Issue #25400: RobotFileParser now correctly returns default values for crawl_delay...

Initial patch by Peter Wirtz.

diff --git a/Lib/test/test_robotparser.py b/Lib/test/test_robotparser.py
index 51b48ce53c91481e7ef5d5922f3f218b2b9eaaf4..0f64ba8b060defbbc129423f063f18d4aa2ac498 100644 (file)
--- a/Lib/test/test_robotparser.py
+++ b/Lib/test/test_robotparser.py
@@ -79,32 +79,17 @@ Disallow: /
     bad = ['/cyberworld/map/index.html', '/', '/tmp/']
 
 
-class CrawlDelayAndRequestRateTest(BaseRobotTest, unittest.TestCase):
-    robots_txt = """\
-User-agent: figtree
-Crawl-delay: 3
-Request-rate: 9/30
-Disallow: /tmp
-Disallow: /a%3cd.html
-Disallow: /a%2fb.html
-Disallow: /%7ejoe/index.html
-    """
-    agent = 'figtree'
-    request_rate = namedtuple('req_rate', 'requests seconds')(9, 30)
-    crawl_delay = 3
-    good = [('figtree', '/foo.html')]
-    bad = ['/tmp', '/tmp.html', '/tmp/a.html', '/a%3cd.html', '/a%3Cd.html',
-           '/a%2fb.html', '/~joe/index.html']
+class BaseRequestRateTest(BaseRobotTest):
 
     def test_request_rate(self):
-        for url in self.good:
+        for url in self.good + self.bad:
             agent, url = self.get_agent_and_url(url)
             with self.subTest(url=url, agent=agent):
                 if self.crawl_delay:
                     self.assertEqual(
                         self.parser.crawl_delay(agent), self.crawl_delay
                     )
-                if self.request_rate and self.parser.request_rate(agent):
+                if self.request_rate:
                     self.assertEqual(
                         self.parser.request_rate(agent).requests,
                         self.request_rate.requests
@@ -115,6 +100,24 @@ Disallow: /%7ejoe/index.html
                     )
 
 
+class CrawlDelayAndRequestRateTest(BaseRequestRateTest, unittest.TestCase):
+    robots_txt = """\
+User-agent: figtree
+Crawl-delay: 3
+Request-rate: 9/30
+Disallow: /tmp
+Disallow: /a%3cd.html
+Disallow: /a%2fb.html
+Disallow: /%7ejoe/index.html
+    """
+    agent = 'figtree'
+    request_rate = namedtuple('req_rate', 'requests seconds')(9, 30)
+    crawl_delay = 3
+    good = [('figtree', '/foo.html')]
+    bad = ['/tmp', '/tmp.html', '/tmp/a.html', '/a%3cd.html', '/a%3Cd.html',
+           '/a%2fb.html', '/~joe/index.html']
+
+
 class DifferentAgentTest(CrawlDelayAndRequestRateTest):
     agent = 'FigTree Robot libwww-perl/5.04'
     # these are not actually tested, but we still need to parse it
@@ -230,6 +233,19 @@ Disallow: /another/path?
     bad = ['/another/path?']
 
 
+class DefaultEntryTest(BaseRequestRateTest, unittest.TestCase):
+    robots_txt = """\
+User-agent: *
+Crawl-delay: 1
+Request-rate: 3/15
+Disallow: /cyberworld/map/
+    """
+    request_rate = namedtuple('req_rate', 'requests seconds')(3, 15)
+    crawl_delay = 1
+    good = ['/', '/test.html']
+    bad = ['/cyberworld/map/index.html']
+
+
 class RobotHandler(BaseHTTPRequestHandler):
 
     def do_GET(self):
@@ -309,6 +325,8 @@ class NetworkTestCase(unittest.TestCase):
         self.assertTrue(parser.allow_all)
         self.assertFalse(parser.disallow_all)
         self.assertEqual(parser.mtime(), 0)
+        self.assertIsNone(parser.crawl_delay('*'))
+        self.assertIsNone(parser.request_rate('*'))
 
 if __name__=='__main__':
     unittest.main()

diff --git a/Lib/urllib/robotparser.py b/Lib/urllib/robotparser.py
index 85add1624aed31dfc3f0a39eca6836109308832e..9dab4c1c3a8880c5d306030610d9accad528bf4e 100644 (file)
--- a/Lib/urllib/robotparser.py
+++ b/Lib/urllib/robotparser.py
@@ -175,16 +175,20 @@ class RobotFileParser:
         return True
 
     def crawl_delay(self, useragent):
+        if not self.mtime():
+            return None
         for entry in self.entries:
             if entry.applies_to(useragent):
                 return entry.delay
-        return None
+        return self.default_entry.delay
 
     def request_rate(self, useragent):
+        if not self.mtime():
+            return None
         for entry in self.entries:
             if entry.applies_to(useragent):
                 return entry.req_rate
-        return None
+        return self.default_entry.req_rate
 
     def __str__(self):
         return ''.join([str(entry) + "\n" for entry in self.entries])

diff --git a/Misc/NEWS b/Misc/NEWS
index 671a9b41c96bf21476fe4bcfe4f8850b86f6fc2e..e26a5c05abae1f3c061a59c5f75e7df1723bb375 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -29,6 +29,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #25400: RobotFileParser now correctly returns default values for
+  crawl_delay and request_rate.  Initial patch by Peter Wirtz.
+
 - Issue #27932: Prevent memory leak in win32_ver().
 
 - Fix UnboundLocalError in socket._sendfile_use_sendfile.