def test_extract_relative_urls(self):
page = """
<html>
<body>
<p>
<a href="foo.html">FooPage</a>
</p>
<div>
<div>
<div>
<div>
<a href="/sub/page/../bar.html">BarPage</a>
</div>
</div>
</div>
</div>
</body>
</html>
"""
expected_links = [
Link(self.crawled_page_url, "/foo.html"),
Link(self.crawled_page_url, "/sub/bar.html"),
]
actual_links = LinkExtractor.extract(self.crawled_page_url, page)
self.assertEqual(actual_links, expected_links)
class TestPage(unittest.TestCase):
@patch("crawler.pages.page.LinkExtractor.extract",
return_value=[
Link("http://www.example.com", "foo/index.html"),
Link("http://www.example.com", "bar/index.html"),
])
def test_link(self, mock_link_extractor):
link = Link("http://www.example.com/", "index.html")
page = Page(link, "mocked_page_body")
self.assertEqual(page.link, link)
mock_link_extractor.assert_called_with(
"http://www.example.com/index.html", "mocked_page_body")
@patch("crawler.pages.page.LinkExtractor.extract",
return_value=[
Link("http://www.example.com", "foo/index.html"),
Link("http://www.example.com", "bar/index.html"),
])
def test_out_links(self, mocked_link_extractor):
link = Link("http://www.example.com/", "index.html")
page = Page(link, "mocked_page_body")
self.assertEqual(page.out_links, [
Link("http://www.example.com", "foo/index.html"),
Link("http://www.example.com", "bar/index.html"),
])
def test_extract_multiple_links(self):
page = """
<html>
<head>
<title>Test Page Two Links</title>
</head>
<body>
<p>
<a href="http://www.example.com/foo.html">FooPage</a>
</p>
<div>
<div>
<div>
<div>
<a href="https://www.example.com/sub/page/bar.html">BarPage</a>
</div>
</div>
</div>
</div>
</body>
</html>
"""
expected_links = [
Link(self.crawled_page_url, "/foo.html"),
Link(self.crawled_page_url, "/sub/page/bar.html"),
]
actual_links = LinkExtractor.extract(self.crawled_page_url, page)
self.assertEqual(actual_links, expected_links)
def test_equal_different_not_equivalent_complex_path(self):
self.assertNotEqual(
Link(self.crawled_page,
"http://www.example.com/sub/path/../foo/bar.html"),
Link(self.crawled_page,
"http://www.example.net/sub/path/bar.html"),
)
def test_out_links(self, mocked_link_extractor):
link = Link("http://www.example.com/", "index.html")
page = Page(link, "mocked_page_body")
self.assertEqual(page.out_links, [
Link("http://www.example.com", "foo/index.html"),
Link("http://www.example.com", "bar/index.html"),
])
def test_link(self, mock_link_extractor):
link = Link("http://www.example.com/", "index.html")
page = Page(link, "mocked_page_body")
self.assertEqual(page.link, link)
mock_link_extractor.assert_called_with(
"http://www.example.com/index.html", "mocked_page_body")
def extract(crawled_page_url, page_text):
"""Given a web page will extract all <a> links, turn them into crawler.links.link.Link instances
and return the results.
Note: Will silently ignore all invalid (semantically, not whether they lead somewhere) links, and all
Links with an unknown url scheme
Args:
crawled_page_url (string): The url of the crawled page
page_text (string): The web page text
Returns:
list: List of crawler.links.link.Link instances representing every unique link on the page
"""
parsed_page = PyQuery(page_text)
links = []
anchor_elements = parsed_page("a[href]")
for anchor_element in anchor_elements:
try:
link = Link(crawled_page_url, anchor_element.attrib["href"])
links.append(link)
except (InvalidPathError, UnknownSchemeError):
next
return links
def __init__(self, start_domain):
"""Initialiser
Args:
start_domain (string): The domain to start crawling
"""
self._start_link = Link(start_domain, "/")
self.site_map = SiteMap()
self._links_to_visit = set()
def test_get(self):
responses.add(
**{
"method": responses.GET,
"url": "http://www.example.com/index.html",
"body": TestPageFetcher.MOCK_PAGE,
"status": 200,
"content_type": "application/html",
})
expected_out_links = [
Link("http://www.example.com/index.html", "/foo.html"),
Link("http://www.example.com/index.html", "/sub/page/bar.html"),
]
expected_link = Link("http://www.example.com/", "index.html")
actual_page = PageFetcher.get(
Link("http://www.example.com/", "index.html"))
self.assertEqual(actual_page.link, expected_link)
self.assertEqual(actual_page.out_links, expected_out_links)
def test_extract_includes_external_links(self):
page = """
<html>
<body>
<p>
<a href="http://www.example.com/foo.html">FooPage</a>
<a href="http://example.com/bar.html">BarPage</a>
<a href="http://www.example.net/baz.html">BazPage</a>
</p>
</body>
</html>
"""
expected_links = [
Link(self.crawled_page_url, "/foo.html"),
Link(self.crawled_page_url, "http://example.com/bar.html"),
Link(self.crawled_page_url, "http://www.example.net/baz.html")
]
actual_links = LinkExtractor.extract(self.crawled_page_url, page)
self.assertEqual(actual_links, expected_links)
def test_extract_one_link(self):
page = """
<html>
<head>
<title>Test Page One Link</title>
</head>
<body>
<p>
<a href="http://www.example.com/foo.html">FooPage</a>
</p>
</body>
</html>
"""
expected_links = [
Link(self.crawled_page_url, "/foo.html"),
]
actual_links = LinkExtractor.extract(self.crawled_page_url, page)
self.assertEqual(actual_links, expected_links)
def test_discards_invalid_links(self):
page = """
<html>
<body>
<p>
<a href="ftp://www.example.com/foo.html">FooPage</a>
<a href="example.com/../../bar.html">BarPage</a>
<a href="/baz.html">BazPage</a>
</p>
</body>
</html>
"""
expected_links = [
Link("http://www.example.com", "/baz.html"),
]
actual_links = LinkExtractor.extract(self.crawled_page_url, page)
self.assertEqual(actual_links, expected_links)
def test_in_crawled_domain(self):
link = Link(self.crawled_page, "")
self.assertTrue(link.in_crawled_domain())
def test_url_simple_absolute_url_different_port(self):
self.assertEqual(
Link(self.crawled_page,
"http://www.example.net:123/index.html").url,
"http://www.example.net:123/index.html",
)
def test_url_simple_absolute_url_same_domain(self):
self.assertEqual(
Link(self.crawled_page, "http://www.example.com/index.html").url,
"http://www.example.com/index.html",
)
def test_url_complex_relative_url(self):
self.assertEqual(
Link(self.crawled_page, "/sub/path/../path/index.html").url,
"http://www.example.com/sub/path/index.html",
)
def test_equal_different_port(self):
self.assertNotEqual(
Link(self.crawled_page, "index.html"),
Link(self.crawled_page, "index.htm"),
)
def test_in_crawled_domain_parent_domain(self):
link = Link(self.crawled_page, "http://example.com/foo.html")
self.assertFalse(link.in_crawled_domain())
def test_in_crawled_domain_with_absolute_path_different_port(self):
link = Link(self.crawled_page, "http://www.example.com:123/foo.html")
self.assertTrue(link.in_crawled_domain())
def test_init_with_simple_relative_path_that_escapes_root(self):
with self.assertRaises(InvalidPathError):
Link(self.crawled_page, "../foo.html")
def test_hash(self):
self.assertEqual(
Link(self.crawled_page, "/foo/bar.html").__hash__(),
hash("www.example.com/foo/bar.html"),
)
def test_equal_different_tld(self):
self.assertNotEqual(
Link(self.crawled_page, "http://www.example.com/index.html"),
Link(self.crawled_page, "http://www.example.net/index.html"),
)
def test_equal_different_subdomain(self):
self.assertNotEqual(
Link(self.crawled_page, "http://www.example.com/index.html"),
Link(self.crawled_page, "http://foo.example.com/index.html"),
)
def test_in_crawled_domain_with_relative_path(self):
link = Link(self.crawled_page, "foo.html")
self.assertTrue(link.in_crawled_domain())
def test_init_with_unknown_scheme(self):
with self.assertRaises(UnknownSchemeError):
Link(self.crawled_page, "foo://127.0.0.1")
def test_init_with_complex_relative_path_that_escapes_root(self):
with self.assertRaises(InvalidPathError):
Link(self.crawled_page, "/path/foo/../../../bar.html")
def test_in_crawled_domain_different_scheme(self):
link = Link(self.crawled_page, "https://www.example.com/foo.html")
self.assertTrue(link.in_crawled_domain())
def test_url_simple_relative_url(self):
self.assertEqual(
Link(self.crawled_page, "index.html").url,
"http://www.example.com/index.html",
)
def test_init_with_complex_absolute_path_that_escapes_root(self):
with self.assertRaises(InvalidPathError):
Link(self.crawled_page,
"http://www.example.com/path/../foo/../../index.html")
def test_url_crawled_subpage_relative_url(self):
self.assertEqual(
Link("http://www.example.com/sub/path/index.html", "foo.html").url,
"http://www.example.com/sub/path/foo.html",
)
