From 01e7f1237af1fa33a1feb0d07d5bb2e5e438962e Mon Sep 17 00:00:00 2001
From: Oleg Broytman <phd@phdru.name>
Date: Tue, 12 Jul 2011 14:37:25 +0000
Subject: [PATCH] Split hrefs into domain and path components; recode only
 domain.

git-svn-id: file:///home/phd/archive/SVN/bookmarks_db/trunk@338 fdd5c36f-1aea-0310-aeeb-c58d7e2b6c23
---
 Writers/bkmk_whtml.py |  2 +-
 bkmk_objects.py       | 49 +++++++++++++++++++++++++++++++++++--------
 bkmk_parser.py        |  3 ++-
 doc/ANNOUNCE          | 26 ++---------------------
 4 files changed, 45 insertions(+), 35 deletions(-)

diff --git a/Writers/bkmk_whtml.py b/Writers/bkmk_whtml.py
index 309fdcf..ad109f8 100644
--- a/Writers/bkmk_whtml.py
+++ b/Writers/bkmk_whtml.py
@@ -49,7 +49,7 @@ class writer_html(Writer):
       self.outfile.write(ind_s*level + "</DL><p>\n")
 
    def bookmark(self, b, level):
-      self.outfile.write(ind_s*(level+1) + '<DT><A HREF="%s" ADD_DATE="%s"' % (b.href.encode(default_encoding), b.add_date))
+      self.outfile.write(ind_s*(level+1) + '<DT><A HREF="%s" ADD_DATE="%s"' % (b.href, b.add_date))
       if b.last_visit: self.outfile.write(' LAST_VISIT="%s"' % b.last_visit)
       self.outfile.write(' LAST_MODIFIED="%s"' % b.last_modified)
       if BKMK_FORMAT == "MOZILLA":
diff --git a/bkmk_objects.py b/bkmk_objects.py
index 00e186c..be139ed 100644
--- a/bkmk_objects.py
+++ b/bkmk_objects.py
@@ -16,7 +16,7 @@ __all__ = ['Folder', 'Bookmark', 'Ruler', 'Walker', 'Writer', 'Robot',
 ]
 
 
-import os
+import os, urllib
 
 BKMK_FORMAT = os.environ.get("BKMK_FORMAT", "MOZILLA")
 
@@ -56,14 +56,45 @@ class Bookmark:
    isBookmark = 1
 
    def __init__(self, href, add_date, last_visit=None, last_modified=None,
-         keyword=None, comment='', icon_href=None, icon=None, charset=None):
-      if isinstance(href, str):
-         try:
-            href = href.decode('idna')
-         except UnicodeDecodeError: # Non-ascii href
-            href = href.decode('utf-8')
-      elif not isinstance(href, unicode):
-          raise TypeError("Bookmark's href must be str or unicode, not %r" % type(href))
+         keyword=None, comment='', icon_href=None, icon=None,
+         charset=None, parser_charset=None):
+      protocol, request = urllib.splittype(href)
+      user, password, port = None, None, None
+      host, path = urllib.splithost(request)
+      if host:
+         user, host = urllib.splituser(host)
+         if user:
+            user, password = urllib.splitpasswd(user)
+         host, port = urllib.splitport(host)
+         if port: port = int(port)
+      path, tag = urllib.splittag(path)
+      path, query = urllib.splitquery(path)
+      path = urllib.unquote(path)
+      if tag: tag = urllib.unquote_plus(tag)
+
+      if host: # host can be None for Mozilla's place: URLs
+          host = host.decode(parser_charset or 'utf-8').encode('idna')
+
+      href = protocol + "://"
+      if user:
+         href += urllib.quote(user)
+         if password:
+            href += ':' + urllib.quote(password)
+         href += '@'
+      if host:
+         href += host
+         if port:
+            href += ':%d' % port
+      if path:
+         if protocol == "file":
+            href += urllib.quote(path)
+         else:
+            href += urllib.quote(path)
+      if query:
+         href += '?' + query
+      if tag:
+         href += '#' + urllib.quote_plus(tag)
+
       self.href = href
       self.add_date = add_date
       self.last_visit = last_visit
diff --git a/bkmk_parser.py b/bkmk_parser.py
index 9d5bc0f..d4b6a2f 100644
--- a/bkmk_parser.py
+++ b/bkmk_parser.py
@@ -152,7 +152,8 @@ class BkmkParser(HTMLParser):
 
       debug("Bookmark points to: `%s'" % href)
       bookmark = Bookmark(href, add_date, last_visit, last_modified,
-         keyword=keyword, icon=icon, charset=charset)
+         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
diff --git a/doc/ANNOUNCE b/doc/ANNOUNCE
index 3a63157..2c5ddca 100644
--- a/doc/ANNOUNCE
+++ b/doc/ANNOUNCE
@@ -5,32 +5,10 @@ WHAT IS IT
    A set of classes, libraries, programs and plugins I use to manipulate my
 bookmarks.html.
 
-WHAT'S NEW in version 4.4.0 (2011-01-07).
+WHAT'S NEW in version 4.5.0 (2011-??-??).
 
-Moved BeautifulSoup.py and subproc.py from Robots/ to the top-level directory.
+Recode international domain names.
 
-Moved parse_html.py and its submodules to a separate parse_html package.
-
-Added statistics code to parse_html, got a statistics on parser
-success/failrure rate, reordered parsers.
-
-Removed old cruft.
-
-
-WHAT'S NEW in version 4.3.1 (2011-01-03).
-
-Get favicon before HTML redirect (refresh).
-
-Get favicon even if it's of a wrong type; many sites return favicon as
-text/plain or application/*; the only exception is text/html which is usually
-an error page instead of error 404.
-
-
-WHAT'S NEW in version 4.3.0 (2011-01-01).
-
-Robots no longer have one global temporary file - there are at least two
-(html and favicon), and in the future there will be more for
-asynchronous robot(s) that will test many URLs in parallel.
 
 
 WHERE TO GET
-- 
2.39.5