Fix(parse_html): Do not parse empty strings

diff --git a/Robots/bkmk_robot_base.py b/Robots/bkmk_robot_base.py
index 1d762d94d1713f951d583f120d658f2f9f8d17dd..f32e1b304d45aab959b047d51db4140241b90633 100644 (file)
--- a/Robots/bkmk_robot_base.py
+++ b/Robots/bkmk_robot_base.py
@@ -144,13 +144,12 @@ 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:
+                    if content and is_html:
                         parser = parse_html(content, charset, self.log)
                         if parser:
                             bookmark.real_title = parser.title
@@ -247,6 +246,10 @@ class robot_base(Robot):
                                                   % (url, timeout)
                                                   )
 
+                    if not content:
+                        self.log("   empty response, no content")
+                    if not is_html:
+                        self.log("   not html")
                 except KeyError as key:
                     self.log("   no header: %s" % key)
 

diff --git a/parse_html/bkmk_parse_html.py b/parse_html/bkmk_parse_html.py
index be5daab26e85ec47620838f506b49085c392e78f..7764303a4e754715a86a3a12e5d0b48d1d5acb8c 100644 (file)
--- a/parse_html/bkmk_parse_html.py
+++ b/parse_html/bkmk_parse_html.py
@@ -101,7 +101,7 @@ BKMK_DEBUG_HTML_PARSERS = os.environ.get("BKMK_DEBUG_HTML_PARSERS")
 
 
 def parse_html(html_text, charset=None, log=None):
-    if not parsers:
+    if not html_text or not parsers:
         return None
 
     if charset:

diff --git a/parse_html/bkmk_ph_lxml.py b/parse_html/bkmk_ph_lxml.py
index a02de919fb3cfef336b287ceb753e2d3d0cd980f..24274826c52ee40179a83d96ff1eb07ceb60e6ec 100644 (file)
--- a/parse_html/bkmk_ph_lxml.py
+++ b/parse_html/bkmk_ph_lxml.py
@@ -17,6 +17,8 @@ from .bkmk_ph_util import HTMLParser
 
 
 def parse_html(html_text, charset=None, log=None):
+    if not html_text:
+        return None
     try:
         html_tree = fromstring(html_text)
     except ValueError as e: