| 1 |
import io |
| 2 |
import os |
| 3 |
import threading |
| 4 |
import unittest |
| 5 |
import urllib.robotparser |
| 6 |
from test import support |
| 7 |
from test.support import socket_helper |
| 8 |
from test.support import threading_helper |
| 9 |
from http.server import BaseHTTPRequestHandler, HTTPServer |
| 10 |
|
| 11 |
|
| 12 |
class BaseRobotTest: |
| 13 |
robots_txt = '' |
| 14 |
agent = 'test_robotparser' |
| 15 |
good = [] |
| 16 |
bad = [] |
| 17 |
site_maps = None |
| 18 |
expected_output = None |
| 19 |
|
| 20 |
def __init_subclass__(cls): |
| 21 |
super().__init_subclass__() |
| 22 |
# Remove tests that do nothing. |
| 23 |
if issubclass(cls, unittest.TestCase): |
| 24 |
if not cls.good: |
| 25 |
cls.test_good_urls = None |
| 26 |
if not cls.bad: |
| 27 |
cls.test_bad_urls = None |
| 28 |
if cls.expected_output is None: |
| 29 |
cls.test_string_formatting = None |
| 30 |
|
| 31 |
def setUp(self): |
| 32 |
lines = io.StringIO(self.robots_txt).readlines() |
| 33 |
self.parser = urllib.robotparser.RobotFileParser() |
| 34 |
self.parser.parse(lines) |
| 35 |
|
| 36 |
def get_agent_and_url(self, url): |
| 37 |
if isinstance(url, tuple): |
| 38 |
agent, url = url |
| 39 |
return agent, url |
| 40 |
return self.agent, url |
| 41 |
|
| 42 |
def test_good_urls(self): |
| 43 |
for url in self.good: |
| 44 |
agent, url = self.get_agent_and_url(url) |
| 45 |
with self.subTest(url=url, agent=agent): |
| 46 |
self.assertTrue(self.parser.can_fetch(agent, url)) |
| 47 |
|
| 48 |
def test_bad_urls(self): |
| 49 |
for url in self.bad: |
| 50 |
agent, url = self.get_agent_and_url(url) |
| 51 |
with self.subTest(url=url, agent=agent): |
| 52 |
self.assertFalse(self.parser.can_fetch(agent, url)) |
| 53 |
|
| 54 |
def test_site_maps(self): |
| 55 |
self.assertEqual(self.parser.site_maps(), self.site_maps) |
| 56 |
|
| 57 |
def test_string_formatting(self): |
| 58 |
self.assertEqual(str(self.parser), self.expected_output) |
| 59 |
|
| 60 |
class UserAgentWildcardTest(BaseRobotTest, unittest.TestCase): |
| 61 |
robots_txt = """\ |
| 62 |
User-agent: * |
| 63 |
Disallow: /cyberworld/map/ # This is an infinite virtual URL space |
| 64 |
Disallow: /tmp/ # these will soon disappear |
| 65 |
Disallow: /foo.html |
| 66 |
""" |
| 67 |
good = ['/', '/test.html'] |
| 68 |
bad = ['/cyberworld/map/index.html', '/tmp/xxx', '/foo.html'] |
| 69 |
|
| 70 |
class SimpleExampleTest(BaseRobotTest, unittest.TestCase): |
| 71 |
# Example from RFC 9309, section 5.1. |
| 72 |
robots_txt = """\ |
| 73 |
User-Agent: * |
| 74 |
Disallow: *.gif$ |
| 75 |
Disallow: /example/ |
| 76 |
Allow: /publications/ |
| 77 |
|
| 78 |
User-Agent: foobot |
| 79 |
Disallow:/ |
| 80 |
Allow:/example/page.html |
| 81 |
Allow:/example/allowed.gif |
| 82 |
|
| 83 |
User-Agent: barbot |
| 84 |
User-Agent: bazbot |
| 85 |
Disallow: /example/page.html |
| 86 |
|
| 87 |
User-Agent: quxbot |
| 88 |
""" |
| 89 |
good = [ |
| 90 |
'/', '/publications/', |
| 91 |
('foobot', '/example/page.html'), ('foobot', '/example/allowed.gif'), |
| 92 |
('barbot', '/'), ('barbot', '/example/'), |
| 93 |
('barbot', '/example/allowed.gif'), |
| 94 |
('barbot', '/example/disallowed.gif'), |
| 95 |
('barbot', '/publications/'), |
| 96 |
('barbot', '/publications/allowed.gif'), |
| 97 |
('bazbot', '/'), ('bazbot', '/example/'), |
| 98 |
('bazbot', '/example/allowed.gif'), |
| 99 |
('bazbot', '/example/disallowed.gif'), |
| 100 |
('bazbot', '/publications/'), |
| 101 |
('bazbot', '/publications/allowed.gif'), |
| 102 |
('quxbot', '/'), ('quxbot', '/example/'), |
| 103 |
('quxbot', '/example/page.html'), ('quxbot', '/example/allowed.gif'), |
| 104 |
('quxbot', '/example/disallowed.gif'), |
| 105 |
('quxbot', '/publications/'), |
| 106 |
('quxbot', '/publications/allowed.gif'), |
| 107 |
] |
| 108 |
bad = [ |
| 109 |
'/example/', '/example/page.html', '/example/allowed.gif', |
| 110 |
'/example/disallowed.gif', |
| 111 |
'/publications/allowed.gif', |
| 112 |
('foobot', '/'), ('foobot', '/example/'), |
| 113 |
('foobot', '/example/disallowed.gif'), |
| 114 |
('foobot', '/publications/'), |
| 115 |
('foobot', '/publications/allowed.gif'), |
| 116 |
('barbot', '/example/page.html'), |
| 117 |
('bazbot', '/example/page.html'), |
| 118 |
] |
| 119 |
|
| 120 |
|
| 121 |
class CrawlDelayAndCustomAgentTest(BaseRobotTest, unittest.TestCase): |
| 122 |
robots_txt = """\ |
| 123 |
# robots.txt for http://www.example.com/ |
| 124 |
|
| 125 |
User-agent: * |
| 126 |
Crawl-delay: 1 |
| 127 |
Request-rate: 3/15 |
| 128 |
Disallow: /cyberworld/map/ # This is an infinite virtual URL space |
| 129 |
|
| 130 |
# Cybermapper knows where to go. |
| 131 |
User-agent: cybermapper |
| 132 |
Disallow: |
| 133 |
""" |
| 134 |
good = ['/', '/test.html', ('cybermapper', '/cyberworld/map/index.html')] |
| 135 |
bad = ['/cyberworld/map/index.html'] |
| 136 |
|
| 137 |
|
| 138 |
class SitemapTest(BaseRobotTest, unittest.TestCase): |
| 139 |
robots_txt = """\ |
| 140 |
# robots.txt for http://www.example.com/ |
| 141 |
|
| 142 |
User-agent: * |
| 143 |
Sitemap: http://www.gstatic.com/s2/sitemaps/profiles-sitemap.xml |
| 144 |
Sitemap: http://www.google.com/hostednews/sitemap_index.xml |
| 145 |
Request-rate: 3/15 |
| 146 |
Disallow: /cyberworld/map/ # This is an infinite virtual URL space |
| 147 |
|
| 148 |
""" |
| 149 |
good = ['/', '/test.html'] |
| 150 |
bad = ['/cyberworld/map/index.html'] |
| 151 |
site_maps = ['http://www.gstatic.com/s2/sitemaps/profiles-sitemap.xml', |
| 152 |
'http://www.google.com/hostednews/sitemap_index.xml'] |
| 153 |
|
| 154 |
|
| 155 |
class RejectAllRobotsTest(BaseRobotTest, unittest.TestCase): |
| 156 |
robots_txt = """\ |
| 157 |
# go away |
| 158 |
User-agent: * |
| 159 |
Disallow: / |
| 160 |
""" |
| 161 |
good = ['/robots.txt'] |
| 162 |
bad = ['/cyberworld/map/index.html', '/', '/tmp/'] |
| 163 |
|
| 164 |
|
| 165 |
class BaseRequestRateTest(BaseRobotTest): |
| 166 |
request_rate = None |
| 167 |
crawl_delay = None |
| 168 |
|
| 169 |
def test_request_rate(self): |
| 170 |
parser = self.parser |
| 171 |
for url in self.good + self.bad: |
| 172 |
agent, url = self.get_agent_and_url(url) |
| 173 |
with self.subTest(url=url, agent=agent): |
| 174 |
self.assertEqual(parser.crawl_delay(agent), self.crawl_delay) |
| 175 |
|
| 176 |
parsed_request_rate = parser.request_rate(agent) |
| 177 |
self.assertEqual(parsed_request_rate, self.request_rate) |
| 178 |
if self.request_rate is not None: |
| 179 |
self.assertIsInstance( |
| 180 |
parsed_request_rate, |
| 181 |
urllib.robotparser.RequestRate |
| 182 |
) |
| 183 |
self.assertEqual( |
| 184 |
parsed_request_rate.requests, |
| 185 |
self.request_rate.requests |
| 186 |
) |
| 187 |
self.assertEqual( |
| 188 |
parsed_request_rate.seconds, |
| 189 |
self.request_rate.seconds |
| 190 |
) |
| 191 |
|
| 192 |
|
| 193 |
class EmptyFileTest(BaseRequestRateTest, unittest.TestCase): |
| 194 |
robots_txt = '' |
| 195 |
good = ['/foo'] |
| 196 |
expected_output = '' |
| 197 |
|
| 198 |
|
| 199 |
class CrawlDelayAndRequestRateTest(BaseRequestRateTest, unittest.TestCase): |
| 200 |
robots_txt = """\ |
| 201 |
User-agent: figtree |
| 202 |
Crawl-delay: 3 |
| 203 |
Request-rate: 9/30 |
| 204 |
Disallow: /tmp |
| 205 |
Disallow: /a%3cd.html |
| 206 |
Disallow: /a%2fb.html |
| 207 |
Disallow: /%7ejoe/index.html |
| 208 |
""" |
| 209 |
agent = 'figtree' |
| 210 |
request_rate = urllib.robotparser.RequestRate(9, 30) |
| 211 |
crawl_delay = 3 |
| 212 |
good = [('figtree', '/foo.html')] |
| 213 |
bad = ['/tmp', '/tmp.html', '/tmp/a.html', '/a%3cd.html', '/a%3Cd.html', |
| 214 |
'/a%2fb.html', '/~joe/index.html'] |
| 215 |
|
| 216 |
|
| 217 |
class DifferentAgentTest(CrawlDelayAndRequestRateTest): |
| 218 |
agent = 'FigTree Robot libwww-perl/5.04' |
| 219 |
|
| 220 |
|
| 221 |
class InvalidRequestRateTest(BaseRobotTest, unittest.TestCase): |
| 222 |
robots_txt = """\ |
| 223 |
User-agent: * |
| 224 |
Disallow: /tmp/ |
| 225 |
Disallow: /a%3Cd.html |
| 226 |
Disallow: /a/b.html |
| 227 |
Disallow: /%7ejoe/index.html |
| 228 |
Crawl-delay: 3 |
| 229 |
Request-rate: 9/banana |
| 230 |
""" |
| 231 |
good = ['/tmp'] |
| 232 |
bad = ['/tmp/', '/tmp/a.html', '/a%3cd.html', '/a%3Cd.html', '/a/b.html', |
| 233 |
'/%7Ejoe/index.html'] |
| 234 |
crawl_delay = 3 |
| 235 |
|
| 236 |
|
| 237 |
class InvalidCrawlDelayTest(BaseRobotTest, unittest.TestCase): |
| 238 |
# From bug report #523041 |
| 239 |
robots_txt = """\ |
| 240 |
User-Agent: * |
| 241 |
Disallow: /. |
| 242 |
Crawl-delay: pears |
| 243 |
""" |
| 244 |
good = ['/foo.html'] |
| 245 |
# bug report says "/" should be denied, but that is not in the RFC |
| 246 |
bad = [] |
| 247 |
|
| 248 |
|
| 249 |
class AnotherInvalidRequestRateTest(BaseRobotTest, unittest.TestCase): |
| 250 |
# also test that Allow and Diasallow works well with each other |
| 251 |
robots_txt = """\ |
| 252 |
User-agent: Googlebot |
| 253 |
Allow: /folder1/myfile.html |
| 254 |
Disallow: /folder1/ |
| 255 |
Request-rate: whale/banana |
| 256 |
""" |
| 257 |
agent = 'Googlebot' |
| 258 |
good = ['/folder1/myfile.html'] |
| 259 |
bad = ['/folder1/anotherfile.html'] |
| 260 |
|
| 261 |
|
| 262 |
class UserAgentOrderingTest(BaseRobotTest, unittest.TestCase): |
| 263 |
# the order of User-agent should not matter |
| 264 |
robots_txt = """\ |
| 265 |
User-agent: Googlebot |
| 266 |
Disallow: / |
| 267 |
Allow: /folder1/ |
| 268 |
|
| 269 |
User-agent: Googlebot-Mobile |
| 270 |
Allow: / |
| 271 |
Disallow: /folder1/ |
| 272 |
""" |
| 273 |
agent = 'Googlebot' |
| 274 |
bad = ['/something.jpg'] |
| 275 |
good = ['/folder1/myfile.html'] |
| 276 |
|
| 277 |
|
| 278 |
class UserAgentGoogleMobileTest(UserAgentOrderingTest): |
| 279 |
agent = 'Googlebot-mobile' |
| 280 |
bad = ['/folder1/myfile.html'] |
| 281 |
good = ['/something.jpg'] |
| 282 |
|
| 283 |
|
| 284 |
class LongestMatchTest(BaseRobotTest, unittest.TestCase): |
| 285 |
# Based on example from RFC 9309, section 5.2. |
| 286 |
robots_txt = """\ |
| 287 |
User-agent: * |
| 288 |
Allow: /example/page/ |
| 289 |
Disallow: /example/page/disallowed.gif |
| 290 |
Allow: /example/ |
| 291 |
""" |
| 292 |
good = ['/example/', '/example/page/'] |
| 293 |
bad = ['/example/page/disallowed.gif'] |
| 294 |
|
| 295 |
|
| 296 |
class LongestMatchWildcardTest(BaseRobotTest, unittest.TestCase): |
| 297 |
robots_txt = """\ |
| 298 |
User-agent: * |
| 299 |
Allow: /example/page/ |
| 300 |
Disallow: *.gif |
| 301 |
Allow: /example/ |
| 302 |
""" |
| 303 |
good = ['/example/', '/example/page/'] |
| 304 |
bad = ['/example/page/disallowed.gif', '/x.gif'] |
| 305 |
|
| 306 |
|
| 307 |
class AllowWinsEqualMatchTest(BaseRobotTest, unittest.TestCase): |
| 308 |
robots_txt = """\ |
| 309 |
User-agent: * |
| 310 |
Disallow: /spam |
| 311 |
Allow: /spam |
| 312 |
Disallow: /spam |
| 313 |
""" |
| 314 |
good = ['/spam', '/spam/'] |
| 315 |
|
| 316 |
|
| 317 |
class AllowWinsEqualFullMatchTest(BaseRobotTest, unittest.TestCase): |
| 318 |
robots_txt = """\ |
| 319 |
User-agent: * |
| 320 |
Disallow: /spam |
| 321 |
Allow: /spam$ |
| 322 |
Disallow: /spam |
| 323 |
Disallow: /eggs$ |
| 324 |
Allow: /eggs |
| 325 |
Disallow: /eggs$ |
| 326 |
""" |
| 327 |
good = ['/spam', '/eggs', '/eggs/'] |
| 328 |
bad = ['/spam/'] |
| 329 |
|
| 330 |
|
| 331 |
class AllowWinsEqualMatchWildcardTest(BaseRobotTest, unittest.TestCase): |
| 332 |
robots_txt = """\ |
| 333 |
User-agent: * |
| 334 |
Disallow: /spam |
| 335 |
Allow: *am |
| 336 |
Disallow: /spam |
| 337 |
Disallow: *gs |
| 338 |
Allow: /eggs |
| 339 |
Disallow: *gs |
| 340 |
""" |
| 341 |
good = ['/spam', '/eggs', '/spam/', '/eggs/'] |
| 342 |
|
| 343 |
|
| 344 |
class MergeGroupsTest(BaseRobotTest, unittest.TestCase): |
| 345 |
robots_txt = """\ |
| 346 |
User-agent: spambot |
| 347 |
Disallow: /some/path |
| 348 |
|
| 349 |
User-agent: spambot |
| 350 |
Disallow: /another/path |
| 351 |
""" |
| 352 |
agent = 'spambot' |
| 353 |
bad = ['/some/path', '/another/path'] |
| 354 |
|
| 355 |
|
| 356 |
class UserAgentStartsGroupTest(BaseRobotTest, unittest.TestCase): |
| 357 |
robots_txt = """\ |
| 358 |
User-agent: spambot |
| 359 |
Disallow: /some/path |
| 360 |
User-agent: eggsbot |
| 361 |
Disallow: /another/path |
| 362 |
""" |
| 363 |
good = [('spambot', '/'), ('spambot', '/another/path'), |
| 364 |
('eggsbot', '/'), ('eggsbot', '/some/path')] |
| 365 |
bad = [('spambot', '/some/path'), ('eggsbot', '/another/path')] |
| 366 |
expected_output = """\ |
| 367 |
User-agent: spambot |
| 368 |
Disallow: /some/path |
| 369 |
|
| 370 |
User-agent: eggsbot |
| 371 |
Disallow: /another/path\ |
| 372 |
""" |
| 373 |
|
| 374 |
class IgnoreEmptyLinesTest(BaseRobotTest, unittest.TestCase): |
| 375 |
robots_txt = """\ |
| 376 |
User-agent: spambot |
| 377 |
|
| 378 |
User-agent: eggsbot |
| 379 |
Disallow: /some/path |
| 380 |
|
| 381 |
Disallow: /another/path |
| 382 |
""" |
| 383 |
good = [('spambot', '/'), ('eggsbot', '/')] |
| 384 |
bad = [ |
| 385 |
('spambot', '/some/path'), ('spambot', '/another/path'), |
| 386 |
('eggsbot', '/some/path'), ('eggsbot', '/another/path'), |
| 387 |
] |
| 388 |
expected_output = """\ |
| 389 |
User-agent: spambot |
| 390 |
User-agent: eggsbot |
| 391 |
Disallow: /some/path |
| 392 |
Disallow: /another/path\ |
| 393 |
""" |
| 394 |
|
| 395 |
|
| 396 |
class IgnoreRulesWithoutUserAgentTest(BaseRobotTest, unittest.TestCase): |
| 397 |
robots_txt = """\ |
| 398 |
Disallow: /some/path |
| 399 |
|
| 400 |
User-agent: * |
| 401 |
Disallow: /another/path |
| 402 |
""" |
| 403 |
good = ['/', '/some/path'] |
| 404 |
bad = ['/another/path'] |
| 405 |
expected_output = """\ |
| 406 |
User-agent: * |
| 407 |
Disallow: /another/path\ |
| 408 |
""" |
| 409 |
|
| 410 |
|
| 411 |
class EmptyGroupTest(BaseRobotTest, unittest.TestCase): |
| 412 |
robots_txt = """\ |
| 413 |
User-agent: * |
| 414 |
Disallow: /some/path |
| 415 |
|
| 416 |
User-agent: spambot |
| 417 |
""" |
| 418 |
agent = 'spambot' |
| 419 |
good = ['/', '/some/path'] |
| 420 |
expected_output = """\ |
| 421 |
User-agent: * |
| 422 |
Disallow: /some/path |
| 423 |
|
| 424 |
User-agent: spambot |
| 425 |
Allow:\ |
| 426 |
""" |
| 427 |
|
| 428 |
|
| 429 |
class WeirdPathTest(BaseRobotTest, unittest.TestCase): |
| 430 |
robots_txt = f"""\ |
| 431 |
User-agent: * |
| 432 |
Disallow: /a$$$ |
| 433 |
Disallow: /b$z |
| 434 |
Disallow: /c*** |
| 435 |
Disallow: /d***z |
| 436 |
Disallow: /e*$**$$ |
| 437 |
Disallow: /f*$**$$z |
| 438 |
Disallow: /g$*$$** |
| 439 |
Disallow: /h$*$$**z |
| 440 |
""" |
| 441 |
good = ['/ax', '/a$$', '/b', '/bz', '/b$z', '/d', '/f', '/fz', |
| 442 |
'/f$$$z', '/fx$y$$z', '/gx', '/g$$$', '/g$x$$y', '/h', '/hz', |
| 443 |
'/h$$$z', '/h$x$$yz'] |
| 444 |
bad = ['/a', '/c', '/cxy', '/dz', '/dxyz', '/dxzy', '/e', '/exy', |
| 445 |
'/e$$', '/ex$y$', '/g'] |
| 446 |
expected_output = """\ |
| 447 |
User-agent: * |
| 448 |
Disallow: /a$ |
| 449 |
Disallow: /c* |
| 450 |
Disallow: /d*z |
| 451 |
Disallow: /e*$ |
| 452 |
Disallow: /g$\ |
| 453 |
""" |
| 454 |
|
| 455 |
|
| 456 |
class PathWithManyWildcardsTest(BaseRobotTest, unittest.TestCase): |
| 457 |
# This test would take many years if use naive translation to regular |
| 458 |
# expression (* -> .*). |
| 459 |
N = 50 |
| 460 |
robots_txt = f"""\ |
| 461 |
User-agent: * |
| 462 |
Disallow: /{'*a'*N}*b |
| 463 |
""" |
| 464 |
good = ['/' + 'a'*N + 'a'] |
| 465 |
bad = ['/' + 'a'*N + 'b'] |
| 466 |
|
| 467 |
|
| 468 |
class DisallowQueryStringTest(BaseRobotTest, unittest.TestCase): |
| 469 |
# see issue #6325 for details |
| 470 |
robots_txt = """\ |
| 471 |
User-agent: * |
| 472 |
Disallow: /some/path?name=value |
| 473 |
Disallow: /another/path? |
| 474 |
Disallow: /yet/one/path?name=value&more |
| 475 |
""" |
| 476 |
good = ['/some/path', '/some/path?', |
| 477 |
'/some/path%3Fname=value', '/some/path?name%3Dvalue', |
| 478 |
'/another/path', '/another/path%3F', |
| 479 |
'/yet/one/path?name=value%26more', |
| 480 |
'/some/pathxname=value'] |
| 481 |
bad = ['/some/path?name=value' |
| 482 |
'/another/path?', '/another/path?name=value', |
| 483 |
'/yet/one/path?name=value&more'] |
| 484 |
|
| 485 |
|
| 486 |
class PercentEncodingTest(BaseRobotTest, unittest.TestCase): |
| 487 |
robots_txt = """\ |
| 488 |
User-agent: * |
| 489 |
Disallow: /a1/Z-._~ # unreserved characters |
| 490 |
Disallow: /a2/%5A%2D%2E%5F%7E # percent-encoded unreserved characters |
| 491 |
Disallow: /u1/%F0%9F%90%8D # percent-encoded ASCII Unicode character |
| 492 |
Disallow: /u2/%f0%9f%90%8d |
| 493 |
Disallow: /u3/\U0001f40d # raw non-ASCII Unicode character |
| 494 |
Disallow: /v1/%F0 # percent-encoded non-ASCII octet |
| 495 |
Disallow: /v2/%f0 |
| 496 |
Disallow: /v3/\udcf0 # raw non-ASCII octet |
| 497 |
Disallow: /p1%xy # raw percent |
| 498 |
Disallow: /p2% |
| 499 |
Disallow: /p3%25xy # percent-encoded percent |
| 500 |
Disallow: /p4%2525xy # double percent-encoded percent |
| 501 |
Disallow: /john%20smith # space |
| 502 |
Disallow: /john doe |
| 503 |
Disallow: /trailingspace%20 |
| 504 |
Disallow: /question%3Fq=v # not query |
| 505 |
Disallow: /hash%23f # not fragment |
| 506 |
Disallow: /dollar%24 |
| 507 |
Disallow: /asterisk%2A |
| 508 |
Disallow: /sub/dir |
| 509 |
Disallow: /slash%2F |
| 510 |
Disallow: /query/question?q=%3F |
| 511 |
Disallow: /query/raw/question?q=? |
| 512 |
Disallow: /query/eq?q%3Dv |
| 513 |
Disallow: /query/amp?q=v%26a |
| 514 |
""" |
| 515 |
good = [ |
| 516 |
'/u1/%F0', '/u1/%f0', |
| 517 |
'/u2/%F0', '/u2/%f0', |
| 518 |
'/u3/%F0', '/u3/%f0', |
| 519 |
'/p1%2525xy', '/p2%f0', '/p3%2525xy', '/p4%xy', '/p4%25xy', |
| 520 |
'/question?q=v', |
| 521 |
'/dollar', '/asterisk', |
| 522 |
'/query/eq?q=v', |
| 523 |
'/query/amp?q=v&a', |
| 524 |
] |
| 525 |
bad = [ |
| 526 |
'/a1/Z-._~', '/a1/%5A%2D%2E%5F%7E', |
| 527 |
'/a2/Z-._~', '/a2/%5A%2D%2E%5F%7E', |
| 528 |
'/u1/%F0%9F%90%8D', '/u1/%f0%9f%90%8d', '/u1/\U0001f40d', |
| 529 |
'/u2/%F0%9F%90%8D', '/u2/%f0%9f%90%8d', '/u2/\U0001f40d', |
| 530 |
'/u3/%F0%9F%90%8D', '/u3/%f0%9f%90%8d', '/u3/\U0001f40d', |
| 531 |
'/v1/%F0', '/v1/%f0', '/v1/\udcf0', '/v1/\U0001f40d', |
| 532 |
'/v2/%F0', '/v2/%f0', '/v2/\udcf0', '/v2/\U0001f40d', |
| 533 |
'/v3/%F0', '/v3/%f0', '/v3/\udcf0', '/v3/\U0001f40d', |
| 534 |
'/p1%xy', '/p1%25xy', |
| 535 |
'/p2%', '/p2%25', '/p2%2525', '/p2%xy', |
| 536 |
'/p3%xy', '/p3%25xy', |
| 537 |
'/p4%2525xy', |
| 538 |
'/john%20smith', '/john smith', |
| 539 |
'/john%20doe', '/john doe', |
| 540 |
'/trailingspace%20', '/trailingspace ', |
| 541 |
'/question%3Fq=v', |
| 542 |
'/hash#f', '/hash%23f', |
| 543 |
'/dollar$', '/dollar%24', |
| 544 |
'/asterisk*', '/asterisk%2A', |
| 545 |
'/sub/dir', '/sub%2Fdir', |
| 546 |
'/slash%2F', '/slash/', |
| 547 |
'/query/question?q=?', '/query/question?q=%3F', |
| 548 |
'/query/raw/question?q=?', '/query/raw/question?q=%3F', |
| 549 |
'/query/eq?q%3Dv', |
| 550 |
'/query/amp?q=v%26a', |
| 551 |
] |
| 552 |
# other reserved characters |
| 553 |
for c in ":/#[]@!$&'()*+,;=": |
| 554 |
robots_txt += f'Disallow: /raw{c}\nDisallow: /pc%{ord(c):02X}\n' |
| 555 |
bad.append(f'/raw{c}') |
| 556 |
bad.append(f'/raw%{ord(c):02X}') |
| 557 |
bad.append(f'/pc{c}') |
| 558 |
bad.append(f'/pc%{ord(c):02X}') |
| 559 |
|
| 560 |
|
| 561 |
class DefaultEntryTest(BaseRequestRateTest, unittest.TestCase): |
| 562 |
robots_txt = """\ |
| 563 |
User-agent: * |
| 564 |
Crawl-delay: 1 |
| 565 |
Request-rate: 3/15 |
| 566 |
Disallow: /cyberworld/map/ |
| 567 |
""" |
| 568 |
request_rate = urllib.robotparser.RequestRate(3, 15) |
| 569 |
crawl_delay = 1 |
| 570 |
good = ['/', '/test.html'] |
| 571 |
bad = ['/cyberworld/map/index.html'] |
| 572 |
|
| 573 |
|
| 574 |
class StringFormattingTest(BaseRobotTest, unittest.TestCase): |
| 575 |
robots_txt = """\ |
| 576 |
User-agent: * |
| 577 |
Crawl-delay: 1 |
| 578 |
Request-rate: 3/15 |
| 579 |
Disallow: /cyberworld/map/ # This is an infinite virtual URL space |
| 580 |
|
| 581 |
# Cybermapper knows where to go. |
| 582 |
User-agent: cybermapper |
| 583 |
Disallow: /some/path |
| 584 |
""" |
| 585 |
|
| 586 |
expected_output = """\ |
| 587 |
User-agent: * |
| 588 |
Crawl-delay: 1 |
| 589 |
Request-rate: 3/15 |
| 590 |
Disallow: /cyberworld/map/ |
| 591 |
|
| 592 |
User-agent: cybermapper |
| 593 |
Disallow: /some/path\ |
| 594 |
""" |
| 595 |
|
| 596 |
|
| 597 |
class ConstructedStringFormattingTest(unittest.TestCase): |
| 598 |
def test_empty(self): |
| 599 |
parser = urllib.robotparser.RobotFileParser() |
| 600 |
self.assertEqual(str(parser), '') |
| 601 |
|
| 602 |
def test_group_without_rules(self): |
| 603 |
parser = urllib.robotparser.RobotFileParser() |
| 604 |
entry = urllib.robotparser.Entry() |
| 605 |
entry.useragents = ['spambot'] |
| 606 |
parser._add_entry(entry) |
| 607 |
entry = urllib.robotparser.Entry() |
| 608 |
entry.useragents = ['hambot'] |
| 609 |
entry.rulelines = [urllib.robotparser.RuleLine('/ham', False)] |
| 610 |
parser._add_entry(entry) |
| 611 |
entry = urllib.robotparser.Entry() |
| 612 |
entry.useragents = ['eggsbot'] |
| 613 |
parser._add_entry(entry) |
| 614 |
self.assertEqual(str(parser), """\ |
| 615 |
User-agent: spambot |
| 616 |
Allow: |
| 617 |
|
| 618 |
User-agent: hambot |
| 619 |
Disallow: /ham |
| 620 |
|
| 621 |
User-agent: eggsbot |
| 622 |
Allow:\ |
| 623 |
""") |
| 624 |
|
| 625 |
def test_group_without_user_agent(self): |
| 626 |
parser = urllib.robotparser.RobotFileParser() |
| 627 |
entry = urllib.robotparser.Entry() |
| 628 |
entry.rulelines = [urllib.robotparser.RuleLine('/ham', False)] |
| 629 |
parser._add_entry(entry) |
| 630 |
entry = urllib.robotparser.Entry() |
| 631 |
entry.useragents = ['spambot'] |
| 632 |
entry.rulelines = [urllib.robotparser.RuleLine('/spam', False)] |
| 633 |
parser._add_entry(entry) |
| 634 |
entry = urllib.robotparser.Entry() |
| 635 |
entry.rulelines = [urllib.robotparser.RuleLine('/eggs', False)] |
| 636 |
parser._add_entry(entry) |
| 637 |
self.assertEqual(str(parser), """\ |
| 638 |
User-agent: spambot |
| 639 |
Disallow: /spam\ |
| 640 |
""") |
| 641 |
|
| 642 |
|
| 643 |
@unittest.skipUnless( |
| 644 |
support.has_socket_support, |
| 645 |
"Socket server requires working socket." |
| 646 |
) |
| 647 |
class BaseLocalNetworkTestCase: |
| 648 |
|
| 649 |
@classmethod |
| 650 |
def setUpClass(cls): |
| 651 |
# clear _opener global variable |
| 652 |
cls.addClassCleanup(urllib.request.urlcleanup) |
| 653 |
|
| 654 |
cls.server = HTTPServer((socket_helper.HOST, 0), cls.RobotHandler) |
| 655 |
cls.addClassCleanup(cls.server.server_close) |
| 656 |
|
| 657 |
t = threading.Thread( |
| 658 |
name='HTTPServer serving', |
| 659 |
target=cls.server.serve_forever, |
| 660 |
# Short poll interval to make the test finish quickly. |
| 661 |
# Time between requests is short enough that we won't wake |
| 662 |
# up spuriously too many times. |
| 663 |
kwargs={'poll_interval':0.01}) |
| 664 |
cls.enterClassContext(threading_helper.start_threads([t])) |
| 665 |
cls.addClassCleanup(cls.server.shutdown) |
| 666 |
|
| 667 |
|
| 668 |
SAMPLE_ROBOTS_TXT = b'''\ |
| 669 |
User-agent: test_robotparser |
| 670 |
Disallow: /utf8/\xf0\x9f\x90\x8d |
| 671 |
Disallow: /non-utf8/\xf0 |
| 672 |
Disallow: //[spam]/path |
| 673 |
''' |
| 674 |
|
| 675 |
|
| 676 |
class LocalNetworkTestCase(BaseLocalNetworkTestCase, unittest.TestCase): |
| 677 |
class RobotHandler(BaseHTTPRequestHandler): |
| 678 |
|
| 679 |
def do_GET(self): |
| 680 |
self.send_response(200) |
| 681 |
self.end_headers() |
| 682 |
self.wfile.write(SAMPLE_ROBOTS_TXT) |
| 683 |
|
| 684 |
def log_message(self, format, *args): |
| 685 |
pass |
| 686 |
|
| 687 |
def testRead(self): |
| 688 |
# Test that reading a weird robots.txt doesn't fail. |
| 689 |
addr = self.server.server_address |
| 690 |
url = f'http://{socket_helper.HOST}:{addr[1]}' |
| 691 |
robots_url = url + '/robots.txt' |
| 692 |
parser = urllib.robotparser.RobotFileParser() |
| 693 |
parser.set_url(robots_url) |
| 694 |
parser.read() |
| 695 |
# And it can even interpret the weird paths in some reasonable way. |
| 696 |
agent = 'test_robotparser' |
| 697 |
self.assertTrue(parser.can_fetch(agent, robots_url)) |
| 698 |
self.assertTrue(parser.can_fetch(agent, url + '/utf8/')) |
| 699 |
self.assertFalse(parser.can_fetch(agent, url + '/utf8/\U0001f40d')) |
| 700 |
self.assertFalse(parser.can_fetch(agent, url + '/utf8/%F0%9F%90%8D')) |
| 701 |
self.assertTrue(parser.can_fetch(agent, url + '/non-utf8/')) |
| 702 |
self.assertFalse(parser.can_fetch(agent, url + '/non-utf8/%F0')) |
| 703 |
self.assertFalse(parser.can_fetch(agent, url + '/non-utf8/\U0001f40d')) |
| 704 |
self.assertFalse(parser.can_fetch(agent, url + '/%2F[spam]/path')) |
| 705 |
|
| 706 |
|
| 707 |
class HttpErrorsTestCase(BaseLocalNetworkTestCase, unittest.TestCase): |
| 708 |
class RobotHandler(BaseHTTPRequestHandler): |
| 709 |
|
| 710 |
def do_GET(self): |
| 711 |
self.send_error(self.server.return_code) |
| 712 |
|
| 713 |
def log_message(self, format, *args): |
| 714 |
pass |
| 715 |
|
| 716 |
def setUp(self): |
| 717 |
# Make sure that a valid code is set in the test. |
| 718 |
self.server.return_code = None |
| 719 |
|
| 720 |
def testUnauthorized(self): |
| 721 |
self.server.return_code = 401 |
| 722 |
addr = self.server.server_address |
| 723 |
url = f'http://{socket_helper.HOST}:{addr[1]}' |
| 724 |
robots_url = url + "/robots.txt" |
| 725 |
parser = urllib.robotparser.RobotFileParser() |
| 726 |
parser.set_url(url) |
| 727 |
parser.read() |
| 728 |
self.assertFalse(parser.can_fetch("*", robots_url)) |
| 729 |
self.assertFalse(parser.can_fetch("*", url + '/some/file.html')) |
| 730 |
|
| 731 |
def testForbidden(self): |
| 732 |
self.server.return_code = 403 |
| 733 |
addr = self.server.server_address |
| 734 |
url = f'http://{socket_helper.HOST}:{addr[1]}' |
| 735 |
robots_url = url + "/robots.txt" |
| 736 |
parser = urllib.robotparser.RobotFileParser() |
| 737 |
parser.set_url(url) |
| 738 |
parser.read() |
| 739 |
self.assertFalse(parser.can_fetch("*", robots_url)) |
| 740 |
self.assertFalse(parser.can_fetch("*", url + '/some/file.html')) |
| 741 |
|
| 742 |
def testNotFound(self): |
| 743 |
self.server.return_code = 404 |
| 744 |
addr = self.server.server_address |
| 745 |
url = f'http://{socket_helper.HOST}:{addr[1]}' |
| 746 |
robots_url = url + "/robots.txt" |
| 747 |
parser = urllib.robotparser.RobotFileParser() |
| 748 |
parser.set_url(url) |
| 749 |
parser.read() |
| 750 |
self.assertTrue(parser.can_fetch("*", robots_url)) |
| 751 |
self.assertTrue(parser.can_fetch("*", url + '/path/file.html')) |
| 752 |
|
| 753 |
def testTeapot(self): |
| 754 |
self.server.return_code = 418 |
| 755 |
addr = self.server.server_address |
| 756 |
url = f'http://{socket_helper.HOST}:{addr[1]}' |
| 757 |
robots_url = url + "/robots.txt" |
| 758 |
parser = urllib.robotparser.RobotFileParser() |
| 759 |
parser.set_url(url) |
| 760 |
parser.read() |
| 761 |
self.assertTrue(parser.can_fetch("*", robots_url)) |
| 762 |
self.assertTrue(parser.can_fetch("*", url + '/pot-1?milk-type=Cream')) |
| 763 |
|
| 764 |
def testServiceUnavailable(self): |
| 765 |
self.server.return_code = 503 |
| 766 |
addr = self.server.server_address |
| 767 |
url = f'http://{socket_helper.HOST}:{addr[1]}' |
| 768 |
robots_url = url + "/robots.txt" |
| 769 |
parser = urllib.robotparser.RobotFileParser() |
| 770 |
parser.set_url(url) |
| 771 |
parser.read() |
| 772 |
self.assertFalse(parser.can_fetch("*", robots_url)) |
| 773 |
self.assertFalse(parser.can_fetch("*", url + '/path/file.html')) |
| 774 |
|
| 775 |
|
| 776 |
@support.requires_working_socket() |
| 777 |
class NetworkTestCase(unittest.TestCase): |
| 778 |
|
| 779 |
base_url = 'http://www.pythontest.net/' |
| 780 |
robots_txt = '{}elsewhere/robots.txt'.format(base_url) |
| 781 |
|
| 782 |
@classmethod |
| 783 |
def setUpClass(cls): |
| 784 |
support.requires('network') |
| 785 |
cls.addClassCleanup(urllib.request.urlcleanup) |
| 786 |
with socket_helper.transient_internet(cls.base_url): |
| 787 |
cls.parser = urllib.robotparser.RobotFileParser(cls.robots_txt) |
| 788 |
cls.parser.read() |
| 789 |
|
| 790 |
def url(self, path): |
| 791 |
return '{}{}{}'.format( |
| 792 |
self.base_url, path, '/' if not os.path.splitext(path)[1] else '' |
| 793 |
) |
| 794 |
|
| 795 |
def test_basic(self): |
| 796 |
self.assertFalse(self.parser.disallow_all) |
| 797 |
self.assertFalse(self.parser.allow_all) |
| 798 |
self.assertGreater(self.parser.mtime(), 0) |
| 799 |
self.assertFalse(self.parser.crawl_delay('*')) |
| 800 |
self.assertFalse(self.parser.request_rate('*')) |
| 801 |
|
| 802 |
def test_can_fetch(self): |
| 803 |
self.assertTrue(self.parser.can_fetch('*', self.url('elsewhere'))) |
| 804 |
self.assertFalse(self.parser.can_fetch('Nutch', self.base_url)) |
| 805 |
self.assertTrue(self.parser.can_fetch('Nutch', self.url('brian'))) |
| 806 |
self.assertFalse(self.parser.can_fetch('Nutch', self.url('webstats'))) |
| 807 |
self.assertFalse(self.parser.can_fetch('*', self.url('webstats'))) |
| 808 |
self.assertTrue(self.parser.can_fetch('*', self.base_url)) |
| 809 |
|
| 810 |
def test_read_404(self): |
| 811 |
parser = urllib.robotparser.RobotFileParser(self.url('i-robot.txt')) |
| 812 |
parser.read() |
| 813 |
self.assertTrue(parser.allow_all) |
| 814 |
self.assertFalse(parser.disallow_all) |
| 815 |
self.assertEqual(parser.mtime(), 0) |
| 816 |
self.assertIsNone(parser.crawl_delay('*')) |
| 817 |
self.assertIsNone(parser.request_rate('*')) |
| 818 |
|
| 819 |
if __name__=='__main__': |
| 820 |
unittest.main() |