Return to indexCDLI.py CVS log

Up to [Repository] / cdli

new indices

#!/usr/bin/env python

import sys, os, PyLucene, threading, time
import xmlrpclib

from datetime import datetime

"""
This class is loosely based on the Lucene (java implementation) demo class 
org.apache.lucene.demo.IndexFiles.  It will take a directory as an argument
and will index all of the files in that directory and downward recursively.
It will index on the file path, the file name and the file contents.  The
resulting Lucene index will be placed in the current directory and called
'index'.
"""

class Ticker(object):

    def __init__(self):
        self.tick = True

    def run(self):
        while self.tick:
            sys.stdout.write('.')
            sys.stdout.flush()
            time.sleep(1.0)

class IndexFiles(object):
    """Usage: python IndexFiles"""

    def __init__(self, storeDir, analyzer):

        if not os.path.exists(storeDir):
            os.mkdir(storeDir)
        store = PyLucene.FSDirectory.getDirectory(storeDir, True)
        writer = PyLucene.IndexWriter(store, analyzer, True)
        writer.setMaxFieldLength(1048576)
        self.indexDocs(writer)
        ticker = Ticker()
        print 'optimizing index',
        threading.Thread(target=ticker.run).start()
        writer.optimize()
        writer.close()
        ticker.tick = False
        print 'done'

    def indexDocs(self, writer):
        s=xmlrpclib.Server("http://127.0.0.1:8080/cdliRoot/cdli_main")
        pns=s.getAllPNumbers()
        for pn in pns:

            txt=s.getFile(pn)
            i=0
            for line in txt.split("\n"):
                doc = PyLucene.Document()
                doc.add(PyLucene.Field("name", pn,
                                       PyLucene.Field.Store.YES,
                                       PyLucene.Field.Index.UN_TOKENIZED))
              
                doc.add(PyLucene.Field("line", str(i),
                                       PyLucene.Field.Store.YES,
                                       PyLucene.Field.Index.UN_TOKENIZED))
              
                        
                doc.add(PyLucene.Field("contents", line,
                                       PyLucene.Field.Store.YES,
                                       PyLucene.Field.Index.TOKENIZED))
                
                writer.addDocument(doc)
                i+=1
                print "    %s"%i
            print "indexed %s"%pn

if __name__ == '__main__':
    
    print 'PyLucene', PyLucene.VERSION, 'Lucene', PyLucene.LUCENE_VERSION
    start = datetime.now()
    try:
        IndexFiles("index", PyLucene.StandardAnalyzer())
        end = datetime.now()
        print end - start
    except Exception, e:
        print "Failed: ", e