X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_ph_html5.py;h=111e1ed4b4b38360fef84ca41a9276351d0e6c5e;hb=cb9c36b39ed72cd1fa272130d2bcf162a89c3013;hp=6400d02c20b924b0d77e65709c9136ab1647d41e;hpb=00549b1c0622ee6ed0ac12249097cf4562bc486e;p=bookmarks_db.git

diff --git a/parse_html/bkmk_ph_html5.py b/parse_html/bkmk_ph_html5.py
index 6400d02..111e1ed 100644
--- a/parse_html/bkmk_ph_html5.py
+++ b/parse_html/bkmk_ph_html5.py
@@ -17,14 +17,16 @@ from .bkmk_ph_util import HTMLParser
 
 def parse_html(html_text, charset=None, log=None):
     parser = HTML5Parser()
-    html_tree = parser.parse(html_text, encoding=charset, parseMeta=bool(charset))
-
-    for node in html_tree.childNodes:
-        if (node.name == 'html') and (node.type != 3): # Skip DocType element
-            html = node
-            break
-    else:
-        html = None
+    html_tree = parser.parse(
+        html_text, encoding=charset, parseMeta=bool(charset))
+
+    html = None
+    if hasattr(html_tree, 'childNodes'):
+        for node in html_tree.childNodes:
+            # Skip DocType element
+            if (node.name == 'html') and (node.type != 3):
+                html = node
+                break
 
     if not html:
         return None
@@ -58,7 +60,8 @@ def parse_html(html_text, charset=None, log=None):
                 if meta_content:
                     try:
                         meta_charset = \
-                            meta_content.lower().split('charset=')[1].split(';')[0]
+                            meta_content.lower().split('charset=')[1].\
+                            split(';')[0]
                     except IndexError:
                         meta_charset = False
                     else: