Poogle - Web Search

It will take a series of search terms (the items in the list called "inputs") and search through the various web pages (in the "pages" list) and count the number of pages where a match is found.

file: poogle.py
1	# Challenge, return number of matches within a page
2	import mpi
3	import urllib
4
5	# what we are searching for
6	inputs = [
7	"seidel",
8	"beck",
9	"loni",
10	"eastman",
11	"brandt"]
12
13	# collection of web pages our search engine knows about
14	pages = [
15	'http://www.cct.lsu.edu/about/overview/',
16	'http://www.cct.lsu.edu/projects/Coastal_Modeling/',
17	'http://www.cct.lsu.edu/projects/SURASCOOP/',
18	'http://www.cct.lsu.edu/projects/CactusCode/',
19	'http://www.cct.lsu.edu/about/employment/employment.php',
20	'http://www.cct.lsu.edu/projects/GridChemCCG/',
21	'http://www.cct.lsu.edu/about/people/faculty/all.php',
22	'http://www.cct.lsu.edu/about/focus/',
23	'http://www.cct.lsu.edu/projects/Enlightened/',
24	'http://www.loni.org/',
25	'http://www.loni.org/plan/',
26	'http://www.cct.lsu.edu/news/news/289']
27
28	n = len(pages)/mpi.size
29	ilo = mpi.rank*n
30	ihi = (mpi.rank+1)*n-1
31
32	# each mpi proc searches a subset of pages from ilo to ihi
33	c = range(ihi+1)
34	for i in range(ilo,ihi+1):
35	c[i] = urllib.urlopen(pages[i]).read().lower()
36
37	for input in inputs:
38	matches = []
39
40	for i in range(ilo,ihi+1):
41	if c[i].find(input) >= 0:
42	matches.append(pages[i])
43
44	# proc zero receives the results of all searches
45	if mpi.rank == 0:
46	for i in range(1,mpi.size):
47	other_matches = mpi.recv(mpi.ANY_SOURCE)[0]
48	for match in other_matches:
49	matches.append(match)
50	else:
51	mpi.send(matches,0)
52
53	if mpi.rank == 0:
54	print input,len(matches)
> mpirun -np 2 pyMPI poogle.py seidel 2 beck 0 loni 4 eastman 0 brandt 0