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