X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_ph_beautifulsoup4.py;h=7687c755e94aa6c80a6d019239efd170554cc4b6;hb=7212b79ec48ec9947048b9fcbe0903b125fdf542;hp=fbd54ff9cfd207c583d5119ba06c20faf23f9d4d;hpb=8a33ee42d7f8edff076734eba8c1718da8127b22;p=bookmarks_db.git

diff --git a/parse_html/bkmk_ph_beautifulsoup4.py b/parse_html/bkmk_ph_beautifulsoup4.py
index fbd54ff..7687c75 100644
--- a/parse_html/bkmk_ph_beautifulsoup4.py
+++ b/parse_html/bkmk_ph_beautifulsoup4.py
@@ -11,23 +11,36 @@ __license__ = "GNU GPL"
 __all__ = ['parse_html']
 
 
+import warnings
+
 from bs4 import BeautifulSoup
 
 from .bkmk_ph_util import HTMLParser
 from compat import string_type
 
+warnings.filterwarnings(
+    'ignore', 'No parser was explicitly specified')
+warnings.filterwarnings(
+    'ignore',
+    "It looks like you're parsing an XML document using an HTML parser.")
+
 universal_charset = "utf-8"
 DEFAULT_CHARSET = "cp1251"  # Stupid default for Russian Cyrillic
 
 
 def _parse_html(html_text, charset):
     try:
-        return BeautifulSoup(html_text, from_encoding=charset)
+        if isinstance(html_text, bytes):
+            return BeautifulSoup(html_text, from_encoding=charset)
+        else:
+            return BeautifulSoup(html_text)
     except TypeError:
         return None
 
 
 def parse_html(html_text, charset=None, log=None):
+    if not html_text:
+        return None
     root = _parse_html(html_text, charset)
     if root is None:
         return None
@@ -56,9 +69,12 @@ def parse_html(html_text, charset=None, log=None):
         else:
             parts = []
             for part in title:
-                if not isinstance(part, string_type):
-                    part = part.decode()
-                parts.append(part.strip())
+                #if not isinstance(part, string_type):
+                #    part = part.decode()
+                if part.strip:
+                    parts.append(part.strip())
+                else:
+                    parts.append(' ')  # Skip tags, they're usually `<br>`
             title = ''.join(parts)
 
     meta = head.find(_find_contenttype, recursive=False)
@@ -84,12 +100,12 @@ def parse_html(html_text, charset=None, log=None):
             if meta_content:
                 meta_charset = _charset = meta_content.lower()
 
-    if title and (_charset or meta_charset):
-        try:
-            title = title.encode(_charset or meta_charset)
-        except LookupError:
-            title = title.encode(universal_charset)
-            _charset = universal_charset
+    #if title and (_charset or meta_charset):
+    #    try:
+    #        title = title.encode(_charset or meta_charset)
+    #    except LookupError:
+    #        title = title.encode(universal_charset)
+    #        _charset = universal_charset
 
     meta = head.find(_find_refresh, recursive=False)
     if meta: