# HG changeset patch
# User Paul Boddie <paul@boddie.org.uk>
# Date 1252263220 -7200
# Node ID dec7d2ac1cc916ae976db663d70b391ab15c0c7b
# Parent  251e042dada73ea117a196d7db189f9766f23d3e
Made separate methods for sorted and unsorted position data.

diff -r 251e042dada7 -r dec7d2ac1cc9 iixr.py
--- a/iixr.py	Sun Sep 06 02:12:16 2009 +0200
+++ b/iixr.py	Sun Sep 06 20:53:40 2009 +0200
@@ -203,7 +203,7 @@
     def reset(self):
         self.last_docnum = 0
 
-    def write_positions(self, docnum, positions):
+    def write_sorted_positions(self, docnum, positions):
 
         """
         Write for the document 'docnum' the given 'positions'.
@@ -225,10 +225,6 @@
 
         self.write_number(len(positions))
 
-        # Make sure that the positions are sorted.
-
-        positions.sort()
-
         # Write the position deltas.
 
         last = 0
@@ -241,6 +237,18 @@
 
         return offset
 
+    def write_positions(self, docnum, positions):
+
+        """
+        Write for the document 'docnum' the given 'positions'.
+        Return the offset of the written record.
+        """
+
+        # Make sure that the positions are sorted.
+
+        positions.sort()
+        return self.write_sorted_positions(docnum, positions)
+
 class PositionReader(FileReader):
 
     "Reading position information from files."
@@ -433,7 +441,7 @@
         self.position_index_writer = position_index_writer
         self.interval = interval
 
-    def write_term_positions(self, doc_positions):
+    def write_sorted_term_positions(self, doc_positions):
 
         """
         Write all 'doc_positions' - a collection of tuples of the form (document
@@ -460,8 +468,6 @@
         first_offset = None
         count = 0
 
-        doc_positions.sort()
-
         for docnum, positions in doc_positions:
             pos_offset = self.position_writer.write_positions(docnum, positions)
 
@@ -506,6 +512,22 @@
 
         return index_offset, frequency, len(doc_positions)
 
+    def write_term_positions(self, doc_positions):
+
+        """
+        Write all 'doc_positions' - a collection of tuples of the form (document
+        number, position list) - to the file.
+
+        Add some records to the index, making dictionary entries.
+
+        Return a tuple containing the offset of the written data, the frequency
+        (number of positions), and document frequency (number of documents) for
+        the term involved.
+        """
+
+        doc_positions.sort()
+        return self.write_sorted_term_positions(doc_positions)
+
     def close(self):
         self.position_writer.close()
         self.position_index_writer.close()