Fix(Robots/bkmk_rrequests): Add forgotten spaces in log

[bookmarks_db.git] / parse_html / bkmk_ph_lxml.py

diff --git a/parse_html/bkmk_ph_lxml.py b/parse_html/bkmk_ph_lxml.py
index 03dd6f4c0d90bb60627a442d1dfb743eb7a07961..24274826c52ee40179a83d96ff1eb07ceb60e6ec 100644 (file)
--- a/parse_html/bkmk_ph_lxml.py
+++ b/parse_html/bkmk_ph_lxml.py
@@ -11,15 +11,34 @@ __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:
+    if not html_text:
         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: