X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=ae7c39e481be8e30b144981bf18a1a8aa32b19d2;hb=b9b055b270749b6af5d83f1e6390793d7b951888;hp=58879b01bc3a76f33b059dd7e7f272bd8375900e;hpb=a8c8b37737468dc990f12724ec21dca1404a1557;p=bookmarks_db.git

diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py
index 58879b0..ae7c39e 100644
--- a/Robots/bkmk_rsimple.py
+++ b/Robots/bkmk_rsimple.py
@@ -21,6 +21,7 @@ class RedirectException(Exception):
    }
    def __init__(self, errcode, newurl):
       Exception.__init__(self, "(%s) to %s" % (self.reloc_dict[errcode], newurl))
+      self.url = newurl
 
 
 class MyURLopener(urllib.URLopener):
@@ -44,8 +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)
 
 
@@ -83,13 +83,16 @@ 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)
 
@@ -142,9 +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:
@@ -152,22 +153,33 @@ 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("   icon           : %s" % icon)
+                     if self.log: self.log("   looking for icon at : %s" % icon)
                      try:
-                        fname, headers = urllib.urlretrieve(icon)
+                        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))
@@ -175,12 +187,16 @@ class robot_simple(Robot):
                         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))
+                        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