]> git.phdru.name Git - bookmarks_db.git/blobdiff - Robots/bkmk_rsimple.py
Mark the error output from the subprocess with a prefix.
[bookmarks_db.git] / Robots / bkmk_rsimple.py
index 58879b01bc3a76f33b059dd7e7f272bd8375900e..77c7a150109509583c2eb3d5fb5f538d45806283 100644 (file)
@@ -44,8 +44,7 @@ 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 +82,16 @@ 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 +144,7 @@ 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:
@@ -152,13 +152,11 @@ 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))
+                           raise RedirectException("html", "%s (%s sec)" % (url, timeout))
 
                      # Get favicon.ico
                      icon = parser.icon
@@ -175,12 +173,15 @@ 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))
+                        else:
+                           if self.log: self.log("   no icon        : %s" % content_type)
+                        os.remove(fname)
 
                except KeyError:
                   pass