third_party/brotli/dec/transform.h

   1 /* Copyright 2013 Google Inc. All Rights Reserved.
   2
   3    Licensed under the Apache License, Version 2.0 (the "License");
   4    you may not use this file except in compliance with the License.
   5    You may obtain a copy of the License at
   6
   7    http://www.apache.org/licenses/LICENSE-2.0
   8
   9    Unless required by applicable law or agreed to in writing, software
  10    distributed under the License is distributed on an "AS IS" BASIS,
  11    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  12    See the License for the specific language governing permissions and
  13    limitations under the License.
  14
  15    Transformations on dictionary words.
  16 */
  17
  18 #ifndef BROTLI_DEC_TRANSFORM_H_
  19 #define BROTLI_DEC_TRANSFORM_H_
  20
  21 #include <stdio.h>
  22 #include <ctype.h>
  23 #include "./types.h"
  24
  25 #if defined(__cplusplus) || defined(c_plusplus)
  26 extern "C" {
  27 #endif
  28
  29 enum WordTransformType {
  30   kIdentity       = 0,
  31   kOmitLast1      = 1,
  32   kOmitLast2      = 2,
  33   kOmitLast3      = 3,
  34   kOmitLast4      = 4,
  35   kOmitLast5      = 5,
  36   kOmitLast6      = 6,
  37   kOmitLast7      = 7,
  38   kOmitLast8      = 8,
  39   kOmitLast9      = 9,
  40   kUppercaseFirst = 10,
  41   kUppercaseAll   = 11,
  42   kOmitFirst1     = 12,
  43   kOmitFirst2     = 13,
  44   kOmitFirst3     = 14,
  45   kOmitFirst4     = 15,
  46   kOmitFirst5     = 16,
  47   kOmitFirst6     = 17,
  48   kOmitFirst7     = 18,
  49   kOmitFirst8     = 19,
  50   kOmitFirst9     = 20
  51 };
  52
  53 typedef struct {
  54   const char* prefix;
  55   enum WordTransformType transform;
  56   const char* suffix;
  57 } Transform;
  58
  59 static const Transform kTransforms[] = {
  60      {         "", kIdentity,       ""           },
  61      {         "", kIdentity,       " "          },
  62      {        " ", kIdentity,       " "          },
  63      {         "", kOmitFirst1,     ""           },
  64      {         "", kUppercaseFirst, " "          },
  65      {         "", kIdentity,       " the "      },
  66      {        " ", kIdentity,       ""           },
  67      {       "s ", kIdentity,       " "          },
  68      {         "", kIdentity,       " of "       },
  69      {         "", kUppercaseFirst, ""           },
  70      {         "", kIdentity,       " and "      },
  71      {         "", kOmitFirst2,     ""           },
  72      {         "", kOmitLast1,      ""           },
  73      {       ", ", kIdentity,       " "          },
  74      {         "", kIdentity,       ", "         },
  75      {        " ", kUppercaseFirst, " "          },
  76      {         "", kIdentity,       " in "       },
  77      {         "", kIdentity,       " to "       },
  78      {       "e ", kIdentity,       " "          },
  79      {         "", kIdentity,       "\""         },
  80      {         "", kIdentity,       "."          },
  81      {         "", kIdentity,       "\">"        },
  82      {         "", kIdentity,       "\n"         },
  83      {         "", kOmitLast3,      ""           },
  84      {         "", kIdentity,       "]"          },
  85      {         "", kIdentity,       " for "      },
  86      {         "", kOmitFirst3,     ""           },
  87      {         "", kOmitLast2,      ""           },
  88      {         "", kIdentity,       " a "        },
  89      {         "", kIdentity,       " that "     },
  90      {        " ", kUppercaseFirst, ""           },
  91      {         "", kIdentity,       ". "         },
  92      {        ".", kIdentity,       ""           },
  93      {        " ", kIdentity,       ", "         },
  94      {         "", kOmitFirst4,     ""           },
  95      {         "", kIdentity,       " with "     },
  96      {         "", kIdentity,       "'"          },
  97      {         "", kIdentity,       " from "     },
  98      {         "", kIdentity,       " by "       },
  99      {         "", kOmitFirst5,     ""           },
 100      {         "", kOmitFirst6,     ""           },
 101      {    " the ", kIdentity,       ""           },
 102      {         "", kOmitLast4,      ""           },
 103      {         "", kIdentity,       ". The "     },
 104      {         "", kUppercaseAll,   ""           },
 105      {         "", kIdentity,       " on "       },
 106      {         "", kIdentity,       " as "       },
 107      {         "", kIdentity,       " is "       },
 108      {         "", kOmitLast7,      ""           },
 109      {         "", kOmitLast1,      "ing "       },
 110      {         "", kIdentity,       "\n\t"       },
 111      {         "", kIdentity,       ":"          },
 112      {        " ", kIdentity,       ". "         },
 113      {         "", kIdentity,       "ed "        },
 114      {         "", kOmitFirst9,     ""           },
 115      {         "", kOmitFirst7,     ""           },
 116      {         "", kOmitLast6,      ""           },
 117      {         "", kIdentity,       "("          },
 118      {         "", kUppercaseFirst, ", "         },
 119      {         "", kOmitLast8,      ""           },
 120      {         "", kIdentity,       " at "       },
 121      {         "", kIdentity,       "ly "        },
 122      {    " the ", kIdentity,       " of "       },
 123      {         "", kOmitLast5,      ""           },
 124      {         "", kOmitLast9,      ""           },
 125      {        " ", kUppercaseFirst, ", "         },
 126      {         "", kUppercaseFirst, "\""         },
 127      {        ".", kIdentity,       "("          },
 128      {         "", kUppercaseAll,   " "          },
 129      {         "", kUppercaseFirst, "\">"        },
 130      {         "", kIdentity,       "=\""        },
 131      {        " ", kIdentity,       "."          },
 132      {    ".com/", kIdentity,       ""           },
 133      {    " the ", kIdentity,       " of the "   },
 134      {         "", kUppercaseFirst, "'"          },
 135      {         "", kIdentity,       ". This "    },
 136      {         "", kIdentity,       ","          },
 137      {        ".", kIdentity,       " "          },
 138      {         "", kUppercaseFirst, "("          },
 139      {         "", kUppercaseFirst, "."          },
 140      {         "", kIdentity,       " not "      },
 141      {        " ", kIdentity,       "=\""        },
 142      {         "", kIdentity,       "er "        },
 143      {        " ", kUppercaseAll,   " "          },
 144      {         "", kIdentity,       "al "        },
 145      {        " ", kUppercaseAll,   ""           },
 146      {         "", kIdentity,       "='"         },
 147      {         "", kUppercaseAll,   "\""         },
 148      {         "", kUppercaseFirst, ". "         },
 149      {        " ", kIdentity,       "("          },
 150      {         "", kIdentity,       "ful "       },
 151      {        " ", kUppercaseFirst, ". "         },
 152      {         "", kIdentity,       "ive "       },
 153      {         "", kIdentity,       "less "      },
 154      {         "", kUppercaseAll,   "'"          },
 155      {         "", kIdentity,       "est "       },
 156      {        " ", kUppercaseFirst, "."          },
 157      {         "", kUppercaseAll,   "\">"        },
 158      {        " ", kIdentity,       "='"         },
 159      {         "", kUppercaseFirst, ","          },
 160      {         "", kIdentity,       "ize "       },
 161      {         "", kUppercaseAll,   "."          },
 162      { "\xc2\xa0", kIdentity,       ""           },
 163      {        " ", kIdentity,       ","          },
 164      {         "", kUppercaseFirst, "=\""        },
 165      {         "", kUppercaseAll,   "=\""        },
 166      {         "", kIdentity,       "ous "       },
 167      {         "", kUppercaseAll,   ", "         },
 168      {         "", kUppercaseFirst, "='"         },
 169      {        " ", kUppercaseFirst, ","          },
 170      {        " ", kUppercaseAll,   "=\""        },
 171      {        " ", kUppercaseAll,   ", "         },
 172      {         "", kUppercaseAll,   ","          },
 173      {         "", kUppercaseAll,   "("          },
 174      {         "", kUppercaseAll,   ". "         },
 175      {        " ", kUppercaseAll,   "."          },
 176      {         "", kUppercaseAll,   "='"         },
 177      {        " ", kUppercaseAll,   ". "         },
 178      {        " ", kUppercaseFirst, "=\""        },
 179      {        " ", kUppercaseAll,   "='"         },
 180      {        " ", kUppercaseFirst, "='"         },
 181 };
 182
 183 static const int kNumTransforms = sizeof(kTransforms) / sizeof(kTransforms[0]);
 184
 185 static int ToUpperCase(uint8_t *p) {
 186   if (p[0] < 0xc0) {
 187     if (p[0] >= 'a' && p[0] <= 'z') {
 188       p[0] ^= 32;
 189     }
 190     return 1;
 191   }
 192   /* An overly simplified uppercasing model for utf-8. */
 193   if (p[0] < 0xe0) {
 194     p[1] ^= 32;
 195     return 2;
 196   }
 197   /* An arbitrary transform for three byte characters. */
 198   p[2] ^= 5;
 199   return 3;
 200 }
 201
 202 static BROTLI_INLINE int TransformDictionaryWord(
 203     uint8_t* dst, const uint8_t* word, int len, int transform) {
 204   const char* prefix = kTransforms[transform].prefix;
 205   const char* suffix = kTransforms[transform].suffix;
 206   const int t = kTransforms[transform].transform;
 207   int skip = t < kOmitFirst1 ? 0 : t - (kOmitFirst1 - 1);
 208   int idx = 0;
 209   int i = 0;
 210   uint8_t* uppercase;
 211   if (skip > len) {
 212     skip = len;
 213   }
 214   while (*prefix) { dst[idx++] = (uint8_t)*prefix++; }
 215   word += skip;
 216   len -= skip;
 217   if (t <= kOmitLast9) {
 218     len -= t;
 219   }
 220   while (i < len) { dst[idx++] = word[i++]; }
 221   uppercase = &dst[idx - len];
 222   if (t == kUppercaseFirst) {
 223     ToUpperCase(uppercase);
 224   } else if (t == kUppercaseAll) {
 225     while (len > 0) {
 226       int step = ToUpperCase(uppercase);
 227       uppercase += step;
 228       len -= step;
 229     }
 230   }
 231   while (*suffix) { dst[idx++] = (uint8_t)*suffix++; }
 232   return idx;
 233 }
 234
 235 #if defined(__cplusplus) || defined(c_plusplus)
 236 }    /* extern "C" */
 237 #endif
 238
 239 #endif  /* BROTLI_DEC_TRANSFORM_H_ */