def test_check_link_requires_python__invalid_requires(caplog):
"""
Test the log message for an invalid Requires-Python.
"""
link = Link('https://example.com', requires_python='invalid')
caplog.set_level(logging.DEBUG)
actual = _check_link_requires_python(link, version_info=(3, 6, 5))
assert actual
expected_message = (
"Ignoring invalid Requires-Python ('invalid') for link: "
"https://example.com")
check_caplog(caplog, 'DEBUG', expected_message)
def test_check_link_requires_python__invalid_requires(
caplog: pytest.LogCaptureFixture, ) -> None:
"""
Test the log message for an invalid Requires-Python.
"""
link = Link("https://example.com", requires_python="invalid")
caplog.set_level(logging.DEBUG)
actual = _check_link_requires_python(link, version_info=(3, 6, 5))
assert actual
expected_message = (
"Ignoring invalid Requires-Python ('invalid') for link: https://example.com"
)
check_caplog(caplog, "DEBUG", expected_message)
def test_check_link_requires_python__incompatible_python(
caplog, ignore_requires_python, expected,
):
"""
Test an incompatible Python.
"""
expected_return, expected_level, expected_message = expected
link = Link('https://example.com', requires_python='== 3.6.4')
caplog.set_level(logging.DEBUG)
actual = _check_link_requires_python(
link, version_info=(3, 6, 5),
ignore_requires_python=ignore_requires_python,
)
assert actual == expected_return
check_caplog(caplog, expected_level, expected_message)
def test_check_link_requires_python__incompatible_python(
caplog: pytest.LogCaptureFixture,
ignore_requires_python: bool,
expected: Tuple[bool, str, str],
) -> None:
"""
Test an incompatible Python.
"""
expected_return, expected_level, expected_message = expected
link = Link("https://example.com", requires_python="== 3.6.4")
caplog.set_level(logging.DEBUG)
actual = _check_link_requires_python(
link,
version_info=(3, 6, 5),
ignore_requires_python=ignore_requires_python,
)
assert actual == expected_return
check_caplog(caplog, expected_level, expected_message)
def process_url(self, url, retrieve=False):
"""Evaluate a URL as a possible download, and maybe retrieve it"""
if url in self.scanned_urls and not retrieve:
return
self.scanned_urls[url] = True
if not URL_SCHEME(url):
self.process_filename(url)
return
else:
dists = list(distros_for_url(url))
if dists:
if not self.url_ok(url):
return
self.debug("Found link: %s", url)
if dists or not retrieve or url in self.fetched_urls:
list(map(self.add, dists))
return # don't need the actual page
if not self.url_ok(url):
self.fetched_urls[url] = True
return
self.info("Reading %s", url)
self.fetched_urls[url] = True # prevent multiple fetch attempts
tmpl = "Download error on %s: %%s -- Some packages may not be found!"
f = self.open_url(url, tmpl % url)
if f is None:
return
if isinstance(f, urllib.error.HTTPError) and f.code == 401:
self.info("Authentication error: %s" % f.msg)
self.fetched_urls[f.url] = True
if 'html' not in f.headers.get('content-type', '').lower():
f.close() # not html, we can't process it
return
base = f.url # handle redirects
page = f.read()
# --- LOCAL CHANGES MADE HERE: ---
if isinstance(page, six.text_type):
page = page.encode('utf8')
charset = 'utf8'
else:
if isinstance(f, urllib.error.HTTPError):
# Errors have no charset, assume latin1:
charset = 'latin-1'
else:
try:
charset = f.headers.get_param('charset') or 'latin-1'
except AttributeError:
# Python 2
charset = f.headers.getparam('charset') or 'latin-1'
try:
html_page = HTMLPage(page, charset, base, cache_link_parsing=False)
except TypeError:
html_page = HTMLPage(page, charset, base)
# https://github.com/buildout/buildout/issues/598
# use_deprecated_html5lib is a required addition in pip 22.
try:
plinks = parse_links(html_page, use_deprecated_html5lib=False)
except TypeError:
plinks = parse_links(html_page)
plinks = list(plinks)
pip_links = [l.url for l in plinks]
# --- END OF LOCAL CHANGES ---
if not isinstance(page, str):
# In Python 3 and got bytes but want str.
page = page.decode(charset, "ignore")
f.close()
# --- LOCAL CHANGES MADE HERE: ---
links = []
for match in HREF.finditer(page):
link = urllib.parse.urljoin(base, htmldecode(match.group(1)))
links.append(_clean_link(link))
# TODO: remove assertion and double index page parsing before releasing.
assert set(pip_links) == set(links)
for link in plinks:
if _check_link_requires_python(link, PY_VERSION_INFO):
self.process_url(link.url)
# --- END OF LOCAL CHANGES ---
if url.startswith(self.index_url) and getattr(f, 'code', None) != 404:
page = self.process_index(url, page)
def test_check_link_requires_python(requires_python, expected):
version_info = (3, 6, 5)
link = Link('https://example.com', requires_python=requires_python)
actual = _check_link_requires_python(link, version_info)
assert actual == expected
def test_check_link_requires_python(requires_python: str,
expected: bool) -> None:
version_info = (3, 6, 5)
link = Link("https://example.com", requires_python=requires_python)
actual = _check_link_requires_python(link, version_info)
assert actual == expected
