Merge multiple files

merge_files.py

#!/usr/local/bin/python3

import heapq
import os
import sys

class MergeFiles:
	"""
	Given a list of files, output their numbers in sorted order.

	Each files has one integer per line and those numbers are
	sorted in ascending order. There are no comments or blank
	lines.

		numbers = MergeFiles(
			'odds.txt',
			'evens.txt',
			'repeats.txt',
		)

		for i in numbers:
			print( "%s: %d" % (i[0], i[1]) )
	"""
	def __init__(self, *filenames):
		"""Create an iterable from a list of filenames

		Each call to the iterator returns the next smallest
		number from any of the files until it exhausts all of
		the files.

		If one of the files does not exist, this blows up.
		"""
		self.heap  = [];
		self.count = 0;
		for f in filenames:
			print( f )
			fh = open( f, 'r' )
			n = int(fh.readline().rstrip('\n'))
			# the count is a tie breaker argument. Otherwise, the
			# heap will try to compare next items in the tuple,
			# which are the filehandle and file name.
			self.heap.append( (n, self.count, fh, f) )
			self.count += 1
		heapq.heapify( self.heap )

	def __iter__(self):
		return self

	# the Python 3 version of the method (just next in 2)
	def __next__(self):

		# perhaps we've read the last thing in the heap.
		try:
			tuple = heapq.heappop( self.heap )
		except IndexError:
			raise StopIteration

		self.count += 1

		# might not be a numer, so we'll stop processing that
		# file
		try:
			new_tuple = (
				int(tuple[2].readline().rstrip('\n')),
				self.count,
				tuple[2],
				tuple[3],
			)
			heapq.heappush( self.heap, new_tuple )
		except:
			pass

		return [ tuple[-1], tuple[0] ]

numbers = MergeFiles( *sys.argv[1:] )
for i in numbers:
	print( "%s: %d" % (i[0], i[1]) )