pyhg/sys/lib/python/robotparser.py

292 lines
9.7 KiB
Python

""" robotparser.py
Copyright (C) 2000 Bastian Kleineidam
You can choose between two licenses when using this package:
1) GNU GPLv2
2) PSF license for Python 2.2
The robots.txt Exclusion Protocol is implemented as specified in
http://info.webcrawler.com/mak/projects/robots/norobots-rfc.html
"""
import urlparse,urllib
__all__ = ["RobotFileParser"]
debug = 0
def _debug(msg):
if debug: print msg
class RobotFileParser:
""" This class provides a set of methods to read, parse and answer
questions about a single robots.txt file.
"""
def __init__(self, url=''):
self.entries = []
self.default_entry = None
self.disallow_all = False
self.allow_all = False
self.set_url(url)
self.last_checked = 0
def mtime(self):
"""Returns the time the robots.txt file was last fetched.
This is useful for long-running web spiders that need to
check for new robots.txt files periodically.
"""
return self.last_checked
def modified(self):
"""Sets the time the robots.txt file was last fetched to the
current time.
"""
import time
self.last_checked = time.time()
def set_url(self, url):
"""Sets the URL referring to a robots.txt file."""
self.url = url
self.host, self.path = urlparse.urlparse(url)[1:3]
def read(self):
"""Reads the robots.txt URL and feeds it to the parser."""
opener = URLopener()
f = opener.open(self.url)
lines = []
line = f.readline()
while line:
lines.append(line.strip())
line = f.readline()
self.errcode = opener.errcode
if self.errcode == 401 or self.errcode == 403:
self.disallow_all = True
_debug("disallow all")
elif self.errcode >= 400:
self.allow_all = True
_debug("allow all")
elif self.errcode == 200 and lines:
_debug("parse lines")
self.parse(lines)
def _add_entry(self, entry):
if "*" in entry.useragents:
# the default entry is considered last
self.default_entry = entry
else:
self.entries.append(entry)
def parse(self, lines):
"""parse the input lines from a robots.txt file.
We allow that a user-agent: line is not preceded by
one or more blank lines."""
state = 0
linenumber = 0
entry = Entry()
for line in lines:
linenumber = linenumber + 1
if not line:
if state==1:
_debug("line %d: warning: you should insert"
" allow: or disallow: directives below any"
" user-agent: line" % linenumber)
entry = Entry()
state = 0
elif state==2:
self._add_entry(entry)
entry = Entry()
state = 0
# remove optional comment and strip line
i = line.find('#')
if i>=0:
line = line[:i]
line = line.strip()
if not line:
continue
line = line.split(':', 1)
if len(line) == 2:
line[0] = line[0].strip().lower()
line[1] = urllib.unquote(line[1].strip())
if line[0] == "user-agent":
if state==2:
_debug("line %d: warning: you should insert a blank"
" line before any user-agent"
" directive" % linenumber)
self._add_entry(entry)
entry = Entry()
entry.useragents.append(line[1])
state = 1
elif line[0] == "disallow":
if state==0:
_debug("line %d: error: you must insert a user-agent:"
" directive before this line" % linenumber)
else:
entry.rulelines.append(RuleLine(line[1], False))
state = 2
elif line[0] == "allow":
if state==0:
_debug("line %d: error: you must insert a user-agent:"
" directive before this line" % linenumber)
else:
entry.rulelines.append(RuleLine(line[1], True))
else:
_debug("line %d: warning: unknown key %s" % (linenumber,
line[0]))
else:
_debug("line %d: error: malformed line %s"%(linenumber, line))
if state==2:
self.entries.append(entry)
_debug("Parsed rules:\n%s" % str(self))
def can_fetch(self, useragent, url):
"""using the parsed robots.txt decide if useragent can fetch url"""
_debug("Checking robots.txt allowance for:\n user agent: %s\n url: %s" %
(useragent, url))
if self.disallow_all:
return False
if self.allow_all:
return True
# search for given user agent matches
# the first match counts
url = urllib.quote(urlparse.urlparse(urllib.unquote(url))[2]) or "/"
for entry in self.entries:
if entry.applies_to(useragent):
return entry.allowance(url)
# try the default entry last
if self.default_entry:
return self.default_entry.allowance(url)
# agent not found ==> access granted
return True
def __str__(self):
ret = ""
for entry in self.entries:
ret = ret + str(entry) + "\n"
return ret
class RuleLine:
"""A rule line is a single "Allow:" (allowance==True) or "Disallow:"
(allowance==False) followed by a path."""
def __init__(self, path, allowance):
if path == '' and not allowance:
# an empty value means allow all
allowance = True
self.path = urllib.quote(path)
self.allowance = allowance
def applies_to(self, filename):
return self.path=="*" or filename.startswith(self.path)
def __str__(self):
return (self.allowance and "Allow" or "Disallow")+": "+self.path
class Entry:
"""An entry has one or more user-agents and zero or more rulelines"""
def __init__(self):
self.useragents = []
self.rulelines = []
def __str__(self):
ret = ""
for agent in self.useragents:
ret = ret + "User-agent: "+agent+"\n"
for line in self.rulelines:
ret = ret + str(line) + "\n"
return ret
def applies_to(self, useragent):
"""check if this entry applies to the specified agent"""
# split the name token and make it lower case
useragent = useragent.split("/")[0].lower()
for agent in self.useragents:
if agent=='*':
# we have the catch-all agent
return True
agent = agent.lower()
if agent in useragent:
return True
return False
def allowance(self, filename):
"""Preconditions:
- our agent applies to this entry
- filename is URL decoded"""
for line in self.rulelines:
_debug((filename, str(line), line.allowance))
if line.applies_to(filename):
return line.allowance
return True
class URLopener(urllib.FancyURLopener):
def __init__(self, *args):
urllib.FancyURLopener.__init__(self, *args)
self.errcode = 200
def http_error_default(self, url, fp, errcode, errmsg, headers):
self.errcode = errcode
return urllib.FancyURLopener.http_error_default(self, url, fp, errcode,
errmsg, headers)
def _check(a,b):
if not b:
ac = "access denied"
else:
ac = "access allowed"
if a!=b:
print "failed"
else:
print "ok (%s)" % ac
print
def _test():
global debug
rp = RobotFileParser()
debug = 1
# robots.txt that exists, gotten to by redirection
rp.set_url('http://www.musi-cal.com/robots.txt')
rp.read()
# test for re.escape
_check(rp.can_fetch('*', 'http://www.musi-cal.com/'), 1)
# this should match the first rule, which is a disallow
_check(rp.can_fetch('', 'http://www.musi-cal.com/'), 0)
# various cherry pickers
_check(rp.can_fetch('CherryPickerSE',
'http://www.musi-cal.com/cgi-bin/event-search'
'?city=San+Francisco'), 0)
_check(rp.can_fetch('CherryPickerSE/1.0',
'http://www.musi-cal.com/cgi-bin/event-search'
'?city=San+Francisco'), 0)
_check(rp.can_fetch('CherryPickerSE/1.5',
'http://www.musi-cal.com/cgi-bin/event-search'
'?city=San+Francisco'), 0)
# case sensitivity
_check(rp.can_fetch('ExtractorPro', 'http://www.musi-cal.com/blubba'), 0)
_check(rp.can_fetch('extractorpro', 'http://www.musi-cal.com/blubba'), 0)
# substring test
_check(rp.can_fetch('toolpak/1.1', 'http://www.musi-cal.com/blubba'), 0)
# tests for catch-all * agent
_check(rp.can_fetch('spam', 'http://www.musi-cal.com/search'), 0)
_check(rp.can_fetch('spam', 'http://www.musi-cal.com/Musician/me'), 1)
_check(rp.can_fetch('spam', 'http://www.musi-cal.com/'), 1)
_check(rp.can_fetch('spam', 'http://www.musi-cal.com/'), 1)
# robots.txt that does not exist
rp.set_url('http://www.lycos.com/robots.txt')
rp.read()
_check(rp.can_fetch('Mozilla', 'http://www.lycos.com/search'), 1)
if __name__ == '__main__':
_test()