samples/LuceneInAction/lia/meetlucene/Indexer.py

   1 # ====================================================================
   2 #   Licensed under the Apache License, Version 2.0 (the "License");
   3 #   you may not use this file except in compliance with the License.
   4 #   You may obtain a copy of the License at
   5 #
   6 #       http://www.apache.org/licenses/LICENSE-2.0
   7 #
   8 #   Unless required by applicable law or agreed to in writing, software
   9 #   distributed under the License is distributed on an "AS IS" BASIS,
  10 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  11 #   See the License for the specific language governing permissions and
  12 #   limitations under the License.
  13 # ====================================================================
  14
  15 import os
  16
  17 from time import time
  18 from datetime import timedelta
  19 from lucene import \
  20     IndexWriter, StandardAnalyzer, Document, Field, \
  21     InputStreamReader, FileInputStream, Version, SimpleFSDirectory, File
  22
  23
  24 class Indexer(object):
  25
  26     def main(cls, argv):
  27
  28         if len(argv) != 3:
  29             print "Usage: python Indexer.py <index dir> <data dir>"
  30
  31         else:
  32             indexDir = argv[1]
  33             dataDir = argv[2]
  34
  35             start = time()
  36             numIndexed = cls.index(indexDir, dataDir)
  37             duration = timedelta(seconds=time() - start)
  38
  39             print "Indexing %s files took %s" %(numIndexed, duration)
  40
  41     def index(cls, indexDir, dataDir):
  42
  43         if not (os.path.exists(dataDir) and os.path.isdir(dataDir)):
  44             raise IOError, "%s does not exist or is not a directory" %(dataDir)
  45
  46         dir = SimpleFSDirectory(File(indexDir))
  47         writer = IndexWriter(dir, StandardAnalyzer(Version.LUCENE_CURRENT),
  48                              True, IndexWriter.MaxFieldLength.LIMITED)
  49         writer.setUseCompoundFile(False)
  50
  51         cls.indexDirectory(writer, dataDir)
  52
  53         numIndexed = writer.numDocs()
  54         writer.optimize()
  55         writer.close()
  56         dir.close()
  57
  58         return numIndexed
  59
  60     def indexDirectory(cls, writer, dir):
  61
  62         for name in os.listdir(dir):
  63             path = os.path.join(dir, name)
  64             if os.path.isfile(path):
  65                 if path.endswith('.txt'):
  66                     cls.indexFile(writer, path)
  67             elif os.path.isdir(path):
  68                 cls.indexDirectory(writer, path)
  69
  70     def indexFile(cls, writer, path):
  71
  72         try:
  73             reader = InputStreamReader(FileInputStream(path), 'iso-8859-1')
  74         except IOError, e:
  75             print 'IOError while opening %s: %s' %(path, e)
  76         else:
  77             print 'Indexing', path
  78             doc = Document()
  79             doc.add(Field("contents", reader))
  80             doc.add(Field("path", os.path.abspath(path),
  81                           Field.Store.YES, Field.Index.NOT_ANALYZED))
  82             writer.addDocument(doc)
  83             reader.close()
  84
  85     main = classmethod(main)
  86     index = classmethod(index)
  87     indexDirectory = classmethod(indexDirectory)
  88     indexFile = classmethod(indexFile)
  89
  90
  91 if __name__ == "__main__":
  92     import sys
  93     Indexer.main(sys.argv)