optimize()
    writer.close()