| 1 | """ robotparser.py
|
|---|
| 2 |
|
|---|
| 3 | Copyright (C) 2000 Bastian Kleineidam
|
|---|
| 4 |
|
|---|
| 5 | You can choose between two licenses when using this package:
|
|---|
| 6 | 1) GNU GPLv2
|
|---|
| 7 | 2) PSF license for Python 2.2
|
|---|
| 8 |
|
|---|
| 9 | The robots.txt Exclusion Protocol is implemented as specified in
|
|---|
| 10 | http://info.webcrawler.com/mak/projects/robots/norobots-rfc.html
|
|---|
| 11 | """
|
|---|
| 12 | import urlparse
|
|---|
| 13 | import urllib
|
|---|
| 14 |
|
|---|
| 15 | __all__ = ["RobotFileParser"]
|
|---|
| 16 |
|
|---|
| 17 |
|
|---|
| 18 | class RobotFileParser:
|
|---|
| 19 | """ This class provides a set of methods to read, parse and answer
|
|---|
| 20 | questions about a single robots.txt file.
|
|---|
| 21 |
|
|---|
| 22 | """
|
|---|
| 23 |
|
|---|
| 24 | def __init__(self, url=''):
|
|---|
| 25 | self.entries = []
|
|---|
| 26 | self.default_entry = None
|
|---|
| 27 | self.disallow_all = False
|
|---|
| 28 | self.allow_all = False
|
|---|
| 29 | self.set_url(url)
|
|---|
| 30 | self.last_checked = 0
|
|---|
| 31 |
|
|---|
| 32 | def mtime(self):
|
|---|
| 33 | """Returns the time the robots.txt file was last fetched.
|
|---|
| 34 |
|
|---|
| 35 | This is useful for long-running web spiders that need to
|
|---|
| 36 | check for new robots.txt files periodically.
|
|---|
| 37 |
|
|---|
| 38 | """
|
|---|
| 39 | return self.last_checked
|
|---|
| 40 |
|
|---|
| 41 | def modified(self):
|
|---|
| 42 | """Sets the time the robots.txt file was last fetched to the
|
|---|
| 43 | current time.
|
|---|
| 44 |
|
|---|
| 45 | """
|
|---|
| 46 | import time
|
|---|
| 47 | self.last_checked = time.time()
|
|---|
| 48 |
|
|---|
| 49 | def set_url(self, url):
|
|---|
| 50 | """Sets the URL referring to a robots.txt file."""
|
|---|
| 51 | self.url = url
|
|---|
| 52 | self.host, self.path = urlparse.urlparse(url)[1:3]
|
|---|
| 53 |
|
|---|
| 54 | def read(self):
|
|---|
| 55 | """Reads the robots.txt URL and feeds it to the parser."""
|
|---|
| 56 | opener = URLopener()
|
|---|
| 57 | f = opener.open(self.url)
|
|---|
| 58 | lines = [line.strip() for line in f]
|
|---|
| 59 | f.close()
|
|---|
| 60 | self.errcode = opener.errcode
|
|---|
| 61 | if self.errcode in (401, 403):
|
|---|
| 62 | self.disallow_all = True
|
|---|
| 63 | elif self.errcode >= 400:
|
|---|
| 64 | self.allow_all = True
|
|---|
| 65 | elif self.errcode == 200 and lines:
|
|---|
| 66 | self.parse(lines)
|
|---|
| 67 |
|
|---|
| 68 | def _add_entry(self, entry):
|
|---|
| 69 | if "*" in entry.useragents:
|
|---|
| 70 | # the default entry is considered last
|
|---|
| 71 | if self.default_entry is None:
|
|---|
| 72 | # the first default entry wins
|
|---|
| 73 | self.default_entry = entry
|
|---|
| 74 | else:
|
|---|
| 75 | self.entries.append(entry)
|
|---|
| 76 |
|
|---|
| 77 | def parse(self, lines):
|
|---|
| 78 | """parse the input lines from a robots.txt file.
|
|---|
| 79 | We allow that a user-agent: line is not preceded by
|
|---|
| 80 | one or more blank lines."""
|
|---|
| 81 | # states:
|
|---|
| 82 | # 0: start state
|
|---|
| 83 | # 1: saw user-agent line
|
|---|
| 84 | # 2: saw an allow or disallow line
|
|---|
| 85 | state = 0
|
|---|
| 86 | linenumber = 0
|
|---|
| 87 | entry = Entry()
|
|---|
| 88 |
|
|---|
| 89 | for line in lines:
|
|---|
| 90 | linenumber += 1
|
|---|
| 91 | if not line:
|
|---|
| 92 | if state == 1:
|
|---|
| 93 | entry = Entry()
|
|---|
| 94 | state = 0
|
|---|
| 95 | elif state == 2:
|
|---|
| 96 | self._add_entry(entry)
|
|---|
| 97 | entry = Entry()
|
|---|
| 98 | state = 0
|
|---|
| 99 | # remove optional comment and strip line
|
|---|
| 100 | i = line.find('#')
|
|---|
| 101 | if i >= 0:
|
|---|
| 102 | line = line[:i]
|
|---|
| 103 | line = line.strip()
|
|---|
| 104 | if not line:
|
|---|
| 105 | continue
|
|---|
| 106 | line = line.split(':', 1)
|
|---|
| 107 | if len(line) == 2:
|
|---|
| 108 | line[0] = line[0].strip().lower()
|
|---|
| 109 | line[1] = urllib.unquote(line[1].strip())
|
|---|
| 110 | if line[0] == "user-agent":
|
|---|
| 111 | if state == 2:
|
|---|
| 112 | self._add_entry(entry)
|
|---|
| 113 | entry = Entry()
|
|---|
| 114 | entry.useragents.append(line[1])
|
|---|
| 115 | state = 1
|
|---|
| 116 | elif line[0] == "disallow":
|
|---|
| 117 | if state != 0:
|
|---|
| 118 | entry.rulelines.append(RuleLine(line[1], False))
|
|---|
| 119 | state = 2
|
|---|
| 120 | elif line[0] == "allow":
|
|---|
| 121 | if state != 0:
|
|---|
| 122 | entry.rulelines.append(RuleLine(line[1], True))
|
|---|
| 123 | state = 2
|
|---|
| 124 | if state == 2:
|
|---|
| 125 | self._add_entry(entry)
|
|---|
| 126 |
|
|---|
| 127 |
|
|---|
| 128 | def can_fetch(self, useragent, url):
|
|---|
| 129 | """using the parsed robots.txt decide if useragent can fetch url"""
|
|---|
| 130 | if self.disallow_all:
|
|---|
| 131 | return False
|
|---|
| 132 | if self.allow_all:
|
|---|
| 133 | return True
|
|---|
| 134 | # search for given user agent matches
|
|---|
| 135 | # the first match counts
|
|---|
| 136 | parsed_url = urlparse.urlparse(urllib.unquote(url))
|
|---|
| 137 | url = urlparse.urlunparse(('', '', parsed_url.path,
|
|---|
| 138 | parsed_url.params, parsed_url.query, parsed_url.fragment))
|
|---|
| 139 | url = urllib.quote(url)
|
|---|
| 140 | if not url:
|
|---|
| 141 | url = "/"
|
|---|
| 142 | for entry in self.entries:
|
|---|
| 143 | if entry.applies_to(useragent):
|
|---|
| 144 | return entry.allowance(url)
|
|---|
| 145 | # try the default entry last
|
|---|
| 146 | if self.default_entry:
|
|---|
| 147 | return self.default_entry.allowance(url)
|
|---|
| 148 | # agent not found ==> access granted
|
|---|
| 149 | return True
|
|---|
| 150 |
|
|---|
| 151 |
|
|---|
| 152 | def __str__(self):
|
|---|
| 153 | return ''.join([str(entry) + "\n" for entry in self.entries])
|
|---|
| 154 |
|
|---|
| 155 |
|
|---|
| 156 | class RuleLine:
|
|---|
| 157 | """A rule line is a single "Allow:" (allowance==True) or "Disallow:"
|
|---|
| 158 | (allowance==False) followed by a path."""
|
|---|
| 159 | def __init__(self, path, allowance):
|
|---|
| 160 | if path == '' and not allowance:
|
|---|
| 161 | # an empty value means allow all
|
|---|
| 162 | allowance = True
|
|---|
| 163 | path = urlparse.urlunparse(urlparse.urlparse(path))
|
|---|
| 164 | self.path = urllib.quote(path)
|
|---|
| 165 | self.allowance = allowance
|
|---|
| 166 |
|
|---|
| 167 | def applies_to(self, filename):
|
|---|
| 168 | return self.path == "*" or filename.startswith(self.path)
|
|---|
| 169 |
|
|---|
| 170 | def __str__(self):
|
|---|
| 171 | return (self.allowance and "Allow" or "Disallow") + ": " + self.path
|
|---|
| 172 |
|
|---|
| 173 |
|
|---|
| 174 | class Entry:
|
|---|
| 175 | """An entry has one or more user-agents and zero or more rulelines"""
|
|---|
| 176 | def __init__(self):
|
|---|
| 177 | self.useragents = []
|
|---|
| 178 | self.rulelines = []
|
|---|
| 179 |
|
|---|
| 180 | def __str__(self):
|
|---|
| 181 | ret = []
|
|---|
| 182 | for agent in self.useragents:
|
|---|
| 183 | ret.extend(["User-agent: ", agent, "\n"])
|
|---|
| 184 | for line in self.rulelines:
|
|---|
| 185 | ret.extend([str(line), "\n"])
|
|---|
| 186 | return ''.join(ret)
|
|---|
| 187 |
|
|---|
| 188 | def applies_to(self, useragent):
|
|---|
| 189 | """check if this entry applies to the specified agent"""
|
|---|
| 190 | # split the name token and make it lower case
|
|---|
| 191 | useragent = useragent.split("/")[0].lower()
|
|---|
| 192 | for agent in self.useragents:
|
|---|
| 193 | if agent == '*':
|
|---|
| 194 | # we have the catch-all agent
|
|---|
| 195 | return True
|
|---|
| 196 | agent = agent.lower()
|
|---|
| 197 | if agent in useragent:
|
|---|
| 198 | return True
|
|---|
| 199 | return False
|
|---|
| 200 |
|
|---|
| 201 | def allowance(self, filename):
|
|---|
| 202 | """Preconditions:
|
|---|
| 203 | - our agent applies to this entry
|
|---|
| 204 | - filename is URL decoded"""
|
|---|
| 205 | for line in self.rulelines:
|
|---|
| 206 | if line.applies_to(filename):
|
|---|
| 207 | return line.allowance
|
|---|
| 208 | return True
|
|---|
| 209 |
|
|---|
| 210 | class URLopener(urllib.FancyURLopener):
|
|---|
| 211 | def __init__(self, *args):
|
|---|
| 212 | urllib.FancyURLopener.__init__(self, *args)
|
|---|
| 213 | self.errcode = 200
|
|---|
| 214 |
|
|---|
| 215 | def prompt_user_passwd(self, host, realm):
|
|---|
| 216 | ## If robots.txt file is accessible only with a password,
|
|---|
| 217 | ## we act as if the file wasn't there.
|
|---|
| 218 | return None, None
|
|---|
| 219 |
|
|---|
| 220 | def http_error_default(self, url, fp, errcode, errmsg, headers):
|
|---|
| 221 | self.errcode = errcode
|
|---|
| 222 | return urllib.FancyURLopener.http_error_default(self, url, fp, errcode,
|
|---|
| 223 | errmsg, headers)
|
|---|
