bkmk_parser.py

   1 """Parser for Netscape Navigator's and Mozilla's bookmarks.html
   2
   3 This file is a part of Bookmarks database and Internet robot.
   4 """
   5
   6 __author__ = "Oleg Broytman <phd@phdru.name>"
   7 __copyright__ = "Copyright (C) 1997-2023 PhiloSoft Design"
   8 __license__ = "GNU GPL"
   9
  10 __all__ = ['BkmkParser']
  11
  12
  13 import os
  14 from m_lib.defenc import default_encoding
  15 from m_lib.net.www.html import HTMLParser
  16 from bkmk_objects import Folder, Bookmark, Ruler
  17
  18
  19 DEBUG = "BKMK_DEBUG" in os.environ
  20
  21 if DEBUG:
  22     def debug(note):
  23         print(note)
  24
  25     def dump_names(folder_stack):
  26         _l = []
  27         for object in folder_stack:
  28             if object.isFolder:
  29                 _l.append(object.name)
  30         return "'%s'" % "' '".join(_l)
  31
  32 else:
  33     def debug(note):
  34         pass
  35     dump_names = debug
  36
  37
  38 class BkmkParser(HTMLParser):
  39     def __init__(self):
  40         HTMLParser.__init__(self)
  41
  42         self.urls = 0
  43         self.objects = 0
  44
  45         self.charset = None
  46         self.recode = None
  47
  48     def handle_data(self, data):
  49         if data:
  50             #if self.charset and default_encoding:
  51             #    data = data.decode(self.charset, "replace").\
  52             #        encode(default_encoding, "xmlcharrefreplace")
  53             self.accumulator += data
  54
  55     # Mozilla - get charset
  56     def do_meta(self, attrs):
  57         http_equiv = ""
  58         content = ""
  59
  60         for attrname, value in attrs:
  61             value = value.strip()
  62             if attrname == 'http-equiv':
  63                 http_equiv = value.lower()
  64             elif attrname == 'content':
  65                 content = value
  66
  67         if http_equiv == "content-type":
  68             try:
  69                 # extract charset from "text/html; charset=UTF-8"
  70                 self.charset = content.split('=')[1]
  71             except IndexError:
  72                 pass
  73
  74     def start_title(self, attrs):
  75         if default_encoding:
  76             self.accumulator += '<META HTTP-EQUIV="Content-Type" '
  77             'CONTENT="text/html; charset=%s">\n' % default_encoding
  78         self.accumulator += "<TITLE>"
  79
  80     def end_title(self):
  81         self.accumulator += "</TITLE>"
  82
  83     # Start root folder
  84     def start_h1(self, attrs):
  85         root_folder = Folder()
  86         self.current_object = root_folder
  87         self.root_folder = root_folder
  88         self.current_folder = root_folder
  89         self.folder_stack = [root_folder]
  90
  91         self.root_folder.header = self.accumulator.strip()
  92         self.accumulator = ''
  93
  94     def end_h1(self):
  95         accumulator = self.accumulator
  96         self.accumulator = ''
  97
  98         debug("Root folder name: `%s'" % accumulator)
  99         self.root_folder.name = accumulator
 100
 101     # Start a folder
 102     def start_h3(self, attrs):
 103         last_modified = None
 104         for attrname, value in attrs:
 105             value = value.strip()
 106             if attrname == 'add_date':
 107                 add_date = value
 108             elif attrname == 'last_modified':
 109                 last_modified = value
 110
 111         debug("New folder...")
 112         folder = Folder(add_date, last_modified=last_modified)
 113         self.current_object = folder
 114         self.current_folder.append(folder)
 115         self.folder_stack.append(folder)  # push new folder
 116         self.current_folder = folder
 117         self.objects += 1
 118
 119     def end_h3(self):
 120         accumulator = self.accumulator
 121         self.accumulator = ''
 122
 123         debug("Folder name: `%s'" % accumulator)
 124         self.current_folder.name = accumulator
 125
 126     # Start a bookmark
 127     def start_a(self, attrs):
 128         add_date = None
 129         last_visit = None
 130         last_modified = None
 131         keyword = ''
 132         icon = None
 133         charset = None
 134
 135         for attrname, value in attrs:
 136             value = value.strip()
 137             if attrname == "href":
 138                 href = value
 139             elif attrname == "add_date":
 140                 add_date = value
 141             elif attrname == "last_visit":
 142                 last_visit = value
 143             elif attrname == "last_modified":
 144                 last_modified = value
 145             elif attrname == "shortcuturl":
 146                 keyword = value
 147             elif attrname == "icon":
 148                 icon = value
 149             elif attrname == "last_charset":
 150                 charset = value
 151
 152         debug("Bookmark points to: `%s'" % href)
 153         bookmark = Bookmark(href, add_date, last_visit, last_modified,
 154                             keyword=keyword, icon=icon, charset=charset,
 155                             parser_charset=self.charset or default_encoding)
 156         self.current_object = bookmark
 157         self.current_folder.append(bookmark)
 158         self.urls += 1
 159         self.objects += 1
 160
 161     def end_a(self):
 162         accumulator = self.accumulator
 163         self.accumulator = ''
 164
 165         debug("Bookmark name: `%s'" % accumulator)
 166         bookmark = self.current_folder[-1]
 167         bookmark.name = accumulator
 168
 169     def flush(self):
 170         accumulator = self.accumulator
 171
 172         if accumulator:
 173             self.accumulator = ''
 174
 175             current_object = self.current_object
 176             if current_object:
 177                 current_object.comment += accumulator.strip()
 178                 debug("Comment: `%s'" % current_object.comment)
 179
 180     def start_dl(self, attrs):
 181         self.flush()
 182
 183     do_dt = start_dl
 184
 185     # End of folder
 186     def end_dl(self):
 187         self.flush()
 188         debug("End folder")
 189         debug("Folder stack: %s" % dump_names(self.folder_stack))
 190         if self.folder_stack:
 191             del self.folder_stack[-1]  # pop last folder
 192             if self.folder_stack:
 193                 self.current_folder = self.folder_stack[-1]
 194             else:
 195                 debug("FOLDER STACK is EMPTY!!! (1)")
 196         else:
 197             debug("FOLDER STACK is EMPTY!!! (2)")
 198         self.current_object = None
 199
 200     def close(self):
 201         HTMLParser.close(self)
 202         if self.folder_stack:
 203             raise ValueError("wrong folder stack: %s" % self.folder_stack)
 204
 205     def do_dd(self, attrs):
 206         pass
 207
 208     do_p = do_dd
 209
 210     # Start ruler
 211     def do_hr(self, attrs):
 212         self.flush()
 213         debug("Ruler")
 214         self.current_folder.append(Ruler())
 215         self.current_object = None
 216         self.objects += 1
 217
 218     # BR in comment
 219     def do_br(self, attrs):
 220         self.accumulator += "<BR>"
 221
 222     # Allow < in the text
 223     def unknown_starttag(self, tag, attrs):
 224         self.accumulator += "<%s>" % tag
 225
 226     # Do not allow unknow end tags
 227     def unknown_endtag(self, tag):
 228         raise NotImplementedError("Unknow end tag `%s'" % tag)