diff options
author | Berker Peksag <berker.peksag@gmail.com> | 2016-09-18 17:17:58 (GMT) |
---|---|---|
committer | Berker Peksag <berker.peksag@gmail.com> | 2016-09-18 17:17:58 (GMT) |
commit | 9a7bbb2e3f12faaf4110ecd15fb739e94f4bc8f6 (patch) | |
tree | 0b8430da11ccc8beb8fc9d4c70780543de1e781e /Lib/urllib | |
parent | 85c98bf9682a46f7b15e9c79c68d38af8a9109b0 (diff) | |
download | cpython-9a7bbb2e3f12faaf4110ecd15fb739e94f4bc8f6.zip cpython-9a7bbb2e3f12faaf4110ecd15fb739e94f4bc8f6.tar.gz cpython-9a7bbb2e3f12faaf4110ecd15fb739e94f4bc8f6.tar.bz2 |
Issue #25400: RobotFileParser now correctly returns default values for crawl_delay and request_rate
Initial patch by Peter Wirtz.
Diffstat (limited to 'Lib/urllib')
-rw-r--r-- | Lib/urllib/robotparser.py | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/Lib/urllib/robotparser.py b/Lib/urllib/robotparser.py index 85add16..9dab4c1 100644 --- 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]) |