| 1 | n/a | """Tests for the packaging.pypi.simple module.""" |
|---|
| 2 | n/a | import re |
|---|
| 3 | n/a | import os |
|---|
| 4 | n/a | import sys |
|---|
| 5 | n/a | import http.client |
|---|
| 6 | n/a | import urllib.error |
|---|
| 7 | n/a | import urllib.parse |
|---|
| 8 | n/a | import urllib.request |
|---|
| 9 | n/a | |
|---|
| 10 | n/a | from packaging.pypi.simple import Crawler |
|---|
| 11 | n/a | |
|---|
| 12 | n/a | from packaging.tests import unittest |
|---|
| 13 | n/a | from packaging.tests.support import (TempdirManager, LoggingCatcher, |
|---|
| 14 | n/a | fake_dec) |
|---|
| 15 | n/a | |
|---|
| 16 | n/a | try: |
|---|
| 17 | n/a | import _thread |
|---|
| 18 | n/a | from packaging.tests.pypi_server import (use_pypi_server, PyPIServer, |
|---|
| 19 | n/a | PYPI_DEFAULT_STATIC_PATH) |
|---|
| 20 | n/a | except ImportError: |
|---|
| 21 | n/a | _thread = None |
|---|
| 22 | n/a | use_pypi_server = fake_dec |
|---|
| 23 | n/a | PYPI_DEFAULT_STATIC_PATH = os.path.join( |
|---|
| 24 | n/a | os.path.dirname(os.path.abspath(__file__)), 'pypiserver') |
|---|
| 25 | n/a | |
|---|
| 26 | n/a | |
|---|
| 27 | n/a | |
|---|
| 28 | n/a | class SimpleCrawlerTestCase(TempdirManager, |
|---|
| 29 | n/a | LoggingCatcher, |
|---|
| 30 | n/a | unittest.TestCase): |
|---|
| 31 | n/a | |
|---|
| 32 | n/a | def _get_simple_crawler(self, server, base_url="/simple/", hosts=None, |
|---|
| 33 | n/a | *args, **kwargs): |
|---|
| 34 | n/a | """Build and return a SimpleIndex with the test server urls""" |
|---|
| 35 | n/a | if hosts is None: |
|---|
| 36 | n/a | hosts = (server.full_address.replace("http://", ""),) |
|---|
| 37 | n/a | kwargs['hosts'] = hosts |
|---|
| 38 | n/a | return Crawler(server.full_address + base_url, *args, |
|---|
| 39 | n/a | **kwargs) |
|---|
| 40 | n/a | |
|---|
| 41 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 42 | n/a | @use_pypi_server() |
|---|
| 43 | n/a | def test_bad_urls(self, server): |
|---|
| 44 | n/a | crawler = Crawler() |
|---|
| 45 | n/a | url = 'http://127.0.0.1:0/nonesuch/test_simple' |
|---|
| 46 | n/a | try: |
|---|
| 47 | n/a | v = crawler._open_url(url) |
|---|
| 48 | n/a | except Exception as v: |
|---|
| 49 | n/a | self.assertIn(url, str(v)) |
|---|
| 50 | n/a | else: |
|---|
| 51 | n/a | v.close() |
|---|
| 52 | n/a | self.assertIsInstance(v, urllib.error.HTTPError) |
|---|
| 53 | n/a | |
|---|
| 54 | n/a | # issue 16 |
|---|
| 55 | n/a | # easy_install inquant.contentmirror.plone breaks because of a typo |
|---|
| 56 | n/a | # in its home URL |
|---|
| 57 | n/a | crawler = Crawler(hosts=('example.org',)) |
|---|
| 58 | n/a | url = ('url:%20https://svn.plone.org/svn/collective/' |
|---|
| 59 | n/a | 'inquant.contentmirror.plone/trunk') |
|---|
| 60 | n/a | try: |
|---|
| 61 | n/a | v = crawler._open_url(url) |
|---|
| 62 | n/a | except Exception as v: |
|---|
| 63 | n/a | self.assertIn(url, str(v)) |
|---|
| 64 | n/a | else: |
|---|
| 65 | n/a | v.close() |
|---|
| 66 | n/a | self.assertIsInstance(v, urllib.error.HTTPError) |
|---|
| 67 | n/a | |
|---|
| 68 | n/a | def _urlopen(*args): |
|---|
| 69 | n/a | raise http.client.BadStatusLine('line') |
|---|
| 70 | n/a | |
|---|
| 71 | n/a | old_urlopen = urllib.request.urlopen |
|---|
| 72 | n/a | urllib.request.urlopen = _urlopen |
|---|
| 73 | n/a | url = 'http://example.org' |
|---|
| 74 | n/a | try: |
|---|
| 75 | n/a | v = crawler._open_url(url) |
|---|
| 76 | n/a | except Exception as v: |
|---|
| 77 | n/a | self.assertIn('line', str(v)) |
|---|
| 78 | n/a | else: |
|---|
| 79 | n/a | v.close() |
|---|
| 80 | n/a | # TODO use self.assertRaises |
|---|
| 81 | n/a | raise AssertionError('Should have raise here!') |
|---|
| 82 | n/a | finally: |
|---|
| 83 | n/a | urllib.request.urlopen = old_urlopen |
|---|
| 84 | n/a | |
|---|
| 85 | n/a | # issue 20 |
|---|
| 86 | n/a | url = 'http://http://svn.pythonpaste.org/Paste/wphp/trunk' |
|---|
| 87 | n/a | try: |
|---|
| 88 | n/a | crawler._open_url(url) |
|---|
| 89 | n/a | except Exception as v: |
|---|
| 90 | n/a | self.assertIn('Download error', str(v)) |
|---|
| 91 | n/a | |
|---|
| 92 | n/a | # issue #160 |
|---|
| 93 | n/a | url = server.full_address |
|---|
| 94 | n/a | page = ('<a href="http://www.famfamfam.com](' |
|---|
| 95 | n/a | 'http://www.famfamfam.com/">') |
|---|
| 96 | n/a | crawler._process_url(url, page) |
|---|
| 97 | n/a | |
|---|
| 98 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 99 | n/a | @use_pypi_server("test_found_links") |
|---|
| 100 | n/a | def test_found_links(self, server): |
|---|
| 101 | n/a | # Browse the index, asking for a specified release version |
|---|
| 102 | n/a | # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1 |
|---|
| 103 | n/a | crawler = self._get_simple_crawler(server) |
|---|
| 104 | n/a | last_release = crawler.get_release("foobar") |
|---|
| 105 | n/a | |
|---|
| 106 | n/a | # we have scanned the index page |
|---|
| 107 | n/a | self.assertIn(server.full_address + "/simple/foobar/", |
|---|
| 108 | n/a | crawler._processed_urls) |
|---|
| 109 | n/a | |
|---|
| 110 | n/a | # we have found 4 releases in this page |
|---|
| 111 | n/a | self.assertEqual(len(crawler._projects["foobar"]), 4) |
|---|
| 112 | n/a | |
|---|
| 113 | n/a | # and returned the most recent one |
|---|
| 114 | n/a | self.assertEqual("%s" % last_release.version, '2.0.1') |
|---|
| 115 | n/a | |
|---|
| 116 | n/a | def test_is_browsable(self): |
|---|
| 117 | n/a | crawler = Crawler(follow_externals=False) |
|---|
| 118 | n/a | self.assertTrue(crawler._is_browsable(crawler.index_url + "test")) |
|---|
| 119 | n/a | |
|---|
| 120 | n/a | # Now, when following externals, we can have a list of hosts to trust. |
|---|
| 121 | n/a | # and don't follow other external links than the one described here. |
|---|
| 122 | n/a | crawler = Crawler(hosts=["pypi.python.org", "example.org"], |
|---|
| 123 | n/a | follow_externals=True) |
|---|
| 124 | n/a | good_urls = ( |
|---|
| 125 | n/a | "http://pypi.python.org/foo/bar", |
|---|
| 126 | n/a | "http://pypi.python.org/simple/foobar", |
|---|
| 127 | n/a | "http://example.org", |
|---|
| 128 | n/a | "http://example.org/", |
|---|
| 129 | n/a | "http://example.org/simple/", |
|---|
| 130 | n/a | ) |
|---|
| 131 | n/a | bad_urls = ( |
|---|
| 132 | n/a | "http://python.org", |
|---|
| 133 | n/a | "http://example.tld", |
|---|
| 134 | n/a | ) |
|---|
| 135 | n/a | |
|---|
| 136 | n/a | for url in good_urls: |
|---|
| 137 | n/a | self.assertTrue(crawler._is_browsable(url)) |
|---|
| 138 | n/a | |
|---|
| 139 | n/a | for url in bad_urls: |
|---|
| 140 | n/a | self.assertFalse(crawler._is_browsable(url)) |
|---|
| 141 | n/a | |
|---|
| 142 | n/a | # allow all hosts |
|---|
| 143 | n/a | crawler = Crawler(follow_externals=True, hosts=("*",)) |
|---|
| 144 | n/a | self.assertTrue(crawler._is_browsable("http://an-external.link/path")) |
|---|
| 145 | n/a | self.assertTrue(crawler._is_browsable("pypi.example.org/a/path")) |
|---|
| 146 | n/a | |
|---|
| 147 | n/a | # specify a list of hosts we want to allow |
|---|
| 148 | n/a | crawler = Crawler(follow_externals=True, |
|---|
| 149 | n/a | hosts=("*.example.org",)) |
|---|
| 150 | n/a | self.assertFalse(crawler._is_browsable("http://an-external.link/path")) |
|---|
| 151 | n/a | self.assertTrue( |
|---|
| 152 | n/a | crawler._is_browsable("http://pypi.example.org/a/path")) |
|---|
| 153 | n/a | |
|---|
| 154 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 155 | n/a | @use_pypi_server("with_externals") |
|---|
| 156 | n/a | def test_follow_externals(self, server): |
|---|
| 157 | n/a | # Include external pages |
|---|
| 158 | n/a | # Try to request the package index, wich contains links to "externals" |
|---|
| 159 | n/a | # resources. They have to be scanned too. |
|---|
| 160 | n/a | crawler = self._get_simple_crawler(server, follow_externals=True) |
|---|
| 161 | n/a | crawler.get_release("foobar") |
|---|
| 162 | n/a | self.assertIn(server.full_address + "/external/external.html", |
|---|
| 163 | n/a | crawler._processed_urls) |
|---|
| 164 | n/a | |
|---|
| 165 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 166 | n/a | @use_pypi_server("with_real_externals") |
|---|
| 167 | n/a | def test_restrict_hosts(self, server): |
|---|
| 168 | n/a | # Only use a list of allowed hosts is possible |
|---|
| 169 | n/a | # Test that telling the simple pyPI client to not retrieve external |
|---|
| 170 | n/a | # works |
|---|
| 171 | n/a | crawler = self._get_simple_crawler(server, follow_externals=False) |
|---|
| 172 | n/a | crawler.get_release("foobar") |
|---|
| 173 | n/a | self.assertNotIn(server.full_address + "/external/external.html", |
|---|
| 174 | n/a | crawler._processed_urls) |
|---|
| 175 | n/a | |
|---|
| 176 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 177 | n/a | @use_pypi_server(static_filesystem_paths=["with_externals"], |
|---|
| 178 | n/a | static_uri_paths=["simple", "external"]) |
|---|
| 179 | n/a | def test_links_priority(self, server): |
|---|
| 180 | n/a | # Download links from the pypi simple index should be used before |
|---|
| 181 | n/a | # external download links. |
|---|
| 182 | n/a | # http://bitbucket.org/tarek/distribute/issue/163/md5-validation-error |
|---|
| 183 | n/a | # |
|---|
| 184 | n/a | # Usecase : |
|---|
| 185 | n/a | # - someone uploads a package on pypi, a md5 is generated |
|---|
| 186 | n/a | # - someone manually coindexes this link (with the md5 in the url) onto |
|---|
| 187 | n/a | # an external page accessible from the package page. |
|---|
| 188 | n/a | # - someone reuploads the package (with a different md5) |
|---|
| 189 | n/a | # - while easy_installing, an MD5 error occurs because the external |
|---|
| 190 | n/a | # link is used |
|---|
| 191 | n/a | # -> The index should use the link from pypi, not the external one. |
|---|
| 192 | n/a | |
|---|
| 193 | n/a | # start an index server |
|---|
| 194 | n/a | index_url = server.full_address + '/simple/' |
|---|
| 195 | n/a | |
|---|
| 196 | n/a | # scan a test index |
|---|
| 197 | n/a | crawler = Crawler(index_url, follow_externals=True) |
|---|
| 198 | n/a | releases = crawler.get_releases("foobar") |
|---|
| 199 | n/a | server.stop() |
|---|
| 200 | n/a | |
|---|
| 201 | n/a | # we have only one link, because links are compared without md5 |
|---|
| 202 | n/a | self.assertEqual(1, len(releases)) |
|---|
| 203 | n/a | self.assertEqual(1, len(releases[0].dists)) |
|---|
| 204 | n/a | # the link should be from the index |
|---|
| 205 | n/a | self.assertEqual(2, len(releases[0].dists['sdist'].urls)) |
|---|
| 206 | n/a | self.assertEqual('12345678901234567', |
|---|
| 207 | n/a | releases[0].dists['sdist'].url['hashval']) |
|---|
| 208 | n/a | self.assertEqual('md5', releases[0].dists['sdist'].url['hashname']) |
|---|
| 209 | n/a | |
|---|
| 210 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 211 | n/a | @use_pypi_server(static_filesystem_paths=["with_norel_links"], |
|---|
| 212 | n/a | static_uri_paths=["simple", "external"]) |
|---|
| 213 | n/a | def test_not_scan_all_links(self, server): |
|---|
| 214 | n/a | # Do not follow all index page links. |
|---|
| 215 | n/a | # The links not tagged with rel="download" and rel="homepage" have |
|---|
| 216 | n/a | # to not be processed by the package index, while processing "pages". |
|---|
| 217 | n/a | |
|---|
| 218 | n/a | # process the pages |
|---|
| 219 | n/a | crawler = self._get_simple_crawler(server, follow_externals=True) |
|---|
| 220 | n/a | crawler.get_releases("foobar") |
|---|
| 221 | n/a | # now it should have processed only pages with links rel="download" |
|---|
| 222 | n/a | # and rel="homepage" |
|---|
| 223 | n/a | self.assertIn("%s/simple/foobar/" % server.full_address, |
|---|
| 224 | n/a | crawler._processed_urls) # it's the simple index page |
|---|
| 225 | n/a | self.assertIn("%s/external/homepage.html" % server.full_address, |
|---|
| 226 | n/a | crawler._processed_urls) # the external homepage is rel="homepage" |
|---|
| 227 | n/a | self.assertNotIn("%s/external/nonrel.html" % server.full_address, |
|---|
| 228 | n/a | crawler._processed_urls) # this link contains no rel=* |
|---|
| 229 | n/a | self.assertNotIn("%s/unrelated-0.2.tar.gz" % server.full_address, |
|---|
| 230 | n/a | crawler._processed_urls) # linked from simple index (no rel) |
|---|
| 231 | n/a | self.assertIn("%s/foobar-0.1.tar.gz" % server.full_address, |
|---|
| 232 | n/a | crawler._processed_urls) # linked from simple index (rel) |
|---|
| 233 | n/a | self.assertIn("%s/foobar-2.0.tar.gz" % server.full_address, |
|---|
| 234 | n/a | crawler._processed_urls) # linked from external homepage (rel) |
|---|
| 235 | n/a | |
|---|
| 236 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 237 | n/a | def test_uses_mirrors(self): |
|---|
| 238 | n/a | # When the main repository seems down, try using the given mirrors""" |
|---|
| 239 | n/a | server = PyPIServer("foo_bar_baz") |
|---|
| 240 | n/a | mirror = PyPIServer("foo_bar_baz") |
|---|
| 241 | n/a | mirror.start() # we dont start the server here |
|---|
| 242 | n/a | |
|---|
| 243 | n/a | try: |
|---|
| 244 | n/a | # create the index using both servers |
|---|
| 245 | n/a | crawler = Crawler(server.full_address + "/simple/", hosts=('*',), |
|---|
| 246 | n/a | # set the timeout to 1s for the tests |
|---|
| 247 | n/a | timeout=1, mirrors=[mirror.full_address]) |
|---|
| 248 | n/a | |
|---|
| 249 | n/a | # this should not raise a timeout |
|---|
| 250 | n/a | self.assertEqual(4, len(crawler.get_releases("foo"))) |
|---|
| 251 | n/a | finally: |
|---|
| 252 | n/a | mirror.stop() |
|---|
| 253 | n/a | server.stop() |
|---|
| 254 | n/a | |
|---|
| 255 | n/a | def test_simple_link_matcher(self): |
|---|
| 256 | n/a | # Test that the simple link matcher finds the right links""" |
|---|
| 257 | n/a | crawler = Crawler(follow_externals=False) |
|---|
| 258 | n/a | |
|---|
| 259 | n/a | # Here, we define: |
|---|
| 260 | n/a | # 1. one link that must be followed, cause it's a download one |
|---|
| 261 | n/a | # 2. one link that must *not* be followed, cause the is_browsable |
|---|
| 262 | n/a | # returns false for it. |
|---|
| 263 | n/a | # 3. one link that must be followed cause it's a homepage that is |
|---|
| 264 | n/a | # browsable |
|---|
| 265 | n/a | # 4. one link that must be followed, because it contain a md5 hash |
|---|
| 266 | n/a | self.assertTrue(crawler._is_browsable("%stest" % crawler.index_url)) |
|---|
| 267 | n/a | self.assertFalse(crawler._is_browsable("http://dl-link2")) |
|---|
| 268 | n/a | content = """ |
|---|
| 269 | n/a | <a href="http://dl-link1" rel="download">download_link1</a> |
|---|
| 270 | n/a | <a href="http://dl-link2" rel="homepage">homepage_link1</a> |
|---|
| 271 | n/a | <a href="%(index_url)stest" rel="homepage">homepage_link2</a> |
|---|
| 272 | n/a | <a href="%(index_url)stest/foobar-1.tar.gz#md5=abcdef>download_link2</a> |
|---|
| 273 | n/a | """ % {'index_url': crawler.index_url} |
|---|
| 274 | n/a | |
|---|
| 275 | n/a | # Test that the simple link matcher yield the good links. |
|---|
| 276 | n/a | generator = crawler._simple_link_matcher(content, crawler.index_url) |
|---|
| 277 | n/a | self.assertEqual(('%stest/foobar-1.tar.gz#md5=abcdef' % |
|---|
| 278 | n/a | crawler.index_url, True), next(generator)) |
|---|
| 279 | n/a | self.assertEqual(('http://dl-link1', True), next(generator)) |
|---|
| 280 | n/a | self.assertEqual(('%stest' % crawler.index_url, False), |
|---|
| 281 | n/a | next(generator)) |
|---|
| 282 | n/a | self.assertRaises(StopIteration, generator.__next__) |
|---|
| 283 | n/a | |
|---|
| 284 | n/a | # Follow the external links is possible (eg. homepages) |
|---|
| 285 | n/a | crawler.follow_externals = True |
|---|
| 286 | n/a | generator = crawler._simple_link_matcher(content, crawler.index_url) |
|---|
| 287 | n/a | self.assertEqual(('%stest/foobar-1.tar.gz#md5=abcdef' % |
|---|
| 288 | n/a | crawler.index_url, True), next(generator)) |
|---|
| 289 | n/a | self.assertEqual(('http://dl-link1', True), next(generator)) |
|---|
| 290 | n/a | self.assertEqual(('http://dl-link2', False), next(generator)) |
|---|
| 291 | n/a | self.assertEqual(('%stest' % crawler.index_url, False), |
|---|
| 292 | n/a | next(generator)) |
|---|
| 293 | n/a | self.assertRaises(StopIteration, generator.__next__) |
|---|
| 294 | n/a | |
|---|
| 295 | n/a | def test_browse_local_files(self): |
|---|
| 296 | n/a | # Test that we can browse local files""" |
|---|
| 297 | n/a | index_url = "file://" + PYPI_DEFAULT_STATIC_PATH |
|---|
| 298 | n/a | if sys.platform == 'win32': |
|---|
| 299 | n/a | # under windows the correct syntax is: |
|---|
| 300 | n/a | # file:///C|\the\path\here |
|---|
| 301 | n/a | # instead of |
|---|
| 302 | n/a | # file://C:\the\path\here |
|---|
| 303 | n/a | fix = re.compile(r'^(file://)([A-Za-z])(:)') |
|---|
| 304 | n/a | index_url = fix.sub('\\1/\\2|', index_url) |
|---|
| 305 | n/a | |
|---|
| 306 | n/a | index_path = os.sep.join([index_url, "test_found_links", "simple"]) |
|---|
| 307 | n/a | crawler = Crawler(index_path) |
|---|
| 308 | n/a | dists = crawler.get_releases("foobar") |
|---|
| 309 | n/a | self.assertEqual(4, len(dists)) |
|---|
| 310 | n/a | |
|---|
| 311 | n/a | def test_get_link_matcher(self): |
|---|
| 312 | n/a | crawler = Crawler("http://example.org") |
|---|
| 313 | n/a | self.assertEqual('_simple_link_matcher', crawler._get_link_matcher( |
|---|
| 314 | n/a | "http://example.org/some/file").__name__) |
|---|
| 315 | n/a | self.assertEqual('_default_link_matcher', crawler._get_link_matcher( |
|---|
| 316 | n/a | "http://other-url").__name__) |
|---|
| 317 | n/a | |
|---|
| 318 | n/a | def test_default_link_matcher(self): |
|---|
| 319 | n/a | crawler = Crawler("http://example.org", mirrors=[]) |
|---|
| 320 | n/a | crawler.follow_externals = True |
|---|
| 321 | n/a | crawler._is_browsable = lambda *args: True |
|---|
| 322 | n/a | base_url = "http://example.org/some/file/" |
|---|
| 323 | n/a | content = """ |
|---|
| 324 | n/a | <a href="../homepage" rel="homepage">link</a> |
|---|
| 325 | n/a | <a href="../download" rel="download">link2</a> |
|---|
| 326 | n/a | <a href="../simpleurl">link2</a> |
|---|
| 327 | n/a | """ |
|---|
| 328 | n/a | found_links = set(uri for uri, _ in |
|---|
| 329 | n/a | crawler._default_link_matcher(content, base_url)) |
|---|
| 330 | n/a | self.assertIn('http://example.org/some/homepage', found_links) |
|---|
| 331 | n/a | self.assertIn('http://example.org/some/simpleurl', found_links) |
|---|
| 332 | n/a | self.assertIn('http://example.org/some/download', found_links) |
|---|
| 333 | n/a | |
|---|
| 334 | n/a | @unittest.skipIf(_thread is None, 'needs threads') |
|---|
| 335 | n/a | @use_pypi_server("project_list") |
|---|
| 336 | n/a | def test_search_projects(self, server): |
|---|
| 337 | n/a | # we can search the index for some projects, on their names |
|---|
| 338 | n/a | # the case used no matters here |
|---|
| 339 | n/a | crawler = self._get_simple_crawler(server) |
|---|
| 340 | n/a | tests = (('Foobar', ['FooBar-bar', 'Foobar-baz', 'Baz-FooBar']), |
|---|
| 341 | n/a | ('foobar*', ['FooBar-bar', 'Foobar-baz']), |
|---|
| 342 | n/a | ('*foobar', ['Baz-FooBar'])) |
|---|
| 343 | n/a | |
|---|
| 344 | n/a | for search, expected in tests: |
|---|
| 345 | n/a | projects = [p.name for p in crawler.search_projects(search)] |
|---|
| 346 | n/a | self.assertListEqual(expected, projects) |
|---|
| 347 | n/a | |
|---|
| 348 | n/a | |
|---|
| 349 | n/a | def test_suite(): |
|---|
| 350 | n/a | return unittest.makeSuite(SimpleCrawlerTestCase) |
|---|
| 351 | n/a | |
|---|
| 352 | n/a | if __name__ == '__main__': |
|---|
| 353 | n/a | unittest.main(defaultTest="test_suite") |
|---|
