def cluster_and_test(urls, pattern_string):
pm = PatternMaker(get_default_config())
for url in urls:
pm.load(url)
for url_meta, clustered in pm.make(combine=True):
for nodes in dump_tree(clustered):
assert pack(url_meta,
[n.value for n in nodes[1:]]) == pattern_string
def test_unpack_pack():
data = [
('http://www.g.com/', '/'),
('http://www.g.com/abc', '/abc'),
('http://www.g.com/abc?a=1#c', '/abc[\\?]a=1#c'),
('http://www.g.com/abc???a=1#c', '/abc[\\?][\\?]{2}a=1#c'),
('http://www.g.com/abc?=1#c', '/abc[\\?]=1#c'),
('http://www.g.com/abc?a=1#', '/abc[\\?]a=1#'),
('http://www.g.com/abc?a=1&b=2#', '/abc[\\?]a=1&b=2#'),
]
for url, expected in data:
assert pack(*analyze_url(url)) == expected
def test_parse_url_pattern():
data = [
'http://www.g.com/',
'http://www.g.com/abc',
'http://www.g.com/abc?a=1#c',
'http://www.g.com/abc???a=1#c',
'http://www.g.com/abc?=1#c',
'http://www.g.com/abc?a=1#',
'http://www.g.com/abc?a=1&b=2#',
]
for url in data:
meta1, parts1 = analyze_url(url)
pattern_string = pack(meta1, parts1)
meta2, parts2 = analyze_url_pattern_string(pattern_string)
assert meta1 == meta2
assert len(parts1) == len(parts2)