1.1 --- a/iixr/merging.py Sat Feb 12 01:23:58 2011 +0100
1.2 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000
1.3 @@ -1,89 +0,0 @@
1.4 -#!/usr/bin/env python
1.5 -
1.6 -"""
1.7 -Dictionary merging classes.
1.8 -
1.9 -Copyright (C) 2009, 2010 Paul Boddie <paul@boddie.org.uk>
1.10 -
1.11 -This program is free software; you can redistribute it and/or modify it under
1.12 -the terms of the GNU General Public License as published by the Free Software
1.13 -Foundation; either version 3 of the License, or (at your option) any later
1.14 -version.
1.15 -
1.16 -This program is distributed in the hope that it will be useful, but WITHOUT ANY
1.17 -WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
1.18 -PARTICULAR PURPOSE. See the GNU General Public License for more details.
1.19 -
1.20 -You should have received a copy of the GNU General Public License along
1.21 -with this program. If not, see <http://www.gnu.org/licenses/>.
1.22 -"""
1.23 -
1.24 -from itermerge import itermerge
1.25 -
1.26 -class Merger:
1.27 -
1.28 - "Merge files."
1.29 -
1.30 - def __init__(self, writer, readers):
1.31 - self.writer = writer
1.32 - self.readers = readers
1.33 -
1.34 - def close(self):
1.35 - for reader in self.readers:
1.36 - reader.close()
1.37 - self.readers = []
1.38 - if self.writer is not None:
1.39 - self.writer.close()
1.40 - self.writer = None
1.41 -
1.42 -class TermDictionaryMerger(Merger):
1.43 -
1.44 - "Merge term and position files."
1.45 -
1.46 - def merge(self):
1.47 -
1.48 - """
1.49 - Merge terms and positions from the readers, sending them to the writer.
1.50 - """
1.51 -
1.52 - last_term = None
1.53 - current_readers = []
1.54 -
1.55 - for term, frequency, doc_frequency, positions in itermerge(self.readers):
1.56 - if term == last_term:
1.57 - current_readers.append(positions)
1.58 - else:
1.59 - if current_readers:
1.60 - self.writer.write_term_positions(last_term, itermerge(current_readers))
1.61 - last_term = term
1.62 - current_readers = [positions]
1.63 - else:
1.64 - if current_readers:
1.65 - self.writer.write_term_positions(last_term, itermerge(current_readers))
1.66 -
1.67 -class FieldDictionaryMerger(Merger):
1.68 -
1.69 - "Merge field files."
1.70 -
1.71 - def merge(self):
1.72 -
1.73 - """
1.74 - Merge fields from the readers, sending them to the writer.
1.75 - """
1.76 -
1.77 - last_docnum = None
1.78 - current_fields = []
1.79 -
1.80 - for docnum, fields in itermerge(self.readers):
1.81 - if docnum == last_docnum:
1.82 - current_fields += fields
1.83 - else:
1.84 - if current_fields:
1.85 - self.writer.write_fields(last_docnum, current_fields)
1.86 - last_docnum = docnum
1.87 - current_fields = fields
1.88 - else:
1.89 - if current_fields:
1.90 - self.writer.write_fields(last_docnum, current_fields)
1.91 -
1.92 -# vim: tabstop=4 expandtab shiftwidth=4