X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_ph_lxml.py;h=a02de919fb3cfef336b287ceb753e2d3d0cd980f;hb=03d2e88efc7b2046a404679a36a8a65d6faae98c;hp=03dd6f4c0d90bb60627a442d1dfb743eb7a07961;hpb=9e2bf3eaee5a8ee84c61ee5eac9c55090d45f63f;p=bookmarks_db.git

diff --git a/parse_html/bkmk_ph_lxml.py b/parse_html/bkmk_ph_lxml.py
index 03dd6f4..a02de91 100644
--- a/parse_html/bkmk_ph_lxml.py
+++ b/parse_html/bkmk_ph_lxml.py
@@ -11,15 +11,32 @@ __license__ = "GNU GPL"
 __all__ = ['parse_html']
 
 
-from lxml.html import fromtring
+import re
+from lxml.html import fromstring
 from .bkmk_ph_util import HTMLParser
 
 
 def parse_html(html_text, charset=None, log=None):
-    html_tree = fromtring(html_text)
-
-    if html_tree.getroot() is None:
-        return None
+    try:
+        html_tree = fromstring(html_text)
+    except ValueError as e:
+        if e.args[0].startswith(
+            'Unicode strings with encoding declaration are not supported.'
+            ' Please use bytes input'
+        ):
+            if not charset:
+                match = re.search(
+                    '<\\?xml version="(\\d|.)+" encoding="([^"]+)"\\?>',
+                    html_text, re.U)
+                if match:
+                    charset = match.group(2)
+            if charset:
+                html_text = html_text.encode(charset)
+                html_tree = fromstring(html_text)
+            else:
+                return None
+        else:
+            raise
 
     title = html_tree.findtext('head/title')
     if title is None: