xplia 1.0.1__py3-none-any.whl

venv/Lib/site-packages/pip/_vendor/chardet/sjisprober.py

@@ -0,0 +1,105 @@
+######################## BEGIN LICENSE BLOCK ########################
+# The Original Code is mozilla.org code.
+#
+# The Initial Developer of the Original Code is
+# Netscape Communications Corporation.
+# Portions created by the Initial Developer are Copyright (C) 1998
+# the Initial Developer. All Rights Reserved.
+#
+# Contributor(s):
+#   Mark Pilgrim - port to Python
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 of the License, or (at your option) any later version.
+#
+# This library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+# 02110-1301  USA
+######################### END LICENSE BLOCK #########################
+
+from typing import Union
+
+from .chardistribution import SJISDistributionAnalysis
+from .codingstatemachine import CodingStateMachine
+from .enums import MachineState, ProbingState
+from .jpcntx import SJISContextAnalysis
+from .mbcharsetprober import MultiByteCharSetProber
+from .mbcssm import SJIS_SM_MODEL
+
+
+class SJISProber(MultiByteCharSetProber):
+    def __init__(self) -> None:
+        super().__init__()
+        self.coding_sm = CodingStateMachine(SJIS_SM_MODEL)
+        self.distribution_analyzer = SJISDistributionAnalysis()
+        self.context_analyzer = SJISContextAnalysis()
+        self.reset()
+
+    def reset(self) -> None:
+        super().reset()
+        self.context_analyzer.reset()
+
+    @property
+    def charset_name(self) -> str:
+        return self.context_analyzer.charset_name
+
+    @property
+    def language(self) -> str:
+        return "Japanese"
+
+    def feed(self, byte_str: Union[bytes, bytearray]) -> ProbingState:
+        assert self.coding_sm is not None
+        assert self.distribution_analyzer is not None
+
+        for i, byte in enumerate(byte_str):
+            coding_state = self.coding_sm.next_state(byte)
+            if coding_state == MachineState.ERROR:
+                self.logger.debug(
+                    "%s %s prober hit error at byte %s",
+                    self.charset_name,
+                    self.language,
+                    i,
+                )
+                self._state = ProbingState.NOT_ME
+                break
+            if coding_state == MachineState.ITS_ME:
+                self._state = ProbingState.FOUND_IT
+                break
+            if coding_state == MachineState.START:
+                char_len = self.coding_sm.get_current_charlen()
+                if i == 0:
+                    self._last_char[1] = byte
+                    self.context_analyzer.feed(
+                        self._last_char[2 - char_len :], char_len
+                    )
+                    self.distribution_analyzer.feed(self._last_char, char_len)
+                else:
+                    self.context_analyzer.feed(
+                        byte_str[i + 1 - char_len : i + 3 - char_len], char_len
+                    )
+                    self.distribution_analyzer.feed(byte_str[i - 1 : i + 1], char_len)
+
+        self._last_char[0] = byte_str[-1]
+
+        if self.state == ProbingState.DETECTING:
+            if self.context_analyzer.got_enough_data() and (
+                self.get_confidence() > self.SHORTCUT_THRESHOLD
+            ):
+                self._state = ProbingState.FOUND_IT
+
+        return self.state
+
+    def get_confidence(self) -> float:
+        assert self.distribution_analyzer is not None
+
+        context_conf = self.context_analyzer.get_confidence()
+        distrib_conf = self.distribution_analyzer.get_confidence()
+        return max(context_conf, distrib_conf)

venv/Lib/site-packages/pip/_vendor/chardet/universaldetector.py

@@ -0,0 +1,362 @@
+######################## BEGIN LICENSE BLOCK ########################
+# The Original Code is Mozilla Universal charset detector code.
+#
+# The Initial Developer of the Original Code is
+# Netscape Communications Corporation.
+# Portions created by the Initial Developer are Copyright (C) 2001
+# the Initial Developer. All Rights Reserved.
+#
+# Contributor(s):
+#   Mark Pilgrim - port to Python
+#   Shy Shalom - original C code
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 of the License, or (at your option) any later version.
+#
+# This library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+# 02110-1301  USA
+######################### END LICENSE BLOCK #########################
+"""
+Module containing the UniversalDetector detector class, which is the primary
+class a user of ``chardet`` should use.
+
+:author: Mark Pilgrim (initial port to Python)
+:author: Shy Shalom (original C code)
+:author: Dan Blanchard (major refactoring for 3.0)
+:author: Ian Cordasco
+"""
+
+
+import codecs
+import logging
+import re
+from typing import List, Optional, Union
+
+from .charsetgroupprober import CharSetGroupProber
+from .charsetprober import CharSetProber
+from .enums import InputState, LanguageFilter, ProbingState
+from .escprober import EscCharSetProber
+from .latin1prober import Latin1Prober
+from .macromanprober import MacRomanProber
+from .mbcsgroupprober import MBCSGroupProber
+from .resultdict import ResultDict
+from .sbcsgroupprober import SBCSGroupProber
+from .utf1632prober import UTF1632Prober
+
+
+class UniversalDetector:
+    """
+    The ``UniversalDetector`` class underlies the ``chardet.detect`` function
+    and coordinates all of the different charset probers.
+
+    To get a ``dict`` containing an encoding and its confidence, you can simply
+    run:
+
+    .. code::
+
+            u = UniversalDetector()
+            u.feed(some_bytes)
+            u.close()
+            detected = u.result
+
+    """
+
+    MINIMUM_THRESHOLD = 0.20
+    HIGH_BYTE_DETECTOR = re.compile(b"[\x80-\xFF]")
+    ESC_DETECTOR = re.compile(b"(\033|~{)")
+    WIN_BYTE_DETECTOR = re.compile(b"[\x80-\x9F]")
+    ISO_WIN_MAP = {
+        "iso-8859-1": "Windows-1252",
+        "iso-8859-2": "Windows-1250",
+        "iso-8859-5": "Windows-1251",
+        "iso-8859-6": "Windows-1256",
+        "iso-8859-7": "Windows-1253",
+        "iso-8859-8": "Windows-1255",
+        "iso-8859-9": "Windows-1254",
+        "iso-8859-13": "Windows-1257",
+    }
+    # Based on https://encoding.spec.whatwg.org/#names-and-labels
+    # but altered to match Python names for encodings and remove mappings
+    # that break tests.
+    LEGACY_MAP = {
+        "ascii": "Windows-1252",
+        "iso-8859-1": "Windows-1252",
+        "tis-620": "ISO-8859-11",
+        "iso-8859-9": "Windows-1254",
+        "gb2312": "GB18030",
+        "euc-kr": "CP949",
+        "utf-16le": "UTF-16",
+    }
+
+    def __init__(
+        self,
+        lang_filter: LanguageFilter = LanguageFilter.ALL,
+        should_rename_legacy: bool = False,
+    ) -> None:
+        self._esc_charset_prober: Optional[EscCharSetProber] = None
+        self._utf1632_prober: Optional[UTF1632Prober] = None
+        self._charset_probers: List[CharSetProber] = []
+        self.result: ResultDict = {
+            "encoding": None,
+            "confidence": 0.0,
+            "language": None,
+        }
+        self.done = False
+        self._got_data = False
+        self._input_state = InputState.PURE_ASCII
+        self._last_char = b""
+        self.lang_filter = lang_filter
+        self.logger = logging.getLogger(__name__)
+        self._has_win_bytes = False
+        self.should_rename_legacy = should_rename_legacy
+        self.reset()
+
+    @property
+    def input_state(self) -> int:
+        return self._input_state
+
+    @property
+    def has_win_bytes(self) -> bool:
+        return self._has_win_bytes
+
+    @property
+    def charset_probers(self) -> List[CharSetProber]:
+        return self._charset_probers
+
+    def reset(self) -> None:
+        """
+        Reset the UniversalDetector and all of its probers back to their
+        initial states.  This is called by ``__init__``, so you only need to
+        call this directly in between analyses of different documents.
+        """
+        self.result = {"encoding": None, "confidence": 0.0, "language": None}
+        self.done = False
+        self._got_data = False
+        self._has_win_bytes = False
+        self._input_state = InputState.PURE_ASCII
+        self._last_char = b""
+        if self._esc_charset_prober:
+            self._esc_charset_prober.reset()
+        if self._utf1632_prober:
+            self._utf1632_prober.reset()
+        for prober in self._charset_probers:
+            prober.reset()
+
+    def feed(self, byte_str: Union[bytes, bytearray]) -> None:
+        """
+        Takes a chunk of a document and feeds it through all of the relevant
+        charset probers.
+
+        After calling ``feed``, you can check the value of the ``done``
+        attribute to see if you need to continue feeding the
+        ``UniversalDetector`` more data, or if it has made a prediction
+        (in the ``result`` attribute).
+
+        .. note::
+           You should always call ``close`` when you're done feeding in your
+           document if ``done`` is not already ``True``.
+        """
+        if self.done:
+            return
+
+        if not byte_str:
+            return
+
+        if not isinstance(byte_str, bytearray):
+            byte_str = bytearray(byte_str)
+
+        # First check for known BOMs, since these are guaranteed to be correct
+        if not self._got_data:
+            # If the data starts with BOM, we know it is UTF
+            if byte_str.startswith(codecs.BOM_UTF8):
+                # EF BB BF  UTF-8 with BOM
+                self.result = {
+                    "encoding": "UTF-8-SIG",
+                    "confidence": 1.0,
+                    "language": "",
+                }
+            elif byte_str.startswith((codecs.BOM_UTF32_LE, codecs.BOM_UTF32_BE)):
+                # FF FE 00 00  UTF-32, little-endian BOM
+                # 00 00 FE FF  UTF-32, big-endian BOM
+                self.result = {"encoding": "UTF-32", "confidence": 1.0, "language": ""}
+            elif byte_str.startswith(b"\xFE\xFF\x00\x00"):
+                # FE FF 00 00  UCS-4, unusual octet order BOM (3412)
+                self.result = {
+                    # TODO: This encoding is not supported by Python. Should remove?
+                    "encoding": "X-ISO-10646-UCS-4-3412",
+                    "confidence": 1.0,
+                    "language": "",
+                }
+            elif byte_str.startswith(b"\x00\x00\xFF\xFE"):
+                # 00 00 FF FE  UCS-4, unusual octet order BOM (2143)
+                self.result = {
+                    # TODO: This encoding is not supported by Python. Should remove?
+                    "encoding": "X-ISO-10646-UCS-4-2143",
+                    "confidence": 1.0,
+                    "language": "",
+                }
+            elif byte_str.startswith((codecs.BOM_LE, codecs.BOM_BE)):
+                # FF FE  UTF-16, little endian BOM
+                # FE FF  UTF-16, big endian BOM
+                self.result = {"encoding": "UTF-16", "confidence": 1.0, "language": ""}
+
+            self._got_data = True
+            if self.result["encoding"] is not None:
+                self.done = True
+                return
+
+        # If none of those matched and we've only see ASCII so far, check
+        # for high bytes and escape sequences
+        if self._input_state == InputState.PURE_ASCII:
+            if self.HIGH_BYTE_DETECTOR.search(byte_str):
+                self._input_state = InputState.HIGH_BYTE
+            elif (
+                self._input_state == InputState.PURE_ASCII
+                and self.ESC_DETECTOR.search(self._last_char + byte_str)
+            ):
+                self._input_state = InputState.ESC_ASCII
+
+        self._last_char = byte_str[-1:]
+
+        # next we will look to see if it is appears to be either a UTF-16 or
+        # UTF-32 encoding
+        if not self._utf1632_prober:
+            self._utf1632_prober = UTF1632Prober()
+
+        if self._utf1632_prober.state == ProbingState.DETECTING:
+            if self._utf1632_prober.feed(byte_str) == ProbingState.FOUND_IT:
+                self.result = {
+                    "encoding": self._utf1632_prober.charset_name,
+                    "confidence": self._utf1632_prober.get_confidence(),
+                    "language": "",
+                }
+                self.done = True
+                return
+
+        # If we've seen escape sequences, use the EscCharSetProber, which
+        # uses a simple state machine to check for known escape sequences in
+        # HZ and ISO-2022 encodings, since those are the only encodings that
+        # use such sequences.
+        if self._input_state == InputState.ESC_ASCII:
+            if not self._esc_charset_prober:
+                self._esc_charset_prober = EscCharSetProber(self.lang_filter)
+            if self._esc_charset_prober.feed(byte_str) == ProbingState.FOUND_IT:
+                self.result = {
+                    "encoding": self._esc_charset_prober.charset_name,
+                    "confidence": self._esc_charset_prober.get_confidence(),
+                    "language": self._esc_charset_prober.language,
+                }
+                self.done = True
+        # If we've seen high bytes (i.e., those with values greater than 127),
+        # we need to do more complicated checks using all our multi-byte and
+        # single-byte probers that are left.  The single-byte probers
+        # use character bigram distributions to determine the encoding, whereas
+        # the multi-byte probers use a combination of character unigram and
+        # bigram distributions.
+        elif self._input_state == InputState.HIGH_BYTE:
+            if not self._charset_probers:
+                self._charset_probers = [MBCSGroupProber(self.lang_filter)]
+                # If we're checking non-CJK encodings, use single-byte prober
+                if self.lang_filter & LanguageFilter.NON_CJK:
+                    self._charset_probers.append(SBCSGroupProber())
+                self._charset_probers.append(Latin1Prober())
+                self._charset_probers.append(MacRomanProber())
+            for prober in self._charset_probers:
+                if prober.feed(byte_str) == ProbingState.FOUND_IT:
+                    self.result = {
+                        "encoding": prober.charset_name,
+                        "confidence": prober.get_confidence(),
+                        "language": prober.language,
+                    }
+                    self.done = True
+                    break
+            if self.WIN_BYTE_DETECTOR.search(byte_str):
+                self._has_win_bytes = True
+
+    def close(self) -> ResultDict:
+        """
+        Stop analyzing the current document and come up with a final
+        prediction.
+
+        :returns:  The ``result`` attribute, a ``dict`` with the keys
+                   `encoding`, `confidence`, and `language`.
+        """
+        # Don't bother with checks if we're already done
+        if self.done:
+            return self.result
+        self.done = True
+
+        if not self._got_data:
+            self.logger.debug("no data received!")
+
+        # Default to ASCII if it is all we've seen so far
+        elif self._input_state == InputState.PURE_ASCII:
+            self.result = {"encoding": "ascii", "confidence": 1.0, "language": ""}
+
+        # If we have seen non-ASCII, return the best that met MINIMUM_THRESHOLD
+        elif self._input_state == InputState.HIGH_BYTE:
+            prober_confidence = None
+            max_prober_confidence = 0.0
+            max_prober = None
+            for prober in self._charset_probers:
+                if not prober:
+                    continue
+                prober_confidence = prober.get_confidence()
+                if prober_confidence > max_prober_confidence:
+                    max_prober_confidence = prober_confidence
+                    max_prober = prober
+            if max_prober and (max_prober_confidence > self.MINIMUM_THRESHOLD):
+                charset_name = max_prober.charset_name
+                assert charset_name is not None
+                lower_charset_name = charset_name.lower()
+                confidence = max_prober.get_confidence()
+                # Use Windows encoding name instead of ISO-8859 if we saw any
+                # extra Windows-specific bytes
+                if lower_charset_name.startswith("iso-8859"):
+                    if self._has_win_bytes:
+                        charset_name = self.ISO_WIN_MAP.get(
+                            lower_charset_name, charset_name
+                        )
+                # Rename legacy encodings with superset encodings if asked
+                if self.should_rename_legacy:
+                    charset_name = self.LEGACY_MAP.get(
+                        (charset_name or "").lower(), charset_name
+                    )
+                self.result = {
+                    "encoding": charset_name,
+                    "confidence": confidence,
+                    "language": max_prober.language,
+                }
+
+        # Log all prober confidences if none met MINIMUM_THRESHOLD
+        if self.logger.getEffectiveLevel() <= logging.DEBUG:
+            if self.result["encoding"] is None:
+                self.logger.debug("no probers hit minimum threshold")
+                for group_prober in self._charset_probers:
+                    if not group_prober:
+                        continue
+                    if isinstance(group_prober, CharSetGroupProber):
+                        for prober in group_prober.probers:
+                            self.logger.debug(
+                                "%s %s confidence = %s",
+                                prober.charset_name,
+                                prober.language,
+                                prober.get_confidence(),
+                            )
+                    else:
+                        self.logger.debug(
+                            "%s %s confidence = %s",
+                            group_prober.charset_name,
+                            group_prober.language,
+                            group_prober.get_confidence(),
+                        )
+        return self.result

venv/Lib/site-packages/pip/_vendor/chardet/utf1632prober.py

@@ -0,0 +1,225 @@
+######################## BEGIN LICENSE BLOCK ########################
+#
+# Contributor(s):
+#   Jason Zavaglia
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 of the License, or (at your option) any later version.
+#
+# This library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+# 02110-1301  USA
+######################### END LICENSE BLOCK #########################
+from typing import List, Union
+
+from .charsetprober import CharSetProber
+from .enums import ProbingState
+
+
+class UTF1632Prober(CharSetProber):
+    """
+    This class simply looks for occurrences of zero bytes, and infers
+    whether the file is UTF16 or UTF32 (low-endian or big-endian)
+    For instance, files looking like ( \0 \0 \0 [nonzero] )+
+    have a good probability to be UTF32BE.  Files looking like ( \0 [nonzero] )+
+    may be guessed to be UTF16BE, and inversely for little-endian varieties.
+    """
+
+    # how many logical characters to scan before feeling confident of prediction
+    MIN_CHARS_FOR_DETECTION = 20
+    # a fixed constant ratio of expected zeros or non-zeros in modulo-position.
+    EXPECTED_RATIO = 0.94
+
+    def __init__(self) -> None:
+        super().__init__()
+        self.position = 0
+        self.zeros_at_mod = [0] * 4
+        self.nonzeros_at_mod = [0] * 4
+        self._state = ProbingState.DETECTING
+        self.quad = [0, 0, 0, 0]
+        self.invalid_utf16be = False
+        self.invalid_utf16le = False
+        self.invalid_utf32be = False
+        self.invalid_utf32le = False
+        self.first_half_surrogate_pair_detected_16be = False
+        self.first_half_surrogate_pair_detected_16le = False
+        self.reset()
+
+    def reset(self) -> None:
+        super().reset()
+        self.position = 0
+        self.zeros_at_mod = [0] * 4
+        self.nonzeros_at_mod = [0] * 4
+        self._state = ProbingState.DETECTING
+        self.invalid_utf16be = False
+        self.invalid_utf16le = False
+        self.invalid_utf32be = False
+        self.invalid_utf32le = False
+        self.first_half_surrogate_pair_detected_16be = False
+        self.first_half_surrogate_pair_detected_16le = False
+        self.quad = [0, 0, 0, 0]
+
+    @property
+    def charset_name(self) -> str:
+        if self.is_likely_utf32be():
+            return "utf-32be"
+        if self.is_likely_utf32le():
+            return "utf-32le"
+        if self.is_likely_utf16be():
+            return "utf-16be"
+        if self.is_likely_utf16le():
+            return "utf-16le"
+        # default to something valid
+        return "utf-16"
+
+    @property
+    def language(self) -> str:
+        return ""
+
+    def approx_32bit_chars(self) -> float:
+        return max(1.0, self.position / 4.0)
+
+    def approx_16bit_chars(self) -> float:
+        return max(1.0, self.position / 2.0)
+
+    def is_likely_utf32be(self) -> bool:
+        approx_chars = self.approx_32bit_chars()
+        return approx_chars >= self.MIN_CHARS_FOR_DETECTION and (
+            self.zeros_at_mod[0] / approx_chars > self.EXPECTED_RATIO
+            and self.zeros_at_mod[1] / approx_chars > self.EXPECTED_RATIO
+            and self.zeros_at_mod[2] / approx_chars > self.EXPECTED_RATIO
+            and self.nonzeros_at_mod[3] / approx_chars > self.EXPECTED_RATIO
+            and not self.invalid_utf32be
+        )
+
+    def is_likely_utf32le(self) -> bool:
+        approx_chars = self.approx_32bit_chars()
+        return approx_chars >= self.MIN_CHARS_FOR_DETECTION and (
+            self.nonzeros_at_mod[0] / approx_chars > self.EXPECTED_RATIO
+            and self.zeros_at_mod[1] / approx_chars > self.EXPECTED_RATIO
+            and self.zeros_at_mod[2] / approx_chars > self.EXPECTED_RATIO
+            and self.zeros_at_mod[3] / approx_chars > self.EXPECTED_RATIO
+            and not self.invalid_utf32le
+        )
+
+    def is_likely_utf16be(self) -> bool:
+        approx_chars = self.approx_16bit_chars()
+        return approx_chars >= self.MIN_CHARS_FOR_DETECTION and (
+            (self.nonzeros_at_mod[1] + self.nonzeros_at_mod[3]) / approx_chars
+            > self.EXPECTED_RATIO
+            and (self.zeros_at_mod[0] + self.zeros_at_mod[2]) / approx_chars
+            > self.EXPECTED_RATIO
+            and not self.invalid_utf16be
+        )
+
+    def is_likely_utf16le(self) -> bool:
+        approx_chars = self.approx_16bit_chars()
+        return approx_chars >= self.MIN_CHARS_FOR_DETECTION and (
+            (self.nonzeros_at_mod[0] + self.nonzeros_at_mod[2]) / approx_chars
+            > self.EXPECTED_RATIO
+            and (self.zeros_at_mod[1] + self.zeros_at_mod[3]) / approx_chars
+            > self.EXPECTED_RATIO
+            and not self.invalid_utf16le
+        )
+
+    def validate_utf32_characters(self, quad: List[int]) -> None:
+        """
+        Validate if the quad of bytes is valid UTF-32.
+
+        UTF-32 is valid in the range 0x00000000 - 0x0010FFFF
+        excluding 0x0000D800 - 0x0000DFFF
+
+        https://en.wikipedia.org/wiki/UTF-32
+        """
+        if (
+            quad[0] != 0
+            or quad[1] > 0x10
+            or (quad[0] == 0 and quad[1] == 0 and 0xD8 <= quad[2] <= 0xDF)
+        ):
+            self.invalid_utf32be = True
+        if (
+            quad[3] != 0
+            or quad[2] > 0x10
+            or (quad[3] == 0 and quad[2] == 0 and 0xD8 <= quad[1] <= 0xDF)
+        ):
+            self.invalid_utf32le = True
+
+    def validate_utf16_characters(self, pair: List[int]) -> None:
+        """
+        Validate if the pair of bytes is  valid UTF-16.
+
+        UTF-16 is valid in the range 0x0000 - 0xFFFF excluding 0xD800 - 0xFFFF
+        with an exception for surrogate pairs, which must be in the range
+        0xD800-0xDBFF followed by 0xDC00-0xDFFF
+
+        https://en.wikipedia.org/wiki/UTF-16
+        """
+        if not self.first_half_surrogate_pair_detected_16be:
+            if 0xD8 <= pair[0] <= 0xDB:
+                self.first_half_surrogate_pair_detected_16be = True
+            elif 0xDC <= pair[0] <= 0xDF:
+                self.invalid_utf16be = True
+        else:
+            if 0xDC <= pair[0] <= 0xDF:
+                self.first_half_surrogate_pair_detected_16be = False
+            else:
+                self.invalid_utf16be = True
+
+        if not self.first_half_surrogate_pair_detected_16le:
+            if 0xD8 <= pair[1] <= 0xDB:
+                self.first_half_surrogate_pair_detected_16le = True
+            elif 0xDC <= pair[1] <= 0xDF:
+                self.invalid_utf16le = True
+        else:
+            if 0xDC <= pair[1] <= 0xDF:
+                self.first_half_surrogate_pair_detected_16le = False
+            else:
+                self.invalid_utf16le = True
+
+    def feed(self, byte_str: Union[bytes, bytearray]) -> ProbingState:
+        for c in byte_str:
+            mod4 = self.position % 4
+            self.quad[mod4] = c
+            if mod4 == 3:
+                self.validate_utf32_characters(self.quad)
+                self.validate_utf16_characters(self.quad[0:2])
+                self.validate_utf16_characters(self.quad[2:4])
+            if c == 0:
+                self.zeros_at_mod[mod4] += 1
+            else:
+                self.nonzeros_at_mod[mod4] += 1
+            self.position += 1
+        return self.state
+
+    @property
+    def state(self) -> ProbingState:
+        if self._state in {ProbingState.NOT_ME, ProbingState.FOUND_IT}:
+            # terminal, decided states
+            return self._state
+        if self.get_confidence() > 0.80:
+            self._state = ProbingState.FOUND_IT
+        elif self.position > 4 * 1024:
+            # if we get to 4kb into the file, and we can't conclude it's UTF,
+            # let's give up
+            self._state = ProbingState.NOT_ME
+        return self._state
+
+    def get_confidence(self) -> float:
+        return (
+            0.85
+            if (
+                self.is_likely_utf16le()
+                or self.is_likely_utf16be()
+                or self.is_likely_utf32le()
+                or self.is_likely_utf32be()
+            )
+            else 0.00
+        )