X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=ae7c39e481be8e30b144981bf18a1a8aa32b19d2;hb=b9b055b270749b6af5d83f1e6390793d7b951888;hp=7a21c4fd1a1e72faf1397df7a853194dab30f696;hpb=89976fb4bb0376958ea17cfe337ecb872221eeb1;p=bookmarks_db.git

diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py
index 7a21c4f..ae7c39e 100644
--- 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 - Mar 2004. Copyright (C) 2000-2004 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.",
@@ -13,12 +21,7 @@ class RedirectException(Exception):
    }
    def __init__(self, errcode, newurl):
       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
+      self.url = newurl
 
 
 class MyURLopener(urllib.URLopener):
@@ -42,8 +45,7 @@ class MyURLopener(urllib.URLopener):
 
 urllib._urlopener = MyURLopener()
 
-# Some sites allow only Mozilla-compatible browsers; way to stop robots?
-server_version = "Mozilla/3.0 (compatible; Python-urllib/%s)" % urllib.__version__
+server_version = "bookmarks_db (Python-urllib/%s)" % urllib.__version__
 urllib._urlopener.addheaders[0] = ('User-agent', server_version)
 
 
@@ -81,17 +83,22 @@ 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())
+            bookmark.icon = None
+
+            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
@@ -138,9 +145,7 @@ class robot_simple(Robot):
                      if self.log: self.log("   no charset in Content-Type header")
                   if content_type == "text/html":
                      parser = parse_html(fname, charset, self.log)
-                     title = parser.title.replace('\r', '').replace('\n', ' ').strip()
-                     bookmark.real_title = parser.unescape(title)
-                     if self.log: self.log("   final title    : %s" % bookmark.real_title)
+                     bookmark.real_title = parser.title
                      if parser.refresh:
                         refresh = parser.refresh
                         try:
@@ -148,13 +153,51 @@ class robot_simple(Robot):
                         except IndexError:
                            url = "self"
                         try:
-                           timeout = int(refresh.split(';')[0])
+                           timeout = float(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))
+                           try:
+                              timeout = int(refresh.split(';')[0])
+                           except ValueError:
+                              pass # float timeout
+                           raise RedirectException("html", "%s (%s 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("   looking for icon at : %s" % icon)
+                     try:
+                        for i in range(8):
+                           try:
+                              fname, headers = urllib.urlretrieve(icon)
+                           except RedirectException, e:
+                              icon = e.url
+                              if self.log: self.log("       redirect to : %s" % icon)
+                           else:
+                              break
+                        else:
+                           raise IOError("Too many redirects")
+                     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:
+                        content_type = headers["Content-Type"]
+                        if content_type.startswith("image/"):
+                           icon_file = open(fname, "rb")
+                           icon = icon_file.read()
+                           icon_file.close()
+                           bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon))
+                           if self.log: self.log("   got icon       : %s" % content_type)
+                        else:
+                           if self.log: self.log("   no icon        : bad content type %s" % content_type)
+                        os.remove(fname)
+
                except KeyError:
                   pass