[2.7] bpo-32861: robotparser fix incomplete __str__ methods. (GH-5711) (GH-6795) (GH-6817)
The robotparser's __str__ representation now includes wildcard
entries.
(cherry picked from commit c3fa1f2b93fa4bf96a8aadc74ee196384cefa31e)
Co-authored-by: Michael Lazar <lazar.michael22@gmail.com>.
diff --git a/Lib/robotparser.py b/Lib/robotparser.py
index a7137a3..4e13f7f 100644
--- a/Lib/robotparser.py
+++ b/Lib/robotparser.py
@@ -160,7 +160,10 @@
def __str__(self):
- return ''.join([str(entry) + "\n" for entry in self.entries])
+ entries = self.entries
+ if self.default_entry is not None:
+ entries = entries + [self.default_entry]
+ return '\n'.join(map(str, entries)) + '\n'
class RuleLine:
diff --git a/Lib/test/test_robotparser.py b/Lib/test/test_robotparser.py
index 8ed5d89..ba7ccf8 100644
--- a/Lib/test/test_robotparser.py
+++ b/Lib/test/test_robotparser.py
@@ -136,6 +136,31 @@
bad = ['/cyberworld/map/index.html']
+class StringFormattingTest(BaseRobotTest, unittest.TestCase):
+ robots_txt = """\
+User-agent: *
+Crawl-delay: 1
+Request-rate: 3/15
+Disallow: /cyberworld/map/ # This is an infinite virtual URL space
+
+# Cybermapper knows where to go.
+User-agent: cybermapper
+Disallow: /some/path
+ """
+
+ expected_output = """\
+User-agent: cybermapper
+Disallow: /some/path
+
+User-agent: *
+Disallow: /cyberworld/map/
+
+"""
+
+ def test_string_formatting(self):
+ self.assertEqual(str(self.parser), self.expected_output)
+
+
class RobotHandler(BaseHTTPRequestHandler):
def do_GET(self):
@@ -226,6 +251,7 @@
UseFirstUserAgentWildcardTest,
EmptyQueryStringTest,
DefaultEntryTest,
+ StringFormattingTest,
PasswordProtectedSiteTestCase,
NetworkTestCase)