Kaydet (Commit) 0522a9f1 authored tarafından Benjamin Peterson's avatar Benjamin Peterson

#1778443 robotparser fixes from Aristotelis Mikropoulos

üst 24122593
...@@ -55,11 +55,8 @@ class RobotFileParser: ...@@ -55,11 +55,8 @@ class RobotFileParser:
"""Reads the robots.txt URL and feeds it to the parser.""" """Reads the robots.txt URL and feeds it to the parser."""
opener = URLopener() opener = URLopener()
f = opener.open(self.url) f = opener.open(self.url)
lines = [] lines = [line.strip() for line in f]
line = f.readline() f.close()
while line:
lines.append(line.strip())
line = f.readline()
self.errcode = opener.errcode self.errcode = opener.errcode
if self.errcode in (401, 403): if self.errcode in (401, 403):
self.disallow_all = True self.disallow_all = True
...@@ -84,7 +81,7 @@ class RobotFileParser: ...@@ -84,7 +81,7 @@ class RobotFileParser:
entry = Entry() entry = Entry()
for line in lines: for line in lines:
linenumber = linenumber + 1 linenumber += 1
if not line: if not line:
if state == 1: if state == 1:
entry = Entry() entry = Entry()
......
...@@ -460,6 +460,7 @@ Luke Mewburn ...@@ -460,6 +460,7 @@ Luke Mewburn
Mike Meyer Mike Meyer
Steven Miale Steven Miale
Trent Mick Trent Mick
Aristotelis Mikropoulos
Damien Miller Damien Miller
Chad Miller Chad Miller
Jay T. Miller Jay T. Miller
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment