def test_find_path(self):
'''Basic test to see if '''
wiki_graph = WikiGraph()
start, end = ('Tom Hanks', 'Will I Am')
path = wiki_graph.find_path(start, end)
self.assertEqual(path.start, start)
self.assertEqual(path.end, end)
self.assertTrue(path, "Path not found")
self.assertTrue(len(path) < 6, "Path too long len=%d" % len(path))
def main():
parser = argparse.ArgumentParser(
"Find a path between two Wikipedia pages via their links.")
parser.add_argument("--start",
help="Title of valid wikipedia page to start from.",
type=str,
required=True)
parser.add_argument("--end",
help="Title of valid wikipedia page to reach.",
type=str,
required=True)
args = parser.parse_args()
wiki_graph = WikiGraph()
print("Searching: '%s' -> '%s'" % (args.start, args.end))
path = wiki_graph.find_path(args.start, args.end)
if path:
print(path.info)
else:
print("Failed Search.")
def test_find_path_benchmark(self):
wiki_graph = WikiGraph(print_requests=True)
total_requests = 0
total_time = 0
failures = []
# Loop through and test if paths exit
for page in samplepages:
(start, end) = (page, "Homunculus")
path = wiki_graph.find_path(start, end)
if path.degree == -1:
failures.append(path)
total_requests += path.requests
total_time += path.time
print("Total Failures:", len(failures))
print(failures)
print("Total requests:", total_requests)
print("Avg number of requests per path: %.2f" %
(total_requests / len(top10pages)))
print("Total time: ", total_time)
print("Avg time per path: %.2f" % (total_time / len(top10pages)))
def main():
# initialise args from cli
parser = argparse.ArgumentParser(
"For a given sample of articles find a path from each to a central end"
" article. Write the output to a given csv file.")
parser.add_argument("-o",
"--outfile",
help="Filename to save the results to.",
type=str,
default="wikiresults.json")
parser.add_argument("-x",
"--center",
help="Title of valid wiki page to center all nodes on",
type=str,
default="Homunculus")
parser.add_argument("-k",
"--sample_size",
help="Sample size of k pages to search from. "
"(Only applies when sample source is not given)",
type=int,
default=1)
parser.add_argument(
"-s",
"--sample_source",
help="Filename containing newline delimited list of valid "
"wiki article titles if not specified sample defaults "
"to random selection from wikimedia api. ",
type=str)
parser.add_argument("-v",
action='store_true',
help="add to display titles of page requests made.")
args = parser.parse_args()
wiki_graph = WikiGraph(print_requests=True)
# resolve any issues with search sample source.
if args.sample_source:
sample = load_sample(args.set)
size = len(sample)
else:
sample = wiki_graph.random_sample(args.sample_size)
size = args.sample_size
with open(args.outfile, mode='w') as outfile:
writer = csv.DictWriter(outfile, ["start", "end", "path", "degree"])
writer.writeheader()
total_time = datetime.now()
total_requests = 0
for i, page in enumerate(sample):
print("%d/%d Searching: '%s' -> '%s'" %
(i + 1, size, page, args.center))
path = wiki_graph.find_path(page, args.center)
print(path.info)
writer.writerow(path.data)
total_requests += path.requests
total_time = (datetime.now() - total_time).total_seconds()
print("Finished Totals: "
"N={}. Time={}. Requests={}.".format(args.sample_size,
total_time, total_requests))