X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_ph_htmlparser.py;h=45e89f5119817787b0b926c7f43627a7dc22e7dd;hb=c88cb7a75e7caf1d67466cfa107981d95115fa0c;hp=5c0a440ff7fea80bba26b58a0e729b75df8567cc;hpb=71712390f4edb041609ff7bc9272d12a5c1a9b1d;p=bookmarks_db.git

diff --git a/parse_html/bkmk_ph_htmlparser.py b/parse_html/bkmk_ph_htmlparser.py
index 5c0a440..45e89f5 100644
--- a/parse_html/bkmk_ph_htmlparser.py
+++ b/parse_html/bkmk_ph_htmlparser.py
@@ -1,10 +1,11 @@
 """HTML Parser using Pythons' HTMLParser
 
 This file is a part of Bookmarks database and Internet robot.
+
 """
 
 __author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 1997-2012 PhiloSoft Design"
+__copyright__ = "Copyright (C) 1997-2017 PhiloSoft Design"
 __license__ = "GNU GPL"
 
 __all__ = ['parse_html']
@@ -18,82 +19,80 @@ class HTMLHeadDone(Exception): pass
 
 
 class HTMLParser(_HTMLParser):
-   def __init__(self, charset=None):
-      _HTMLParser.__init__(self)
-      self.charset = charset
-      self.meta_charset = 0
-      self.title = None
-      self.refresh = None
-      self.icon = None
-
-   def end_head(self):
-      raise HTMLHeadDone()
-
-   def do_meta(self, attrs):
-      http_equiv = ""
-      content = ""
-
-      for attrname, value in attrs:
-         if value:
-            value = value.strip()
-            if attrname == 'http-equiv':
-               http_equiv = value.lower()
-            elif attrname == 'content':
-               content = value
-
-      if (not self.charset) and (http_equiv == "content-type"):
-         try:
-            # extract charset from "text/html; foo; charset=UTF-8, bar; baz;"
-            self.charset = content.lower().split('charset=')[1].split(';')[0].split(',')[0]
-            self.meta_charset = 1 # Remember that the charset was retrieved from
-                                  # META tag, not from the Content-Type header
-         except IndexError:
-            pass
-
-      if http_equiv == "refresh":
-         self.refresh = content
-
-   def start_title(self, attrs):
-      self.accumulator = ''
-
-   def end_title(self):
-      if not self.title: # use only the first title
-         self.title = self.accumulator
-
-   def do_link(self, attrs):
-      has_icon = False
-      href = None
-
-      for attrname, value in attrs:
-         if value:
-            value = value.strip()
-            if (attrname == 'rel') and (value.lower() in ('icon', 'shortcut icon')):
-               has_icon = True
-            elif attrname == 'href':
-               href = value
-
-      if has_icon:
-         self.icon = href
-
-
-def parse_html(filename, charset=None, log=None):
-   infile = open(filename, 'r')
-   parser = HTMLParser(charset)
-
-   for line in infile:
-      try:
-         parser.feed(line)
-      except (HTMLParseError, HTMLHeadDone):
-         break
-
-   infile.close()
-
-   try:
-      parser.close()
-   except (HTMLParseError, HTMLHeadDone):
-      pass
-
-   if parser.title is None:
-      return None
-
-   return parser
+    def __init__(self, charset=None):
+        _HTMLParser.__init__(self)
+        self.charset = charset
+        self.meta_charset = 0
+        self.title = None
+        self.refresh = None
+        self.icon = None
+
+    def end_head(self):
+        raise HTMLHeadDone()
+
+    def do_meta(self, attrs):
+        http_equiv = ""
+        content = ""
+
+        for attrname, value in attrs:
+            if value:
+                value = value.strip()
+                if attrname == 'http-equiv':
+                    http_equiv = value.lower()
+                elif attrname == 'content':
+                    content = value
+                elif (attrname == 'charset') and (not self.charset):
+                    self.charset = value.lower()
+                    self.meta_charset = 1
+
+        if (not self.charset) and (http_equiv == "content-type"):
+            try:
+                # extract charset from "text/html; foo; charset=UTF-8, bar; baz;"
+                self.charset = content.lower().split('charset=')[1].split(';')[0].split(',')[0]
+                self.meta_charset = 1 # Remember that the charset was retrieved from
+                                      # META tag, not from the Content-Type header
+            except IndexError:
+                pass
+
+        if http_equiv == "refresh":
+            self.refresh = content
+
+    def start_title(self, attrs):
+        self.accumulator = ''
+
+    def end_title(self):
+        if not self.title: # use only the first title
+            self.title = self.accumulator
+
+    def do_link(self, attrs):
+        has_icon = False
+        href = None
+
+        for attrname, value in attrs:
+            if value:
+                value = value.strip()
+                if (attrname == 'rel') and (value.lower() in ('icon', 'shortcut icon')):
+                    has_icon = True
+                elif attrname == 'href':
+                    href = value
+
+        if has_icon:
+            self.icon = href
+
+
+def parse_html(html_text, charset=None, log=None):
+    parser = HTMLParser(charset)
+
+    try:
+        parser.feed(html_text)
+    except (HTMLParseError, HTMLHeadDone):
+        pass
+
+    try:
+        parser.close()
+    except (HTMLParseError, HTMLHeadDone):
+        pass
+
+    if (parser.title is None) and (parser.refresh is None) and (parser.icon is None):
+        return None
+    return parser