Minor refactoring

[bookmarks_db.git] / bkmk_parser.py

diff --git a/bkmk_parser.py b/bkmk_parser.py
old mode 100755 (executable)
new mode 100644 (file)
index 7ce99e8..02116cc
--- a/bkmk_parser.py
+++ b/bkmk_parser.py
@@ -1,11 +1,17 @@
-"""
-   Parser for Netscape Navigator's and Mozilla's bookmarks.html
+"""Parser for Netscape Navigator's and Mozilla's bookmarks.html
 
-   Written by BroytMann. Copyright (C) 1997-2005 PhiloSoft Design
+This file is a part of Bookmarks database and Internet robot.
 """
 
+__author__ = "Oleg Broytman <phd@phdru.name>"
+__copyright__ = "Copyright (C) 1997-2012 PhiloSoft Design"
+__license__ = "GNU GPL"
+
+__all__ = ['BkmkParser']
+
 
-import sys, os
+import os
+from m_lib.defenc import default_encoding
 from m_lib.net.www.html import HTMLParser
 from bkmk_objects import Folder, Bookmark, Ruler
 
@@ -29,8 +35,6 @@ else:
    dump_names = debug
 
 
-DEFAULT_CHARSET = None
-
 class BkmkParser(HTMLParser):
    def __init__(self):
       HTMLParser.__init__(self)
@@ -38,17 +42,15 @@ class BkmkParser(HTMLParser):
       self.urls = 0
       self.objects = 0
 
-      self.charset = ""
+      self.charset = None
       self.recode = None
 
-
    def handle_data(self, data):
       if data:
-         if DEFAULT_CHARSET:
-            data = unicode(data, self.charset, "replace").encode(DEFAULT_CHARSET, "replace")
+         if self.charset and default_encoding:
+            data = unicode(data, self.charset, "replace").encode(default_encoding, "xmlcharrefreplace")
          self.accumulator += data
 
-
    # Mozilla - get charset
    def do_meta(self, attrs):
       http_equiv = ""
@@ -67,27 +69,15 @@ class BkmkParser(HTMLParser):
             self.charset = content.split('=')[1]
          except IndexError:
             pass
-         else:
-            global DEFAULT_CHARSET
-            DEFAULT_CHARSET = sys.getdefaultencoding()
-            if DEFAULT_CHARSET == "ascii":
-               try:
-                  import locale
-               except ImportError:
-                  pass
-               else:
-                  DEFAULT_CHARSET = locale.getpreferredencoding()
-
 
    def start_title(self, attrs):
-      if DEFAULT_CHARSET:
-         self.accumulator += '<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=%s">\n' % DEFAULT_CHARSET
+      if default_encoding:
+         self.accumulator += '<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=%s">\n' % default_encoding
       self.accumulator += "<TITLE>"
 
    def end_title(self):
       self.accumulator += "</TITLE>"
 
-
    # Start root folder
    def start_h1(self, attrs):
       root_folder = Folder()
@@ -106,16 +96,18 @@ class BkmkParser(HTMLParser):
       debug("Root folder name: `%s'" % accumulator)
       self.root_folder.name = accumulator
 
-
    # Start a folder
    def start_h3(self, attrs):
+      last_modified = None
       for attrname, value in attrs:
          value = value.strip()
          if attrname == 'add_date':
             add_date = value
+         elif attrname == 'last_modified':
+            last_modified = value
 
       debug("New folder...")
-      folder = Folder(add_date)
+      folder = Folder(add_date, last_modified=last_modified)
       self.current_object = folder
       self.current_folder.append(folder)
       self.folder_stack.append(folder) # push new folder
@@ -129,12 +121,14 @@ class BkmkParser(HTMLParser):
       debug("Folder name: `%s'" % accumulator)
       self.current_folder.name = accumulator
 
-
    # Start a bookmark
    def start_a(self, attrs):
+      add_date = None
       last_visit = None
       last_modified = None
-      keyword = None
+      keyword = ''
+      icon = None
+      charset = None
 
       for attrname, value in attrs:
          value = value.strip()
@@ -148,9 +142,15 @@ class BkmkParser(HTMLParser):
             last_modified = value
          elif attrname == "shortcuturl":
             keyword = value
+         elif attrname == "icon":
+            icon = value
+         elif attrname == "last_charset":
+            charset = value
 
       debug("Bookmark points to: `%s'" % href)
-      bookmark = Bookmark(href, add_date, last_visit, last_modified, keyword or '')
+      bookmark = Bookmark(href, add_date, last_visit, last_modified,
+         keyword=keyword, icon=icon,
+         charset=charset, parser_charset=self.charset or default_encoding)
       self.current_object = bookmark
       self.current_folder.append(bookmark)
       self.urls += 1
@@ -164,7 +164,6 @@ class BkmkParser(HTMLParser):
       bookmark = self.current_folder[-1]
       bookmark.name = accumulator
 
-
    def flush(self):
       accumulator = self.accumulator
 
@@ -176,13 +175,11 @@ class BkmkParser(HTMLParser):
             current_object.comment += accumulator.strip()
             debug("Comment: `%s'" % current_object.comment)
 
-
    def start_dl(self, attrs):
       self.flush()
 
    do_dt = start_dl
 
-
    # End of folder
    def end_dl(self):
       self.flush()
@@ -198,19 +195,16 @@ class BkmkParser(HTMLParser):
          debug("FOLDER STACK is EMPTY!!! (2)")
       self.current_object = None
 
-
    def close(self):
       HTMLParser.close(self)
       if self.folder_stack:
          raise ValueError, "wrong folder stack: %s" % self.folder_stack
 
-
    def do_dd(self, attrs):
       pass
 
    do_p = do_dd
 
-
    # Start ruler
    def do_hr(self, attrs):
       self.flush()
@@ -219,17 +213,14 @@ class BkmkParser(HTMLParser):
       self.current_object = None
       self.objects += 1
 
-
    # BR in comment
    def do_br(self, attrs):
       self.accumulator += "<BR>"
 
-
    # Allow < in the text
    def unknown_starttag(self, tag, attrs):
       self.accumulator += "<%s>" % tag
 
-
    # Do not allow unknow end tags
    def unknown_endtag(self, tag):
       raise NotImplementedError("Unknow end tag `%s'" % tag)