create_regex_tables

   1 # Copyright (C) 2010 Apple Inc. All rights reserved.
   2 #
   3 # Redistribution and use in source and binary forms, with or without
   4 # modification, are permitted provided that the following conditions
   5 # are met:
   6 # 1. Redistributions of source code must retain the above copyright
   7 #    notice, this list of conditions and the following disclaimer.
   8 # 2. Redistributions in binary form must reproduce the above copyright
   9 #    notice, this list of conditions and the following disclaimer in the
  10 #    documentation and/or other materials provided with the distribution.
  11 #
  12 # THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
  13 # EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  14 # IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
  15 # PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
  16 # CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
  17 # EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  18 # PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
  19 # PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
  20 # OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  21 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  22 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  23
  24 import sys
  25
  26 types = {
  27     "wordchar": { "UseTable" : True, "data": ['_', ('0','9'), ('A', 'Z'), ('a','z')]},
  28     "nonwordchar": { "UseTable" : True, "Inverse": "wordchar", "data": ['`', (0, ord('0') - 1), (ord('9') + 1, ord('A') - 1), (ord('Z') + 1, ord('_') - 1), (ord('z') + 1, 0xffff)]},
  29     "newline": { "UseTable" : False, "data": ['\n', '\r', 0x2028, 0x2029]},
  30     "spaces": { "UseTable" : True, "data": [' ', ('\t', '\r'), 0xa0, 0x1680, 0x180e, 0x2028, 0x2029, 0x202f, 0x205f, 0x3000, (0x2000, 0x200a), 0xfeff]},
  31     "nonspaces": { "UseTable" : True, "Inverse": "spaces", "data": [(0, ord('\t') - 1), (ord('\r') + 1, ord(' ') - 1), (ord(' ') + 1, 0x009f), (0x00a1, 0x167f), (0x1681, 0x180d), (0x180f, 0x1fff), (0x200b, 0x2027), (0x202a, 0x202e), (0x2030, 0x205e), (0x2060, 0x2fff), (0x3001, 0xfefe), (0xff00, 0xffff)]},
  32     "digits": { "UseTable" : False, "data": [('0', '9')]},
  33     "nondigits": { "UseTable" : False, "Inverse": "digits", "data": [(0, ord('0') - 1), (ord('9') + 1, 0xffff)] }
  34 }
  35 entriesPerLine = 50
  36 arrays = "";
  37 functions = "";
  38 emitTables = (len(sys.argv) < 2 or sys.argv[1] != "--no-tables")
  39
  40 for name, classes in types.items():
  41     ranges = [];
  42     size = 0;
  43     for _class in classes["data"]:
  44         if type(_class) == str:
  45             ranges.append((ord(_class), ord(_class)))
  46         elif type(_class) == int:
  47             ranges.append((_class, _class))
  48         else:
  49             (min, max) = _class;
  50             if type(min) == str:
  51                 min = ord(min)
  52             if type(max) == str:
  53                 max = ord(max)
  54             if max > 0x7f and min <= 0x7f:
  55                 ranges.append((min, 0x7f))
  56                 min = 0x80
  57             ranges.append((min,max))
  58     ranges.sort();
  59
  60     if emitTables and classes["UseTable"] and (not "Inverse" in classes):
  61         array = ("static const char _%sData[65536] = {\n" % name);
  62         i = 0
  63         for (min,max) in ranges:
  64             while i < min:
  65                 i = i + 1
  66                 array += ('0,')
  67                 if (i % entriesPerLine == 0) and (i != 0):
  68                     array += ('\n')
  69             while i <= max:
  70                 i = i + 1
  71                 if (i == 65536):
  72                     array += ("1")
  73                 else:
  74                     array += ('1,')
  75                 if (i % entriesPerLine == 0) and (i != 0):
  76                     array += ('\n')
  77         while i < 0xffff:
  78             array += ("0,")
  79             i = i + 1;
  80             if (i % entriesPerLine == 0) and (i != 0):
  81                 array += ('\n')
  82         if i == 0xffff:
  83             array += ("0")
  84         array += ("\n};\n\n");
  85         arrays += array
  86
  87     # Generate createFunction:
  88     function = "";
  89     function += ("CharacterClass* %sCreate()\n" % name)
  90     function += ("{\n")
  91     if emitTables and classes["UseTable"]:
  92         if "Inverse" in classes:
  93             function += ("    CharacterClass* characterClass = new CharacterClass(CharacterClassTable::create(_%sData, true));\n" % (classes["Inverse"]))
  94         else:
  95             function += ("    CharacterClass* characterClass = new CharacterClass(CharacterClassTable::create(_%sData, false));\n" % (name))
  96     else:
  97         function += ("    CharacterClass* characterClass = new CharacterClass(0);\n")
  98     for (min, max) in ranges:
  99         if (min == max):
 100             if (min > 127):
 101                 function += ("    characterClass->m_matchesUnicode.append(0x%04x);\n" % min)
 102             else:
 103                 function += ("    characterClass->m_matches.append(0x%02x);\n" % min)
 104             continue
 105         if (min > 127) or (max > 127):
 106             function += ("    characterClass->m_rangesUnicode.append(CharacterRange(0x%04x, 0x%04x));\n" % (min, max))
 107         else:
 108             function += ("    characterClass->m_ranges.append(CharacterRange(0x%02x, 0x%02x));\n" % (min, max))
 109     function += ("    return characterClass;\n")
 110     function += ("}\n\n")
 111     functions += function
 112
 113 if (len(sys.argv) > 1):
 114     f = open(sys.argv[-1], "w")
 115     f.write(arrays)
 116     f.write(functions)
 117     f.close()
 118 else:
 119     print(arrays)
 120     print(functions)
 121