X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html_beautifulsoup.py;h=25719ca9d643ad86e2c2844c572b938d1bf6f792;hb=5f637b5a5fe27098985975928632b9fea5ea3c62;hp=9b0faded06ab92cb951b844421555776690118ff;hpb=0b76120991af955d34a9376d44e1df719f7ac16c;p=bookmarks_db.git

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index 9b0fade..25719ca 100644
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -49,11 +49,12 @@ class BadDeclParser(BeautifulSoup):
          return j
 
 
-def parse_html(filename, charset=None):
+def parse_html(filename, charset=None, log=None):
    infile = open(filename, 'r')
    try:
       root = BadDeclParser(infile, fromEncoding=charset)
    except TypeError:
+      if log: log("TypeError")
       return None
    finally:
       infile.close()
@@ -61,12 +62,16 @@ def parse_html(filename, charset=None):
    try:
       head = root.html.head
    except AttributeError:
+      if log: log("No HTML in root or no HEAD in HTML")
       return None
 
    if head is None:
       head = root.html # Some sites put TITLE in HTML without HEAD
 
    _charset = root.originalEncoding
+   if _charset == "windows-1252": # Replace default
+      _charset = DEFAULT_CHARSET
+
    try:
       title = head.title.string.encode(_charset)
    except AttributeError:
@@ -80,6 +85,17 @@ def parse_html(filename, charset=None):
       except AttributeError:
          title = '' # no TITLE in HTML too
 
+   meta = head.find(_find_contenttype, recursive=False)
+   if meta:
+      try:
+         __charset = meta.get("content").lower().split('charset=')[1].split(';')[0]
+      except IndexError: # No charset in the META Content-Type
+         meta_charset = False
+      else:
+         meta_charset = _charset == __charset
+   else:
+      meta_charset = False
+
    meta = head.find(_find_refresh, recursive=False)
    if meta:
       refresh = meta.get("content")
@@ -92,7 +108,11 @@ def parse_html(filename, charset=None):
    else:
       icon = None
 
-   return BSoupParser(_charset, _charset != charset, title, refresh, icon)
+   return BSoupParser(_charset, meta_charset, title, refresh, icon)
+
+def _find_contenttype(Tag):
+   return (Tag.name == "meta") and \
+      (Tag.get("http-equiv", '').lower() == "content-type")
 
 def _find_refresh(Tag):
    return (Tag.name == "meta") and \