No need to split bookmark.href - a robot will split it itself.

[bookmarks_db.git] / Robots / bkmk_rsimple.py

diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py
index 0d6c04409756be02fa161257b71729b68c2bd740..1c4795c11d6aecb8bfc6a460a260be5b56a65d8b 100644 (file)
--- a/Robots/bkmk_rsimple.py
+++ b/Robots/bkmk_rsimple.py
@@ -1,10 +1,18 @@
 """
-   Simple, strightforward robot; guaranteed to has problems with timeouts :)
+   Simple, strightforward robot
 
-   Written by BroytMann, Mar 2000 - Aug 2002. Copyright (C) 2000-2002 PhiloSoft Design
+   Written by Oleg BroytMann. Copyright (C) 2000-2007 PhiloSoft Design.
 """
 
 
+import sys, os, string
+import time, urllib
+from base64 import b64encode
+from urlparse import urljoin
+from m_lib.net.www.util import parse_time
+from m_lib.md5wrapper import md5wrapper
+
+
 class RedirectException(Exception):
    reloc_dict = {
       301: "perm.",
@@ -15,12 +23,6 @@ class RedirectException(Exception):
       Exception.__init__(self, "(%s) to %s" % (self.reloc_dict[errcode], newurl))
 
 
-import string, os
-import time, urllib
-from m_lib.net.www.util import parse_time
-from m_lib.md5wrapper import md5wrapper
-
-
 class MyURLopener(urllib.URLopener):
    # Error 302 -- relocated (temporarily)
    def http_error_302(self, url, fp, errcode, errmsg, headers, data=None): 
@@ -81,17 +83,20 @@ from bkmk_objects import Robot
 from parse_html import parse_html
 
 class robot_simple(Robot):
-   def check_url(self, bookmark, url_type, url_rest):
+   def check_url(self, bookmark):
       if not self.tempfname:
          self.tempfname = bookmark.tempfname
 
       try:
          try:
             self.start = int(time.time())
+            url_type, url_rest = urllib.splittype(bookmark.href)
             url_host, url_path = urllib.splithost(url_rest)
             url_path, url_tag  = urllib.splittag(url_path)
 
+            if bookmark.charset: urllib._urlopener.addheader('Accept-Charset', bookmark.charset)
             fname, headers = urllib.urlretrieve("%s://%s%s" % (url_type, url_host, url_path), self.tempfname)
+            if bookmark.charset: del urllib._urlopener.addheaders[-1]
 
             size = 0
             last_modified = None
@@ -143,15 +148,41 @@ class robot_simple(Robot):
                      if self.log: self.log("   final title    : %s" % bookmark.real_title)
                      if parser.refresh:
                         refresh = parser.refresh
-                        try:
-                           timeout = int(refresh.split(';')[0])
-                        except (IndexError, ValueError):
-                           timeout = "ERROR"
                         try:
                            url = refresh.split('=', 1)[1]
                         except IndexError:
                            url = "self"
-                        raise RedirectException("html", "%s (%d sec)" % (url, timeout))
+                        try:
+                           timeout = int(refresh.split(';')[0])
+                        except (IndexError, ValueError):
+                           timeout = None
+                        if timeout is None:
+                           raise RedirectException("html", "Bad redirect to %s (%s)" % (url, refresh))
+                        else:
+                           raise RedirectException("html", "%s (%d sec)" % (url, timeout))
+
+                     # Get favicon.ico
+                     icon = parser.icon
+                     if not icon:
+                        icon = "/favicon.ico"
+                     icon = urljoin("%s://%s%s" % (url_type, url_host, url_path), icon)
+                     if self.log: self.log("   icon           : %s" % icon)
+                     try:
+                        fname, headers = urllib.urlretrieve(icon)
+                     except:
+                        etype, emsg, tb = sys.exc_info()
+                        if self.log: self.log("   no icon        : %s %s" % (etype, emsg))
+                        etype = None
+                        emsg = None
+                        tb = None
+                     else:
+                        icon_file = open(fname, "rb")
+                        icon = icon_file.read()
+                        icon_file.close()
+                        os.remove(fname)
+                        content_type = headers["Content-Type"]
+                        bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon))
+
                except KeyError:
                   pass