Added a second way to pass specify encodings to UnicodeDammit and

  EncodingDetector, based on the order of precedence defined in the
  HTML5 spec, starting at:
  https://html.spec.whatwg.org/multipage/parsing.html#parsing-with-a-known-character-encoding

  Encodings in 'known_definite_encodings' are tried first, then
  byte-order-mark sniffing is run, then encodings in 'user_encodings'
  are tried. The old argument, 'override_encodings', is now a
  deprecated alias for 'known_definite_encodings'.

  This changes the default behavior of the html.parser and lxml tree
  builders, in a way that may slightly improve encoding
  detection but will probably have no effect. [bug=1889014]

7 files changed, 412 insertions, 242 deletions

diff --git a/CHANGELOG b/CHANGELOG
index 7954872..93c59ba 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -7,6 +7,20 @@
 * Performance improvement when processing tags that speeds up overall
   tree construction by 2%. Patch by Morotti. [bug=1899358]
 
+* Added a second way to pass specify encodings to UnicodeDammit and
+  EncodingDetector, based on the order of precedence defined in the
+  HTML5 spec, starting at:
+  https://html.spec.whatwg.org/multipage/parsing.html#parsing-with-a-known-character-encoding
+
+  Encodings in 'known_definite_encodings' are tried first, then
+  byte-order-mark sniffing is run, then encodings in 'user_encodings'
+  are tried. The old argument, 'override_encodings', is now a
+  deprecated alias for 'known_definite_encodings'.
+
+  This changes the default behavior of the html.parser and lxml tree
+  builders, in a way that may slightly improve encoding
+  detection but will probably have no effect. [bug=1889014]
+
 * Improve the warning issued when a directory name (as opposed to
   the name of a regular file) is passed as markup into the BeautifulSoup
   constructor. [bug=1913628]
diff --git a/bs4/builder/__init__.py b/bs4/builder/__init__.py
index 03da4c6..b6e2c37 100644
--- a/bs4/builder/__init__.py
+++ b/bs4/builder/__init__.py
@@ -234,7 +234,8 @@ class TreeBuilder(object):
         :param markup: Some markup -- probably a bytestring.
         :param user_specified_encoding: The user asked to try this encoding.
         :param document_declared_encoding: The markup itself claims to be
-            in this encoding.
+            in this encoding. NOTE: This argument is not used by the
+            calling code and can probably be removed.
         :param exclude_encodings: The user asked _not_ to try any of
             these encodings.
 
diff --git a/bs4/builder/_htmlparser.py b/bs4/builder/_htmlparser.py
index 96a7b7d..2f2bf1e 100644
--- a/bs4/builder/_htmlparser.py
+++ b/bs4/builder/_htmlparser.py
@@ -359,9 +359,24 @@ class HTMLParserTreeBuilder(HTMLTreeBuilder):
             return
 
         # Ask UnicodeDammit to sniff the most likely encoding.
+
+        # This was provided by the end-user; treat it as a known
+        # definite encoding per the algorithm laid out in the HTML5
+        # spec.  (See the EncodingDetector class for details.)
+        known_definite_encodings = [user_specified_encoding]
+
+        # This was found in the document; treat it as a slightly lower-priority
+        # user encoding.
+        user_encodings = [document_declared_encoding]
+
         try_encodings = [user_specified_encoding, document_declared_encoding]
-        dammit = UnicodeDammit(markup, try_encodings, is_html=True,
-                               exclude_encodings=exclude_encodings)
+        dammit = UnicodeDammit(
+            markup,
+            known_definite_encodings=known_definite_encodings,
+            user_encodings=user_encodings,
+            is_html=True,
+            exclude_encodings=exclude_encodings
+        )
         yield (dammit.markup, dammit.original_encoding,
                dammit.declared_html_encoding,
                dammit.contains_replacement_characters)
diff --git a/bs4/builder/_lxml.py b/bs4/builder/_lxml.py
index 1b44d75..c670b84 100644
--- a/bs4/builder/_lxml.py
+++ b/bs4/builder/_lxml.py
@@ -180,9 +180,19 @@ class LXMLTreeBuilderForXML(TreeBuilder):
             yield (markup.encode("utf8"), "utf8",
                    document_declared_encoding, False)
 
-        try_encodings = [user_specified_encoding, document_declared_encoding]
+        # This was provided by the end-user; treat it as a known
+        # definite encoding per the algorithm laid out in the HTML5
+        # spec.  (See the EncodingDetector class for details.)
+        known_definite_encodings = [user_specified_encoding]
+
+        # This was found in the document; treat it as a slightly lower-priority
+        # user encoding.
+        user_encodings = [document_declared_encoding]
         detector = EncodingDetector(
-            markup, try_encodings, is_html, exclude_encodings)
+            markup, known_definite_encodings=known_definite_encodings,
+            user_encodings=user_encodings, is_html=is_html,
+            exclude_encodings=exclude_encodings
+        )
         for encoding in detector.encodings:
             yield (detector.markup, encoding, document_declared_encoding, False)
 
diff --git a/bs4/dammit.py b/bs4/dammit.py
index 33f7b7d..7e0a7f8 100644
--- a/bs4/dammit.py
+++ b/bs4/dammit.py
@@ -228,32 +228,65 @@ class EncodingDetector:
     Order of precedence:
 
     1. Encodings you specifically tell EncodingDetector to try first
-    (the override_encodings argument to the constructor).
+    (the known_definite_encodings argument to the constructor).
 
-    2. An encoding declared within the bytestring itself, either in an
+    2. An encoding determined by sniffing the document's byte-order mark.
+
+    3. Encodings you specifically tell EncodingDetector to try if
+    byte-order mark sniffing fails (the user_encodings argument to the
+    constructor).
+
+    4. An encoding declared within the bytestring itself, either in an
     XML declaration (if the bytestring is to be interpreted as an XML
     document), or in a <meta> tag (if the bytestring is to be
     interpreted as an HTML document.)
 
-    3. An encoding detected through textual analysis by chardet,
+    5. An encoding detected through textual analysis by chardet,
     cchardet, or a similar external library.
 
     4. UTF-8.
 
     5. Windows-1252.
+
     """
-    def __init__(self, markup, override_encodings=None, is_html=False,
-                 exclude_encodings=None):
+    def __init__(self, markup, known_definite_encodings=None,
+                 is_html=False, exclude_encodings=None,
+                 user_encodings=None, override_encodings=None):
         """Constructor.
 
         :param markup: Some markup in an unknown encoding.
-        :param override_encodings: These encodings will be tried first.
-        :param is_html: If True, this markup is considered to be HTML. Otherwise
-            it's assumed to be XML.
-        :param exclude_encodings: These encodings will not be tried, even
-            if they otherwise would be.
+
+        :param known_definite_encodings: When determining the encoding
+            of `markup`, these encodings will be tried first, in
+            order. In HTML terms, this corresponds to the "known
+            definite encoding" step defined here:
+            https://html.spec.whatwg.org/multipage/parsing.html#parsing-with-a-known-character-encoding
+
+        :param user_encodings: These encodings will be tried after the
+            `known_definite_encodings` have been tried and failed, and
+            after an attempt to sniff the encoding by looking at a
+            byte order mark has failed. In HTML terms, this
+            corresponds to the step "user has explicitly instructed
+            the user agent to override the document's character
+            encoding", defined here:
+            https://html.spec.whatwg.org/multipage/parsing.html#determining-the-character-encoding
+
+        :param override_encodings: A deprecated alias for
+            known_definite_encodings. Any encodings here will be tried
+            immediately after the encodings in
+            known_definite_encodings.
+
+        :param is_html: If True, this markup is considered to be
+            HTML. Otherwise it's assumed to be XML.
+
+        :param exclude_encodings: These encodings will not be tried,
+            even if they otherwise would be.
+
         """
-        self.override_encodings = override_encodings or []
+        self.known_definite_encodings = list(known_definite_encodings or [])
+        if override_encodings:
+            self.known_definite_encodings += override_encodings
+        self.user_encodings = user_encodings or []
         exclude_encodings = exclude_encodings or []
         self.exclude_encodings = set([x.lower() for x in exclude_encodings])
         self.chardet_encoding = None
@@ -286,7 +319,9 @@ class EncodingDetector:
         :yield: A sequence of strings.
         """
         tried = set()
-        for e in self.override_encodings:
+
+        # First, try the known definite encodings
+        for e in self.known_definite_encodings:
             if self._usable(e, tried):
                 yield e
 
@@ -295,6 +330,12 @@ class EncodingDetector:
         if self._usable(self.sniffed_encoding, tried):
             yield self.sniffed_encoding
 
+        # Sniffing the byte-order mark did nothing; try the user
+        # encodings.
+        for e in self.user_encodings:
+            if self._usable(e, tried):
+                yield e
+            
         # Look within the document for an XML or HTML encoding
         # declaration.
         if self.declared_encoding is None:
@@ -405,13 +446,33 @@ class UnicodeDammit:
         "iso-8859-2",
         ]
 
-    def __init__(self, markup, override_encodings=[],
-                 smart_quotes_to=None, is_html=False, exclude_encodings=[]):
+    def __init__(self, markup, known_definite_encodings=[],
+                 smart_quotes_to=None, is_html=False, exclude_encodings=[],
+                 user_encodings=None, override_encodings=None
+    ):
         """Constructor.
 
         :param markup: A bytestring representing markup in an unknown encoding.
-        :param override_encodings: These encodings will be tried first,
-           before any sniffing code is run.
+
+        :param known_definite_encodings: When determining the encoding
+            of `markup`, these encodings will be tried first, in
+            order. In HTML terms, this corresponds to the "known
+            definite encoding" step defined here:
+            https://html.spec.whatwg.org/multipage/parsing.html#parsing-with-a-known-character-encoding
+
+        :param user_encodings: These encodings will be tried after the
+            `known_definite_encodings` have been tried and failed, and
+            after an attempt to sniff the encoding by looking at a
+            byte order mark has failed. In HTML terms, this
+            corresponds to the step "user has explicitly instructed
+            the user agent to override the document's character
+            encoding", defined here:
+            https://html.spec.whatwg.org/multipage/parsing.html#determining-the-character-encoding
+
+        :param override_encodings: A deprecated alias for
+            known_definite_encodings. Any encodings here will be tried
+            immediately after the encodings in
+            known_definite_encodings.
 
         :param smart_quotes_to: By default, Microsoft smart quotes will, like all other characters, be converted
            to Unicode characters. Setting this to 'ascii' will convert them to ASCII quotes instead.
@@ -421,6 +482,7 @@ class UnicodeDammit:
             it's assumed to be XML.
         :param exclude_encodings: These encodings will not be considered, even
             if the sniffing code thinks they might make sense.
+
         """
         self.smart_quotes_to = smart_quotes_to
         self.tried_encodings = []
@@ -428,7 +490,9 @@ class UnicodeDammit:
         self.is_html = is_html
         self.log = logging.getLogger(__name__)
         self.detector = EncodingDetector(
-            markup, override_encodings, is_html, exclude_encodings)
+            markup, known_definite_encodings, is_html, exclude_encodings,
+            user_encodings, override_encodings
+        )
 
         # Short-circuit if the data is in Unicode to begin with.
         if isinstance(markup, unicode) or markup == '':
diff --git a/bs4/tests/test_dammit.py b/bs4/tests/test_dammit.py
new file mode 100644
index 0000000..b477df8
--- /dev/null
+++ b/bs4/tests/test_dammit.py
@@ -0,0 +1,286 @@
+# encoding: utf-8
+import logging
+import unittest
+import bs4
+from bs4 import BeautifulSoup
+from bs4.dammit import (
+    EncodingDetector,
+    UnicodeDammit,
+)
+
+class TestUnicodeDammit(unittest.TestCase):
+    """Standalone tests of UnicodeDammit."""
+
+    def test_unicode_input(self):
+        markup = u"I'm already Unicode! \N{SNOWMAN}"
+        dammit = UnicodeDammit(markup)
+        self.assertEqual(dammit.unicode_markup, markup)
+
+    def test_smart_quotes_to_unicode(self):
+        markup = b"<foo>\x91\x92\x93\x94</foo>"
+        dammit = UnicodeDammit(markup)
+        self.assertEqual(
+            dammit.unicode_markup, u"<foo>\u2018\u2019\u201c\u201d</foo>")
+
+    def test_smart_quotes_to_xml_entities(self):
+        markup = b"<foo>\x91\x92\x93\x94</foo>"
+        dammit = UnicodeDammit(markup, smart_quotes_to="xml")
+        self.assertEqual(
+            dammit.unicode_markup, "<foo>&#x2018;&#x2019;&#x201C;&#x201D;</foo>")
+
+    def test_smart_quotes_to_html_entities(self):
+        markup = b"<foo>\x91\x92\x93\x94</foo>"
+        dammit = UnicodeDammit(markup, smart_quotes_to="html")
+        self.assertEqual(
+            dammit.unicode_markup, "<foo>&lsquo;&rsquo;&ldquo;&rdquo;</foo>")
+
+    def test_smart_quotes_to_ascii(self):
+        markup = b"<foo>\x91\x92\x93\x94</foo>"
+        dammit = UnicodeDammit(markup, smart_quotes_to="ascii")
+        self.assertEqual(
+            dammit.unicode_markup, """<foo>''""</foo>""")
+
+    def test_detect_utf8(self):
+        utf8 = b"Sacr\xc3\xa9 bleu! \xe2\x98\x83"
+        dammit = UnicodeDammit(utf8)
+        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
+        self.assertEqual(dammit.unicode_markup, u'Sacr\xe9 bleu! \N{SNOWMAN}')
+
+
+    def test_convert_hebrew(self):
+        hebrew = b"\xed\xe5\xec\xf9"
+        dammit = UnicodeDammit(hebrew, ["iso-8859-8"])
+        self.assertEqual(dammit.original_encoding.lower(), 'iso-8859-8')
+        self.assertEqual(dammit.unicode_markup, u'\u05dd\u05d5\u05dc\u05e9')
+
+    def test_dont_see_smart_quotes_where_there_are_none(self):
+        utf_8 = b"\343\202\261\343\203\274\343\202\277\343\202\244 Watch"
+        dammit = UnicodeDammit(utf_8)
+        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
+        self.assertEqual(dammit.unicode_markup.encode("utf-8"), utf_8)
+
+    def test_ignore_inappropriate_codecs(self):
+        utf8_data = u"Räksmörgås".encode("utf-8")
+        dammit = UnicodeDammit(utf8_data, ["iso-8859-8"])
+        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
+
+    def test_ignore_invalid_codecs(self):
+        utf8_data = u"Räksmörgås".encode("utf-8")
+        for bad_encoding in ['.utf8', '...', 'utF---16.!']:
+            dammit = UnicodeDammit(utf8_data, [bad_encoding])
+            self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
+
+    def test_exclude_encodings(self):
+        # This is UTF-8.
+        utf8_data = u"Räksmörgås".encode("utf-8")
+
+        # But if we exclude UTF-8 from consideration, the guess is
+        # Windows-1252.
+        dammit = UnicodeDammit(utf8_data, exclude_encodings=["utf-8"])
+        self.assertEqual(dammit.original_encoding.lower(), 'windows-1252')
+
+        # And if we exclude that, there is no valid guess at all.
+        dammit = UnicodeDammit(
+            utf8_data, exclude_encodings=["utf-8", "windows-1252"])
+        self.assertEqual(dammit.original_encoding, None)
+
+class TestEncodingDetector(unittest.TestCase):
+        
+    def test_encoding_detector_replaces_junk_in_encoding_name_with_replacement_character(self):
+        detected = EncodingDetector(
+            b'<?xml version="1.0" encoding="UTF-\xdb" ?>')
+        encodings = list(detected.encodings)
+        assert u'utf-\N{REPLACEMENT CHARACTER}' in encodings
+
+    def test_detect_html5_style_meta_tag(self):
+
+        for data in (
+            b'<html><meta charset="euc-jp" /></html>',
+            b"<html><meta charset='euc-jp' /></html>",
+            b"<html><meta charset=euc-jp /></html>",
+            b"<html><meta charset=euc-jp/></html>"):
+            dammit = UnicodeDammit(data, is_html=True)
+            self.assertEqual(
+                "euc-jp", dammit.original_encoding)
+
+    def test_last_ditch_entity_replacement(self):
+        # This is a UTF-8 document that contains bytestrings
+        # completely incompatible with UTF-8 (ie. encoded with some other
+        # encoding).
+        #
+        # Since there is no consistent encoding for the document,
+        # Unicode, Dammit will eventually encode the document as UTF-8
+        # and encode the incompatible characters as REPLACEMENT
+        # CHARACTER.
+        #
+        # If chardet is installed, it will detect that the document
+        # can be converted into ISO-8859-1 without errors. This happens
+        # to be the wrong encoding, but it is a consistent encoding, so the
+        # code we're testing here won't run.
+        #
+        # So we temporarily disable chardet if it's present.
+        doc = b"""\357\273\277<?xml version="1.0" encoding="UTF-8"?>
+<html><b>\330\250\330\252\330\261</b>
+<i>\310\322\321\220\312\321\355\344</i></html>"""
+        chardet = bs4.dammit.chardet_dammit
+        logging.disable(logging.WARNING)
+        try:
+            def noop(str):
+                return None
+            bs4.dammit.chardet_dammit = noop
+            dammit = UnicodeDammit(doc)
+            self.assertEqual(True, dammit.contains_replacement_characters)
+            self.assertTrue(u"\ufffd" in dammit.unicode_markup)
+
+            soup = BeautifulSoup(doc, "html.parser")
+            self.assertTrue(soup.contains_replacement_characters)
+        finally:
+            logging.disable(logging.NOTSET)
+            bs4.dammit.chardet_dammit = chardet
+
+    def test_byte_order_mark_removed(self):
+        # A document written in UTF-16LE will have its byte order marker stripped.
+        data = b'\xff\xfe<\x00a\x00>\x00\xe1\x00\xe9\x00<\x00/\x00a\x00>\x00'
+        dammit = UnicodeDammit(data)
+        self.assertEqual(u"<a>áé</a>", dammit.unicode_markup)
+        self.assertEqual("utf-16le", dammit.original_encoding)
+       
+    def test_known_definite_versus_user_encodings(self):
+        # The known_definite_encodings are used before sniffing the
+        # byte-order mark; the user_encodings are used afterwards.
+
+        # Here's a document in UTF-16LE.
+        data = b'\xff\xfe<\x00a\x00>\x00\xe1\x00\xe9\x00<\x00/\x00a\x00>\x00'
+        dammit = UnicodeDammit(data)
+
+        # We can process it as UTF-16 by passing it in as a known
+        # definite encoding.
+        before = UnicodeDammit(data, known_definite_encodings=["utf-16"])
+        self.assertEqual("utf-16", before.original_encoding)
+        
+        # If we pass UTF-18 as a user encoding, it's not even
+        # tried--the encoding sniffed from the byte-order mark takes
+        # precedence.
+        after = UnicodeDammit(data, user_encodings=["utf-8"])
+        self.assertEqual("utf-16le", after.original_encoding)
+        self.assertEqual(
+            ["utf-16le"], [x[0] for x in dammit.tried_encodings]
+        )
+        
+        # Here's a document in ISO-8859-8.
+        hebrew = b"\xed\xe5\xec\xf9"
+        dammit = UnicodeDammit(hebrew, known_definite_encodings=["utf-8"],
+                               user_encodings=["iso-8859-8"])
+        
+        # The known_definite_encodings don't work, BOM sniffing does
+        # nothing (it only works for a few UTF encodings), but one of
+        # the user_encodings does work.
+        self.assertEqual("iso-8859-8", dammit.original_encoding)
+        self.assertEqual(
+            ["utf-8", "iso-8859-8"], [x[0] for x in dammit.tried_encodings]
+        )
+        
+    def test_deprecated_override_encodings(self):
+        # override_encodings is a deprecated alias for
+        # known_definite_encodings.
+        hebrew = b"\xed\xe5\xec\xf9"
+        dammit = UnicodeDammit(
+            hebrew,
+            known_definite_encodings=["shift-jis"],
+            override_encodings=["utf-8"],
+            user_encodings=["iso-8859-8"],
+        )
+        self.assertEqual("iso-8859-8", dammit.original_encoding)
+
+        # known_definite_encodings and override_encodings were tried
+        # before user_encodings.
+        self.assertEqual(
+            ["shift-jis", "utf-8", "iso-8859-8"],
+            [x[0] for x in dammit.tried_encodings]
+        )
+
+    def test_detwingle(self):
+        # Here's a UTF8 document.
+        utf8 = (u"\N{SNOWMAN}" * 3).encode("utf8")
+
+        # Here's a Windows-1252 document.
+        windows_1252 = (
+            u"\N{LEFT DOUBLE QUOTATION MARK}Hi, I like Windows!"
+            u"\N{RIGHT DOUBLE QUOTATION MARK}").encode("windows_1252")
+
+        # Through some unholy alchemy, they've been stuck together.
+        doc = utf8 + windows_1252 + utf8
+
+        # The document can't be turned into UTF-8:
+        self.assertRaises(UnicodeDecodeError, doc.decode, "utf8")
+
+        # Unicode, Dammit thinks the whole document is Windows-1252,
+        # and decodes it into "☃☃☃“Hi, I like Windows!”☃☃☃"
+
+        # But if we run it through fix_embedded_windows_1252, it's fixed:
+
+        fixed = UnicodeDammit.detwingle(doc)
+        self.assertEqual(
+            u"☃☃☃“Hi, I like Windows!”☃☃☃", fixed.decode("utf8"))
+
+    def test_detwingle_ignores_multibyte_characters(self):
+        # Each of these characters has a UTF-8 representation ending
+        # in \x93. \x93 is a smart quote if interpreted as
+        # Windows-1252. But our code knows to skip over multibyte
+        # UTF-8 characters, so they'll survive the process unscathed.
+        for tricky_unicode_char in (
+            u"\N{LATIN SMALL LIGATURE OE}", # 2-byte char '\xc5\x93'
+            u"\N{LATIN SUBSCRIPT SMALL LETTER X}", # 3-byte char '\xe2\x82\x93'
+            u"\xf0\x90\x90\x93", # This is a CJK character, not sure which one.
+            ):
+            input = tricky_unicode_char.encode("utf8")
+            self.assertTrue(input.endswith(b'\x93'))
+            output = UnicodeDammit.detwingle(input)
+            self.assertEqual(output, input)
+
+    def test_find_declared_encoding(self):
+        # Test our ability to find a declared encoding inside an
+        # XML or HTML document.
+        #
+        # Even if the document comes in as Unicode, it may be
+        # interesting to know what encoding was claimed
+        # originally.
+
+        html_unicode = u'<html><head><meta charset="utf-8"></head></html>'
+        html_bytes = html_unicode.encode("ascii")
+
+        xml_unicode= u'<?xml version="1.0" encoding="ISO-8859-1" ?>'
+        xml_bytes = xml_unicode.encode("ascii")
+
+        m = EncodingDetector.find_declared_encoding
+        self.assertEquals(None, m(html_unicode, is_html=False))
+        self.assertEquals("utf-8", m(html_unicode, is_html=True))
+        self.assertEquals("utf-8", m(html_bytes, is_html=True))
+
+        self.assertEquals("iso-8859-1", m(xml_unicode))
+        self.assertEquals("iso-8859-1", m(xml_bytes))
+
+        # Normally, only the first few kilobytes of a document are checked for
+        # an encoding.
+        spacer = b' ' * 5000
+        self.assertEquals(None, m(spacer + html_bytes))
+        self.assertEquals(None, m(spacer + xml_bytes))
+
+        # But you can tell find_declared_encoding to search an entire
+        # HTML document.
+        self.assertEquals(
+            "utf-8",
+            m(spacer + html_bytes, is_html=True, search_entire_document=True)
+        )
+
+        # The XML encoding declaration has to be the very first thing
+        # in the document. We'll allow whitespace before the document
+        # starts, but nothing else.
+        self.assertEquals(
+            "iso-8859-1",
+            m(xml_bytes, search_entire_document=True)
+        )
+        self.assertEquals(
+            None, m(b'a' + xml_bytes, search_entire_document=True)
+        )
+            
diff --git a/bs4/tests/test_soup.py b/bs4/tests/test_soup.py
index 0603ce7..ddb6446 100644
--- a/bs4/tests/test_soup.py
+++ b/bs4/tests/test_soup.py
@@ -32,7 +32,6 @@ import bs4.dammit
 from bs4.dammit import (
     EntitySubstitution,
     UnicodeDammit,
-    EncodingDetector,
 )
 from bs4.testing import (
     default_builder,
@@ -478,226 +477,7 @@ class TestEncodingConversion(SoupTest):
         markup = u'<div><a \N{SNOWMAN}="snowman"></a></div>'
         self.assertEqual(self.soup(markup).div.encode("utf8"), markup.encode("utf8"))
 
-class TestUnicodeDammit(unittest.TestCase):
-    """Standalone tests of UnicodeDammit."""
 
-    def test_unicode_input(self):
-        markup = u"I'm already Unicode! \N{SNOWMAN}"
-        dammit = UnicodeDammit(markup)
-        self.assertEqual(dammit.unicode_markup, markup)
-
-    def test_smart_quotes_to_unicode(self):
-        markup = b"<foo>\x91\x92\x93\x94</foo>"
-        dammit = UnicodeDammit(markup)
-        self.assertEqual(
-            dammit.unicode_markup, u"<foo>\u2018\u2019\u201c\u201d</foo>")
-
-    def test_smart_quotes_to_xml_entities(self):
-        markup = b"<foo>\x91\x92\x93\x94</foo>"
-        dammit = UnicodeDammit(markup, smart_quotes_to="xml")
-        self.assertEqual(
-            dammit.unicode_markup, "<foo>&#x2018;&#x2019;&#x201C;&#x201D;</foo>")
-
-    def test_smart_quotes_to_html_entities(self):
-        markup = b"<foo>\x91\x92\x93\x94</foo>"
-        dammit = UnicodeDammit(markup, smart_quotes_to="html")
-        self.assertEqual(
-            dammit.unicode_markup, "<foo>&lsquo;&rsquo;&ldquo;&rdquo;</foo>")
-
-    def test_smart_quotes_to_ascii(self):
-        markup = b"<foo>\x91\x92\x93\x94</foo>"
-        dammit = UnicodeDammit(markup, smart_quotes_to="ascii")
-        self.assertEqual(
-            dammit.unicode_markup, """<foo>''""</foo>""")
-
-    def test_detect_utf8(self):
-        utf8 = b"Sacr\xc3\xa9 bleu! \xe2\x98\x83"
-        dammit = UnicodeDammit(utf8)
-        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-        self.assertEqual(dammit.unicode_markup, u'Sacr\xe9 bleu! \N{SNOWMAN}')
-
-
-    def test_convert_hebrew(self):
-        hebrew = b"\xed\xe5\xec\xf9"
-        dammit = UnicodeDammit(hebrew, ["iso-8859-8"])
-        self.assertEqual(dammit.original_encoding.lower(), 'iso-8859-8')
-        self.assertEqual(dammit.unicode_markup, u'\u05dd\u05d5\u05dc\u05e9')
-
-    def test_dont_see_smart_quotes_where_there_are_none(self):
-        utf_8 = b"\343\202\261\343\203\274\343\202\277\343\202\244 Watch"
-        dammit = UnicodeDammit(utf_8)
-        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-        self.assertEqual(dammit.unicode_markup.encode("utf-8"), utf_8)
-
-    def test_ignore_inappropriate_codecs(self):
-        utf8_data = u"Räksmörgås".encode("utf-8")
-        dammit = UnicodeDammit(utf8_data, ["iso-8859-8"])
-        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-
-    def test_ignore_invalid_codecs(self):
-        utf8_data = u"Räksmörgås".encode("utf-8")
-        for bad_encoding in ['.utf8', '...', 'utF---16.!']:
-            dammit = UnicodeDammit(utf8_data, [bad_encoding])
-            self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-
-    def test_exclude_encodings(self):
-        # This is UTF-8.
-        utf8_data = u"Räksmörgås".encode("utf-8")
-
-        # But if we exclude UTF-8 from consideration, the guess is
-        # Windows-1252.
-        dammit = UnicodeDammit(utf8_data, exclude_encodings=["utf-8"])
-        self.assertEqual(dammit.original_encoding.lower(), 'windows-1252')
-
-        # And if we exclude that, there is no valid guess at all.
-        dammit = UnicodeDammit(
-            utf8_data, exclude_encodings=["utf-8", "windows-1252"])
-        self.assertEqual(dammit.original_encoding, None)
-
-    def test_encoding_detector_replaces_junk_in_encoding_name_with_replacement_character(self):
-        detected = EncodingDetector(
-            b'<?xml version="1.0" encoding="UTF-\xdb" ?>')
-        encodings = list(detected.encodings)
-        assert u'utf-\N{REPLACEMENT CHARACTER}' in encodings
-
-    def test_detect_html5_style_meta_tag(self):
-
-        for data in (
-            b'<html><meta charset="euc-jp" /></html>',
-            b"<html><meta charset='euc-jp' /></html>",
-            b"<html><meta charset=euc-jp /></html>",
-            b"<html><meta charset=euc-jp/></html>"):
-            dammit = UnicodeDammit(data, is_html=True)
-            self.assertEqual(
-                "euc-jp", dammit.original_encoding)
-
-    def test_last_ditch_entity_replacement(self):
-        # This is a UTF-8 document that contains bytestrings
-        # completely incompatible with UTF-8 (ie. encoded with some other
-        # encoding).
-        #
-        # Since there is no consistent encoding for the document,
-        # Unicode, Dammit will eventually encode the document as UTF-8
-        # and encode the incompatible characters as REPLACEMENT
-        # CHARACTER.
-        #
-        # If chardet is installed, it will detect that the document
-        # can be converted into ISO-8859-1 without errors. This happens
-        # to be the wrong encoding, but it is a consistent encoding, so the
-        # code we're testing here won't run.
-        #
-        # So we temporarily disable chardet if it's present.
-        doc = b"""\357\273\277<?xml version="1.0" encoding="UTF-8"?>
-<html><b>\330\250\330\252\330\261</b>
-<i>\310\322\321\220\312\321\355\344</i></html>"""
-        chardet = bs4.dammit.chardet_dammit
-        logging.disable(logging.WARNING)
-        try:
-            def noop(str):
-                return None
-            bs4.dammit.chardet_dammit = noop
-            dammit = UnicodeDammit(doc)
-            self.assertEqual(True, dammit.contains_replacement_characters)
-            self.assertTrue(u"\ufffd" in dammit.unicode_markup)
-
-            soup = BeautifulSoup(doc, "html.parser")
-            self.assertTrue(soup.contains_replacement_characters)
-        finally:
-            logging.disable(logging.NOTSET)
-            bs4.dammit.chardet_dammit = chardet
-
-    def test_byte_order_mark_removed(self):
-        # A document written in UTF-16LE will have its byte order marker stripped.
-        data = b'\xff\xfe<\x00a\x00>\x00\xe1\x00\xe9\x00<\x00/\x00a\x00>\x00'
-        dammit = UnicodeDammit(data)
-        self.assertEqual(u"<a>áé</a>", dammit.unicode_markup)
-        self.assertEqual("utf-16le", dammit.original_encoding)
-
-    def test_detwingle(self):
-        # Here's a UTF8 document.
-        utf8 = (u"\N{SNOWMAN}" * 3).encode("utf8")
-
-        # Here's a Windows-1252 document.
-        windows_1252 = (
-            u"\N{LEFT DOUBLE QUOTATION MARK}Hi, I like Windows!"
-            u"\N{RIGHT DOUBLE QUOTATION MARK}").encode("windows_1252")
-
-        # Through some unholy alchemy, they've been stuck together.
-        doc = utf8 + windows_1252 + utf8
-
-        # The document can't be turned into UTF-8:
-        self.assertRaises(UnicodeDecodeError, doc.decode, "utf8")
-
-        # Unicode, Dammit thinks the whole document is Windows-1252,
-        # and decodes it into "☃☃☃“Hi, I like Windows!”☃☃☃"
-
-        # But if we run it through fix_embedded_windows_1252, it's fixed:
-
-        fixed = UnicodeDammit.detwingle(doc)
-        self.assertEqual(
-            u"☃☃☃“Hi, I like Windows!”☃☃☃", fixed.decode("utf8"))
-
-    def test_detwingle_ignores_multibyte_characters(self):
-        # Each of these characters has a UTF-8 representation ending
-        # in \x93. \x93 is a smart quote if interpreted as
-        # Windows-1252. But our code knows to skip over multibyte
-        # UTF-8 characters, so they'll survive the process unscathed.
-        for tricky_unicode_char in (
-            u"\N{LATIN SMALL LIGATURE OE}", # 2-byte char '\xc5\x93'
-            u"\N{LATIN SUBSCRIPT SMALL LETTER X}", # 3-byte char '\xe2\x82\x93'
-            u"\xf0\x90\x90\x93", # This is a CJK character, not sure which one.
-            ):
-            input = tricky_unicode_char.encode("utf8")
-            self.assertTrue(input.endswith(b'\x93'))
-            output = UnicodeDammit.detwingle(input)
-            self.assertEqual(output, input)
-
-    def test_find_declared_encoding(self):
-        # Test our ability to find a declared encoding inside an
-        # XML or HTML document.
-        #
-        # Even if the document comes in as Unicode, it may be
-        # interesting to know what encoding was claimed
-        # originally.
-
-        html_unicode = u'<html><head><meta charset="utf-8"></head></html>'
-        html_bytes = html_unicode.encode("ascii")
-
-        xml_unicode= u'<?xml version="1.0" encoding="ISO-8859-1" ?>'
-        xml_bytes = xml_unicode.encode("ascii")
-
-        m = EncodingDetector.find_declared_encoding
-        self.assertEquals(None, m(html_unicode, is_html=False))
-        self.assertEquals("utf-8", m(html_unicode, is_html=True))
-        self.assertEquals("utf-8", m(html_bytes, is_html=True))
-
-        self.assertEquals("iso-8859-1", m(xml_unicode))
-        self.assertEquals("iso-8859-1", m(xml_bytes))
-
-        # Normally, only the first few kilobytes of a document are checked for
-        # an encoding.
-        spacer = b' ' * 5000
-        self.assertEquals(None, m(spacer + html_bytes))
-        self.assertEquals(None, m(spacer + xml_bytes))
-
-        # But you can tell find_declared_encoding to search an entire
-        # HTML document.
-        self.assertEquals(
-            "utf-8",
-            m(spacer + html_bytes, is_html=True, search_entire_document=True)
-        )
-
-        # The XML encoding declaration has to be the very first thing
-        # in the document. We'll allow whitespace before the document
-        # starts, but nothing else.
-        self.assertEquals(
-            "iso-8859-1",
-            m(xml_bytes, search_entire_document=True)
-        )
-        self.assertEquals(
-            None, m(b'a' + xml_bytes, search_entire_document=True)
-        )
-            
 class TestNamedspacedAttribute(SoupTest):
 
     def test_name_may_be_none_or_missing(self):