Fix(robots): Do not parse empty strings

[bookmarks_db.git] / Robots / bkmk_robot_base.py

diff --git a/Robots/bkmk_robot_base.py b/Robots/bkmk_robot_base.py
index 9a74cb4ad67e342f96ec087e99186bb2eb69c4d2..52d6b563f1056c9ffe29a08299d59084d248be4e 100644 (file)
--- a/Robots/bkmk_robot_base.py
+++ b/Robots/bkmk_robot_base.py
@@ -29,10 +29,11 @@ from parse_html import parse_html
 
 
 reloc_dict = {
-  301: "perm.",
+  301: "perm1.",
   302: "temp2.",
   303: "temp3.",
   307: "temp7.",
+  308: "temp8.",
   "html": "html"
 }
 
@@ -125,7 +126,14 @@ class robot_base(Robot):
             if headers:
                 try:
                     content_type = headers["Content-Type"]
-                    self.log("   Content-Type: %s" % content_type)
+                    self.log("   Content-Type   : %s" % content_type)
+                    if content_type is None:
+                        if 'html' in content.lower():
+                            content_type = 'text/html'
+                        else:
+                            content_type = 'text/plain'
+                        self.log("   Set Content-Type to: %s"
+                                 % content_type)
                     try:
                         # extract charset from
                         # "text/html; foo; charset=UTF-8, bar; baz;"
@@ -136,14 +144,14 @@ class robot_base(Robot):
                     except (ValueError, IndexError):
                         charset = None
                         self.log("   no charset in Content-Type header")
+                    is_html = False
                     for ctype in ("text/html", "application/xhtml+xml"):
                         if content_type.startswith(ctype):
-                            html = True
+                            is_html = True
                             break
-                    else:
-                        html = False
-                    if html:
-                        parser = parse_html(content, charset, self.log)
+                    content_stripped = content.strip()
+                    if content_stripped and is_html:
+                        parser = parse_html(content_stripped, charset, self.log)
                         if parser:
                             bookmark.real_title = parser.title
                             icon = parser.icon
@@ -187,9 +195,11 @@ class robot_base(Robot):
                                 icons[icon_url] = None
                             else:
                                 content_type = icon_headers["Content-Type"]
-                                if content_type.startswith("application/") \
-                                   or content_type.startswith("image/") \
-                                   or content_type.startswith("text/plain"):
+                                if content_type and (
+                                    content_type.startswith("application/")
+                                    or content_type.startswith("image/")
+                                    or content_type.startswith("text/plain")
+                                ):
                                     bookmark.icon_href = icon_url
                                     self.log("   got icon       : %s"
                                              % content_type)
@@ -237,6 +247,10 @@ class robot_base(Robot):
                                                   % (url, timeout)
                                                   )
 
+                    if not content_stripped:
+                        self.log("   empty response, no content")
+                    if not is_html:
+                        self.log("   not html")
                 except KeyError as key:
                     self.log("   no header: %s" % key)
 
@@ -265,8 +279,15 @@ class robot_base(Robot):
         return 1
 
     def set_redirect(self, bookmark, errcode, newurl):
-        bookmark.moved = "(%s) to %s" % (reloc_dict[errcode], newurl)
-        self.log('   Moved: %s' % bookmark.moved)
+        bookmark.moved = moved = "(%s) to %s" % (reloc_dict[errcode], newurl)
+        try:
+            moved.encode('ascii')
+        except UnicodeEncodeError:
+            try:
+                moved = moved.encode(bookmark.charset)
+            except (LookupError, TypeError, UnicodeEncodeError):
+                moved = moved.encode('utf-8')
+        self.log('   Moved: %s' % moved)
 
     def finish_check_url(self, bookmark):
         start = self.start