1 """Base class for robots
3 This file is a part of Bookmarks database and Internet robot.
7 __author__ = "Oleg Broytman <phd@phdru.name>"
8 __copyright__ = "Copyright (C) 2000-2023 PhiloSoft Design"
9 __license__ = "GNU GPL"
11 __all__ = ['robot_base', 'get_error']
14 from base64 import b64encode
19 from urllib.parse import splittype, splithost, splittag, urljoin
21 from urllib import splittype, splithost, splittag
22 from urlparse import urljoin
24 from m_lib.md5wrapper import md5wrapper
25 from m_lib.net.www.util import parse_time
27 from bkmk_objects import Robot
28 from parse_html import parse_html
41 if isinstance(e, str):
47 s.append("'%s'" % str(i).replace('\n', "\\n"))
48 return "(%s)" % ' '.join(s)
51 # Icon cache; maps URL to a tuple (content type, data)
52 # or None if there is no icon.
56 class robot_base(Robot):
59 def __init__(self, *args, **kw):
60 Robot.__init__(self, *args, **kw)
61 socket.setdefaulttimeout(int(self.timeout))
63 def check_url(self, bookmark):
65 self.start = int(time.time())
68 url_type, url_rest = splittype(bookmark.href)
69 url_host, url_path = splithost(url_rest)
70 url_path, url_tag = splittag(url_path) # noqa: E221
71 # multiple spaces before operator
73 url = "%s://%s%s" % (url_type, url_host, url_path)
74 error, redirect_code, redirect_to, headers, content = \
75 self.get(bookmark, url, True)
78 bookmark.error = error
82 self.set_redirect(bookmark, redirect_code, redirect_to)
90 size = headers["Content-Length"]
95 last_modified = headers["Last-Modified"]
100 last_modified = parse_time(last_modified)
105 last_modified = str(int(last_modified))
107 last_modified = bookmark.last_visit
110 bookmark.last_modified = last_modified
113 if url_type == "ftp": # Pass welcome message through MD5
114 ftp_welcome = self.get_ftp_welcome()
115 if not isinstance(ftp_welcome, bytes):
116 ftp_welcome = ftp_welcome.encode('utf-8')
117 md5.update(ftp_welcome)
119 if isinstance(content, bytes):
122 md5.update(content.encode('utf-8'))
123 bookmark.md5 = str(md5)
127 content_type = headers["Content-Type"]
128 self.log(" Content-Type: %s" % content_type)
130 # extract charset from
131 # "text/html; foo; charset=UTF-8, bar; baz;"
132 content_type, charset = content_type.split(';', 1)
133 content_type = content_type.strip()
134 charset = charset.split('=')[1].strip().split(',')[0]
135 self.log(" HTTP charset : %s" % charset)
136 except (ValueError, IndexError):
138 self.log(" no charset in Content-Type header")
139 for ctype in ("text/html", "application/xhtml+xml"):
140 if content_type.startswith(ctype):
146 parser = parse_html(content, charset, self.log)
148 bookmark.real_title = parser.title
153 icon = "/favicon.ico"
155 "%s://%s%s" % (url_type, url_host, url_path), icon)
156 self.log(" looking for icon at: %s" % icon_url)
157 if icon_url in icons:
159 bookmark.icon_href = icon_url
160 content_type, bookmark.icon = icons[icon_url]
161 self.log(" cached icon: %s" % content_type)
163 self.log(" cached icon: no icon")
168 error, icon_redirect_code, \
169 icon_redirect_to, icon_headers, \
171 self.get(bookmark, _icon_url)
172 if icon_redirect_code:
173 _icon_url = icon_redirect_to
174 self.log(" redirect to : %s"
177 if icon_data is None:
178 raise IOError("No icon")
181 raise IOError("Too many redirects")
183 etype, emsg, _ = sys.exc_info()
184 self.log(" no icon : %s %s"
186 etype = emsg = _ = None
187 icons[icon_url] = None
189 content_type = icon_headers["Content-Type"]
190 if content_type.startswith("application/") \
191 or content_type.startswith("image/") \
192 or content_type.startswith("text/plain"):
193 bookmark.icon_href = icon_url
194 self.log(" got icon : %s"
197 content_type.startswith("application/")
198 or content_type.startswith(
201 self.log(" non-image content type,"
203 content_type = 'image/x-icon'
204 if not isinstance(icon_data, bytes):
205 icon_data = icon_data.encode('utf-8')
206 bookmark.icon = "data:%s;base64,%s" \
207 % (content_type, b64encode(icon_data))
208 icons[icon_url] = (content_type,
212 self.log(" no icon :"
213 "bad content type '%s'"
216 icons[icon_url] = None
217 if parser and parser.refresh:
218 refresh = parser.refresh
220 url = refresh.split('=', 1)[1]
224 timeout = float(refresh.split(';')[0])
225 except (IndexError, ValueError):
226 self.set_redirect(bookmark, "html",
227 "Bad redirect to %s (%s)"
232 timeout = int(refresh.split(';')[0])
235 self.set_redirect(bookmark, "html",
240 except KeyError as key:
241 self.log(" no header: %s" % key)
244 bookmark.error = "Unexpected EOF (FTP server closed connection)"
245 self.log(' EOF: %s' % bookmark.error)
247 except KeyboardInterrupt:
248 self.log("Keyboard interrupt (^C)")
251 except socket.error as e:
252 bookmark.error = get_error(e)
253 self.log(bookmark.error)
257 traceback.print_exc()
258 bookmark.error = "Exception!"
259 self.log(' Exception: %s' % bookmark.error)
262 self.finish_check_url(bookmark)
267 def set_redirect(self, bookmark, errcode, newurl):
268 bookmark.moved = "(%s) to %s" % (reloc_dict[errcode], newurl)
269 self.log(' Moved: %s' % bookmark.moved)
271 def finish_check_url(self, bookmark):
273 bookmark.last_tested = str(start)
274 now = int(time.time())
275 bookmark.test_time = str(now - start)