def test_parse_index_simple_tilde(self):
result = parse_index(
self.simplepy,
"""<a href="/~user/py-1.4.12.zip#md5=12ab">qwe</a>""")
link, = result.releaselinks
assert link.basename == "py-1.4.12.zip"
assert link.url.endswith("/~user/py-1.4.12.zip#md5=12ab")
def test_parse_index_with_wheel(self):
result = parse_index(self.simplepy,
"""<a href="pkg/py-1.0-cp27-none-linux_x86_64.whl" />
""")
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == "py-1.0-cp27-none-linux_x86_64.whl"
def test_parse_index_two_eggs_same_url(self):
simplepy = URL("https://pypi.org/simple/Py/")
result = parse_index(
simplepy, """<a href="../../pkg/pyzip#egg=py-dev">qwe2</a>
<a href="../../pkg/pyzip#egg=py-dev">qwe</a>
""")
assert len(result.releaselinks) == 1
def test_parse_index_normalized_name(self):
simplepy = URL("https://pypi.org/simple/ndg-httpsclient/")
result = parse_index(simplepy, """
<a href="../../pkg/ndg_httpsclient-1.0.tar.gz" />
""")
assert len(result.releaselinks) == 1
assert result.releaselinks[0].url.endswith("ndg_httpsclient-1.0.tar.gz")
def test_releasefile_md5_matching_and_ordering(self):
""" check that md5-links win over non-md5 links anywhere.
And otherwise the links from the index page win over scraped ones.
"""
result = parse_index(
self.simplepy,
"""<a href="../../pkg/py-1.4.12.zip#md5=12ab">qwe</a>
<a href="../../pkg/py-1.4.11.zip">qwe</a>
<a href="../../pkg/py-1.4.10.zip#md5=2222">qwe</a>
<a href="http://pylib.org" rel="homepage">whatever</a>
<a href="http://pylib2.org" rel="download">whatever2</a>
""")
assert len(result.releaselinks) == 3
assert len(result.crawllinks) == 2
result.parse_index(URL("http://pylib.org"),
"""
<a href="http://pylib.org/py-1.4.12.zip" />
<a href="http://pylib.org/py-1.4.11.zip#md5=1111" />
<a href="http://pylib.org/py-1.4.10.zip#md5=12ab" />
""",
scrape=False)
assert len(result.crawllinks) == 2
assert len(result.releaselinks) == 3
link1, link2, link3 = result.releaselinks
assert link1.url == "https://pypi.org/pkg/py-1.4.12.zip#md5=12ab"
assert link2.url == "http://pylib.org/py-1.4.11.zip#md5=1111"
assert link3.url == "https://pypi.org/pkg/py-1.4.10.zip#md5=2222"
def test_parse_index_simple_dir_egg_issue63(self):
simplepy = URL("https://pypi.org/simple/py/")
result = parse_index(
simplepy, """<a href="../../pkg/py-1.4.12.zip#md5=12ab">qwe</a>
<a href="../../pkg/#egg=py-dev">qwe</a>
""")
assert len(result.releaselinks) == 1
def test_parse_index_simple_nocase(self):
simplepy = URL("https://pypi.org/simple/Py/")
result = parse_index(
simplepy, """<a href="../../pkg/py-1.4.12.zip#md5=12ab">qwe</a>
<a href="../../pkg/PY-1.4.13.zip">qwe</a>
""")
assert len(result.releaselinks) == 2
def test_parse_index_with_requires_python(self):
result = parse_index(self.simplepy,
"""<a href="pkg/py-1.0.zip" data-requires-python="<3" />
""")
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == "py-1.0.zip"
assert link.requires_python == "<3"
def test_parse_index_with_yanked(self):
result = parse_index(
self.simplepy,
"""<a href="pkg/py-1.0.zip" data-yanked="" />""")
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == "py-1.0.zip"
assert link.yanked is True
def test_parse_index_ftp_ignored_for_now(self):
result = parse_index(
self.simplepy, """<a href="http://bb.org/download/py-1.0.zip" />
<a href="ftp://bb.org/download/py-1.0.tar.gz" />
<a rel="download" href="ftp://bb.org/download/py-1.1.tar.gz" />
""")
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == "py-1.0.zip"
def test_parse_index_with_yanked_first_with_hash_spec_kept(self):
result = parse_index(self.simplepy,
"""<a href="pkg/py-1.0.zip#md5=pony"/>
<a href="pkg/py-1.0.zip#md5=pony" data-yanked="" />
""")
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == "py-1.0.zip"
assert link.hash_spec == "md5=pony"
assert link.yanked is False
def test_parse_index_with_rel(self, rel):
result = parse_index(self.simplepy, """
<a href="http://pylib.org" rel="%s">whatever</a>
<a href="http://pylib2.org" rel="%s">whatever2</a>
<a href="http://pylib3.org">py-1.0.zip</a>
<a href="http://pylib2.org/py-1.0.zip" rel="%s">whatever2</a>
""" % (rel,rel, rel))
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link == "http://pylib2.org/py-1.0.zip"
def test_parse_index_with_requires_python_first_with_hash_spec_kept(self):
result = parse_index(
self.simplepy, """<a href="pkg/py-1.0.zip#md5=pony"/>
<a href="pkg/py-1.0.zip#md5=pony" data-requires-python="<3" />
""")
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == "py-1.0.zip"
assert link.hash_spec == "md5=pony"
assert link.requires_python is None
def test_parse_index_simple_hash_types(self, hash_type, hash_value):
result = parse_index(
self.simplepy, """<a href="../../pkg/py-1.4.12.zip#%s=%s" /a>""" %
(hash_type, hash_value))
link, = result.releaselinks
assert link.basename == "py-1.4.12.zip"
assert link.hash_spec == "%s=%s" % (hash_type, hash_value)
if hash_type == "md5":
assert link.md5 == hash_value
else:
assert link.md5 is None
assert link.hash_algo == getattr(hashlib, hash_type)
def test_parse_index_with_egg(self):
result = parse_index(
self.simplepy,
"""<a href="http://bb.org/download/py.zip#egg=py-dev" />
<a href="http://bb.org/download/py-1.0.zip" />
<a href="http://bb.org/download/py.zip#egg=something-dev" />
""")
assert len(result.releaselinks) == 2
link, link2 = result.releaselinks
assert link.basename == "py.zip"
assert link.eggfragment == "py-dev"
assert link2.basename == "py-1.0.zip"
def test_releasefile_and_scrape(self):
result = parse_index(self.simplepy,
"""<a href="../../pkg/py-1.4.12.zip#md5=12ab">qwe</a>
<a href="http://pylib.org" rel="homepage">whatever</a>
<a href="http://pylib2.org" rel="download">whatever2</a>
""")
assert len(result.releaselinks) == 1
result.parse_index(URL("http://pylib.org"), """
<a href="http://pylib.org/py-1.1-py27.egg" />
<a href="http://pylib.org/other" rel="download" />
""")
assert len(result.releaselinks) == 2
links = list(result.releaselinks)
assert links[0].url == "https://pypi.org/pkg/py-1.4.12.zip#md5=12ab"
assert links[1].url == "http://pylib.org/py-1.1-py27.egg"
def test_parse_index_with_two_eggs_ordering(self):
# it seems that pip/easy_install in some cases
# rely on the exact ordering of eggs in the html page
# for example with nose, there are two eggs and the second/last
# one is chosen due to the internal pip/easy_install algorithm
result = parse_index(
self.simplepy,
"""<a href="http://bb.org/download/py.zip#egg=py-dev" />
<a href="http://other/master#egg=py-dev" />
""")
assert len(result.releaselinks) == 2
link1, link2 = result.releaselinks
assert link1.basename == "master"
assert link1.eggfragment == "py-dev"
assert link2.basename == "py.zip"
assert link2.eggfragment == "py-dev"
def test_parse_index_egg_svnurl(self, monkeypatch):
# strange case reported by fschulze/witsch where
# urlparsing will yield a fragment for svn urls.
# it's not exactly clear how urlparse.uses_fragment
# sometimes contains "svn" but it's good to check
# that we are not sensitive to the issue.
try:
import urllib.parse as urlparse
except ImportError:
# PY2
import urlparse
monkeypatch.setattr(urlparse, "uses_fragment",
urlparse.uses_fragment + ["svn"])
simplepy = URL("https://pypi.org/simple/zope.sqlalchemy/")
result = parse_index(
simplepy, '<a href="svn://svn.zope.org/repos/main/'
'zope.sqlalchemy/trunk#egg=zope.sqlalchemy-dev" />')
assert len(result.releaselinks) == 0
assert len(result.egglinks) == 0
def test_parse_index_invalid_link(self, pypistage):
result = parse_index(
self.simplepy, '''
<a rel="download" href="https:/host.com/123" />
''')
assert result.releaselinks == []
def test_parse_index_with_valid_basenames(self, basename):
result = parse_index(self.simplepy, '<a href="pkg/%s" />' % basename)
assert len(result.releaselinks) == 1
link, = result.releaselinks
assert link.basename == basename
def test_parse_index_invalid_link(self, pypistage):
result = parse_index(
self.simplepy, '''
<a rel="download" href="https:/host.com/123" />
''')
assert result.crawllinks == {URL('https://pypi.org/host.com/123')}
def test_parse_index_unparseable_url(self):
simple = URL("https://pypi.org/simple/x123/")
result = parse_index(simple, '<a href="http:" />')
assert len(result.releaselinks) == 0
def test_releasefile_and_scrape_no_ftp(self):
result = parse_index(
self.simplepy, """<a href="ftp://pylib2.org/py-1.0.tar.gz"
rel="download">whatever2</a> """)
assert len(result.releaselinks) == 0
assert len(result.crawllinks) == 0
def test_parse_index_with_non_parseable_hrefs(self):
result = parse_index(
self.simplepy, """<a href="qlkwje 1lk23j123123" />
<a href="http://bb.org/download/py-1.0.zip" />""")
assert len(result.releaselinks) == 1
def test_parse_index_with_matchingproject_no_version(self):
result = parse_index(
self.simplepy, """<a href="http://bb.org/download/p.zip" />
<a href="http://bb.org/download/py-1.0.zip" />""")
assert len(result.releaselinks) == 1
def test_parse_index_simple_nomatch(self):
result = parse_index(self.simplepy,
"""<a href="../../pkg/py-1.3.html">qwe</a>""")
assert not result.releaselinks
def test_parse_index_with_num_in_project(self):
simple = URL("https://pypi.org/simple/py-4chan/")
result = parse_index(simple, '<a href="pkg/py-4chan-1.0.zip"/>')
assert len(result.releaselinks) == 1
assert result.releaselinks[0].basename == "py-4chan-1.0.zip"