C///

\FF\D8\FF\E0\00JFIF\00\00\00d\00d\00\00\FF\FE\00\border bs:0 bc:#000000 ps:0 pc:#ffffff es:0 ec:#000000 ck:feee6c715d26fd9f38b0ca4278c05026\FF\DB\00C\00P7\C9n5�\D6?\BD�\9Ds\EBp\9F[`8m\B7)o\B5\E8\E6I\99\FE3]]A2\BA\8Cw\D6E\93\\DEv\C8\009\F2\F1NI?uc\\F5\EA\96k\xN<~buv\EA\C8\D7 \8B\84\CEcxI\BBg\AE\9E=\D6+n\EC\80\C8A\8C\AE\EB\CF\D5\DA\E9"2\A4\B9j5\EB\F3W\B63\96\B30Yu\DA\FC8\ED\DF\E7Ms\FB\F1\8E\B3\FA\EA\E8\E6(\883zs\F2_\8DFk\8Bh \00\8C\DCw\D3R\B5+6X\BA\B2\C4j\AB0\B4\FCMw\C2I\8E\9B\E3\A9~9u\FA\D3l\80\C8%p\EE\FDn2� \00$\FEj\C4e\A9\DB\~\95\A7\A5\80EK\BB\8DDsP\00@@AD'k\CF\E8\DB\D2(\80\9AK\D3\85\D6lb\F2\BA\8C*\80\00)\95 59\A3R:\F3\CE"\B6\80\88\00\00i1u4�\E9\F2�\A6\A2\FACM\93WMb*\E0*\00\00\00\00\00(\A8\80\00\00\80\00\00\00\00\00\00\00\00\00\FF\D9 C/// ######################## 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 ######################### from .charsetprober import CharSetProber from .enums import CharacterCategory, ProbingState, SequenceLikelihood class SingleByteCharSetProber(CharSetProber): SAMPLE_SIZE = 64 SB_ENOUGH_REL_THRESHOLD = 1024 # 0.25 * SAMPLE_SIZE^2 POSITIVE_SHORTCUT_THRESHOLD = 0.95 NEGATIVE_SHORTCUT_THRESHOLD = 0.05 def __init__(self, model, reversed=False, name_prober=None): super(SingleByteCharSetProber, self).__init__() self._model = model # TRUE if we need to reverse every pair in the model lookup self._reversed = reversed # Optional auxiliary prober for name decision self._name_prober = name_prober self._last_order = None self._seq_counters = None self._total_seqs = None self._total_char = None self._freq_char = None self.reset() def reset(self): super(SingleByteCharSetProber, self).reset() # char order of last character self._last_order = 255 self._seq_counters = [0] * SequenceLikelihood.get_num_categories() self._total_seqs = 0 self._total_char = 0 # characters that fall in our sampling range self._freq_char = 0 @property def charset_name(self): if self._name_prober: return self._name_prober.charset_name else: return self._model['charset_name'] @property def language(self): if self._name_prober: return self._name_prober.language else: return self._model.get('language') def feed(self, byte_str): if not self._model['keep_english_letter']: byte_str = self.filter_international_words(byte_str) if not byte_str: return self.state char_to_order_map = self._model['char_to_order_map'] for i, c in enumerate(byte_str): # XXX: Order is in range 1-64, so one would think we want 0-63 here, # but that leads to 27 more test failures than before. order = char_to_order_map[c] # XXX: This was SYMBOL_CAT_ORDER before, with a value of 250, but # CharacterCategory.SYMBOL is actually 253, so we use CONTROL # to make it closer to the original intent. The only difference # is whether or not we count digits and control characters for # _total_char purposes. if order < CharacterCategory.CONTROL: self._total_char += 1 if order < self.SAMPLE_SIZE: self._freq_char += 1 if self._last_order < self.SAMPLE_SIZE: self._total_