| 1 |
/* |
| 2 |
* Copyright (c) 2024 joshua stein <jcs@jcs.org> |
| 3 |
* |
| 4 |
* Permission to use, copy, modify, and distribute this software for any |
| 5 |
* purpose with or without fee is hereby granted, provided that the above |
| 6 |
* copyright notice and this permission notice appear in all copies. |
| 7 |
* |
| 8 |
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES |
| 9 |
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF |
| 10 |
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR |
| 11 |
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES |
| 12 |
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
| 13 |
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
| 14 |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
| 15 |
*/ |
| 16 |
|
| 17 |
#include "html.h" |
| 18 |
|
| 19 |
#ifdef HTML_ENABLE |
| 20 |
|
| 21 |
#ifdef HTML_ENABLE_DEBUGGING |
| 22 |
const char *html_mode_names[] = { |
| 23 |
"NONE", |
| 24 |
"INITIAL", |
| 25 |
"BEFORE_HTML", |
| 26 |
"BEFORE_HEAD", |
| 27 |
"IN_HEAD", |
| 28 |
"IN_HEAD_NOSCRIPT", |
| 29 |
"AFTER_HEAD", |
| 30 |
"IN_BODY", |
| 31 |
"TEXT", |
| 32 |
"IN_TABLE", |
| 33 |
"IN_TABLE_TEXT", |
| 34 |
"IN_CAPTION", |
| 35 |
"IN_COLUMN_GROUP", |
| 36 |
"IN_TABLE_BODY", |
| 37 |
"IN_ROW", |
| 38 |
"IN_CELL", |
| 39 |
"IN_SELECT", |
| 40 |
"IN_SELECT_IN_TABLE", |
| 41 |
"IN_TEMPLATE", |
| 42 |
"AFTER_BODY", |
| 43 |
"IN_FRAMESET", |
| 44 |
"AFTER_FRAMESET", |
| 45 |
"AFTER_AFTER_BODY", |
| 46 |
"AFTER_AFTER_FRAMESET" |
| 47 |
}; |
| 48 |
|
| 49 |
const char *html_state_names[] = { |
| 50 |
"NONE", |
| 51 |
"DATA", |
| 52 |
"RCDATA", |
| 53 |
"RAWTEXT", |
| 54 |
"SCRIPT_DATA", |
| 55 |
"PLAINTEXT", |
| 56 |
"TAG_OPEN", |
| 57 |
"END_TAG_OPEN", |
| 58 |
"TAG_NAME", |
| 59 |
"RCDATA_LESS_THAN_SIGN", |
| 60 |
"RCDATA_END_TAG_OPEN", |
| 61 |
"RCDATA_END_TAG_NAME", |
| 62 |
"RAWTEXT_LESS_THAN_SIGN", |
| 63 |
"RAWTEXT_END_TAG_OPEN", |
| 64 |
"RAWTEXT_END_TAG_NAME", |
| 65 |
"SCRIPT_DATA_LESS_THAN_SIGN", |
| 66 |
"SCRIPT_DATA_END_TAG_OPEN", |
| 67 |
"SCRIPT_DATA_END_TAG_NAME", |
| 68 |
"SCRIPT_DATA_ESCAPE_START", |
| 69 |
"SCRIPT_DATA_ESCAPE_START_DASH", |
| 70 |
"SCRIPT_DATA_ESCAPED", |
| 71 |
"SCRIPT_DATA_ESCAPED_DASH", |
| 72 |
"SCRIPT_DATA_ESCAPED_DASH_DASH", |
| 73 |
"SCRIPT_DATA_ESCAPED_LESS_THAN_SIGN", |
| 74 |
"SCRIPT_DATA_ESCAPED_END_TAG_OPEN", |
| 75 |
"SCRIPT_DATA_ESCAPED_END_TAG_NAME", |
| 76 |
"SCRIPT_DATA_DOUBLE_ESCAPE_START", |
| 77 |
"SCRIPT_DATA_DOUBLE_ESCAPED", |
| 78 |
"SCRIPT_DATA_DOUBLE_ESCAPED_DASH", |
| 79 |
"SCRIPT_DATA_DOUBLE_ESCAPED_DASH_DASH", |
| 80 |
"SCRIPT_DATA_DOUBLE_ESCAPED_LESS_THAN_SIGN", |
| 81 |
"SCRIPT_DATA_DOUBLE_ESCAPE_END", |
| 82 |
"BEFORE_ATTRIBUTE_NAME", |
| 83 |
"ATTRIBUTE_NAME", |
| 84 |
"AFTER_ATTRIBUTE_NAME", |
| 85 |
"BEFORE_ATTRIBUTE_VALUE", |
| 86 |
"ATTRIBUTE_VALUE_DOUBLE_QUOTED", |
| 87 |
"ATTRIBUTE_VALUE_SINGLE_QUOTED", |
| 88 |
"ATTRIBUTE_VALUE_UNQUOTED", |
| 89 |
"AFTER_ATTRIBUTE_VALUE_QUOTED", |
| 90 |
"SELF_CLOSING_START_TAG", |
| 91 |
"BOGUS_COMMENT", |
| 92 |
"MARKUP_DECLARATION_OPEN", |
| 93 |
"COMMENT_START", |
| 94 |
"COMMENT_START_DASH", |
| 95 |
"COMMENT", |
| 96 |
"COMMENT_LESS_THAN_SIGN", |
| 97 |
"COMMENT_LESS_THAN_SIGN_BANG", |
| 98 |
"COMMENT_LESS_THAN_SIGN_BANG_DASH", |
| 99 |
"COMMENT_LESS_THAN_SIGN_BANG_DASH_DASH", |
| 100 |
"COMMENT_END_DASH", |
| 101 |
"COMMENT_END", |
| 102 |
"COMMENT_END_BANG", |
| 103 |
"DOCTYPE", |
| 104 |
"BEFORE_DOCTYPE_NAME", |
| 105 |
"DOCTYPE_NAME", |
| 106 |
"AFTER_DOCTYPE_NAME", |
| 107 |
"AFTER_DOCTYPE_PUBLIC_KEYWORD", |
| 108 |
"BEFORE_DOCTYPE_PUBLIC_IDENTIFIER", |
| 109 |
"DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED", |
| 110 |
"DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED", |
| 111 |
"AFTER_DOCTYPE_PUBLIC_IDENTIFIER", |
| 112 |
"BETWEEN_DOCTYPE_PUBLIC_AND_SYSTEM_IDENTIFIERS", |
| 113 |
"AFTER_DOCTYPE_SYSTEM_KEYWORD", |
| 114 |
"BEFORE_DOCTYPE_SYSTEM_IDENTIFIER", |
| 115 |
"DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED", |
| 116 |
"DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED", |
| 117 |
"AFTER_DOCTYPE_SYSTEM_IDENTIFIER", |
| 118 |
"BOGUS_DOCTYPE", |
| 119 |
"CDATA_SECTION", |
| 120 |
"CDATA_SECTION_BRACKET", |
| 121 |
"CDATA_SECTION_END", |
| 122 |
"CHARACTER_REFERENCE", |
| 123 |
"NAMED_CHARACTER_REFERENCE", |
| 124 |
"AMBIGUOUS_AMPERSAND", |
| 125 |
"NUMERIC_CHARACTER_REFERENCE", |
| 126 |
"HEXADECIMAL_CHARACTER_REFERENCE_START", |
| 127 |
"DECIMAL_CHARACTER_REFERENCE_START", |
| 128 |
"HEXADECIMAL_CHARACTER_REFERENCE", |
| 129 |
"DECIMAL_CHARACTER_REFERENCE", |
| 130 |
"NUMERIC_CHARACTER_REFERENCE_END" |
| 131 |
}; |
| 132 |
|
| 133 |
const char *html_error_strings[] = { |
| 134 |
"NONE", |
| 135 |
"ABRUPT_CLOSING_OF_EMPTY_COMMENT", |
| 136 |
"ABRUPT_DOCTYPE_PUBLIC_IDENTIFIER", |
| 137 |
"ABRUPT_DOCTYPE_SYSTEM_IDENTIFIER", |
| 138 |
"ABSENCE_OF_DIGITS_IN_NUMERIC_CHARACTER_REFERENCE", |
| 139 |
"CDATA_IN_HTML_CONTENT", |
| 140 |
"CHARACTER_REFERENCE_OUTSIDE_UNICODE_RANGE", |
| 141 |
"CONTROL_CHARACTER_IN_INPUT_STREAM", |
| 142 |
"CONTROL_CHARACTER_REFERENCE", |
| 143 |
"DUPLICATE_ATTRIBUTE", |
| 144 |
"END_TAG_WITH_ATTRIBUTES", |
| 145 |
"END_TAG_WITH_TRAILING_SOLIDUS", |
| 146 |
"EOF_BEFORE_TAG_NAME", |
| 147 |
"EOF_IN_CDATA", |
| 148 |
"EOF_IN_COMMENT", |
| 149 |
"EOF_IN_DOCTYPE", |
| 150 |
"EOF_IN_SCRIPT_HTML_COMMENT_LIKE_TEXT", |
| 151 |
"EOF_IN_TAG", |
| 152 |
"INCORRECTLY_CLOSED_COMMENT", |
| 153 |
"INCORRECTLY_OPENED_COMMENT", |
| 154 |
"INVALID_CHARACTER_SEQUENCE_AFTER_DOCTYPE_NAME", |
| 155 |
"INVALID_FIRST_CHARACTER_OF_TAG_NAME", |
| 156 |
"MISSING_ATTRIBUTE_VALUE", |
| 157 |
"MISSING_DOCTYPE_NAME", |
| 158 |
"MISSING_DOCTYPE_PUBLIC_IDENTIFIER", |
| 159 |
"MISSING_DOCTYPE_SYSTEM_IDENTIFIER", |
| 160 |
"MISSING_END_TAG_NAME", |
| 161 |
"MISSING_QUOTE_BEFORE_DOCTYPE_PUBLIC_IDENTIFIER", |
| 162 |
"MISSING_QUOTE_BEFORE_DOCTYPE_SYSTEM_IDENTIFIER", |
| 163 |
"MISSING_SEMICOLON_AFTER_CHARACTER_REFERENCE", |
| 164 |
"MISSING_WHITESPACE_AFTER_DOCTYPE_PUBLIC_KEYWORD", |
| 165 |
"MISSING_WHITESPACE_AFTER_DOCTYPE_SYSTEM_KEYWORD", |
| 166 |
"MISSING_WHITESPACE_BEFORE_DOCTYPE_NAME", |
| 167 |
"MISSING_WHITESPACE_BETWEEN_ATTRIBUTES", |
| 168 |
"MISSING_WHITESPACE_BETWEEN_DOCTYPE_PUBLIC_AND_SYSTEM_IDENTIFIERS", |
| 169 |
"NESTED_COMMENT", |
| 170 |
"NONCHARACTER_CHARACTER_REFERENCE", |
| 171 |
"NONCHARACTER_IN_INPUT_STREAM", |
| 172 |
"NON_VOID_HTML_ELEMENT_START_TAG_WITH_TRAILING_SOLIDUS", |
| 173 |
"NULL_CHARACTER_REFERENCE", |
| 174 |
"SURROGATE_CHARACTER_REFERENCE", |
| 175 |
"SURROGATE_IN_INPUT_STREAM", |
| 176 |
"UNEXPECTED_CHARACTER_AFTER_DOCTYPE_SYSTEM_IDENTIFIER", |
| 177 |
"UNEXPECTED_CHARACTER_IN_ATTRIBUTE_NAME", |
| 178 |
"UNEXPECTED_CHARACTER_IN_UNQUOTED_ATTRIBUTE_VALUE", |
| 179 |
"UNEXPECTED_EQUALS_SIGN_BEFORE_ATTRIBUTE_NAME", |
| 180 |
"UNEXPECTED_NULL_CHARACTER", |
| 181 |
"UNEXPECTED_QUESTION_MARK_INSTEAD_OF_TAG_NAME", |
| 182 |
"UNEXPECTED_SOLIDUS_IN_TAG", |
| 183 |
"UNKNOWN_NAMED_CHARACTER_REFERENCE" |
| 184 |
}; |
| 185 |
|
| 186 |
const char *html_token_names[] = { |
| 187 |
"INVALID", |
| 188 |
|
| 189 |
"DOCTYPE", |
| 190 |
"START_TAG", |
| 191 |
"END_TAG", |
| 192 |
"COMMENT", |
| 193 |
"CHARACTER", |
| 194 |
"EOF" |
| 195 |
}; |
| 196 |
#endif |
| 197 |
|
| 198 |
/* this doesn't have to list all tags, just ones the docs reference */ |
| 199 |
const char *html_tag_names[] = { |
| 200 |
"\1", /* bogus just so 0 isn't a valid tag id */ |
| 201 |
"a", |
| 202 |
"address", |
| 203 |
"applet", |
| 204 |
"area", |
| 205 |
"article", |
| 206 |
"aside", |
| 207 |
"b", |
| 208 |
"base", |
| 209 |
"basefont", |
| 210 |
"bgsound", |
| 211 |
"big", |
| 212 |
"blockquote", |
| 213 |
"body", |
| 214 |
"br", |
| 215 |
"button", |
| 216 |
"caption", |
| 217 |
"center", |
| 218 |
"cite", |
| 219 |
"code", |
| 220 |
"col", |
| 221 |
"colgroup", |
| 222 |
"dd", |
| 223 |
"details", |
| 224 |
"dfn", |
| 225 |
"dialog", |
| 226 |
"dir", |
| 227 |
"div", |
| 228 |
"dl", |
| 229 |
"dt", |
| 230 |
"em", |
| 231 |
"embed", |
| 232 |
"fieldset", |
| 233 |
"figcaption", |
| 234 |
"figure", |
| 235 |
"font", |
| 236 |
"footer", |
| 237 |
"form", |
| 238 |
"frame", |
| 239 |
"frameset", |
| 240 |
"h1", |
| 241 |
"h2", |
| 242 |
"h3", |
| 243 |
"h4", |
| 244 |
"h5", |
| 245 |
"h6", |
| 246 |
"head", |
| 247 |
"header", |
| 248 |
"hgroup", |
| 249 |
"hr", |
| 250 |
"html", |
| 251 |
"i", |
| 252 |
"iframe", |
| 253 |
"image", |
| 254 |
"img", |
| 255 |
"input", |
| 256 |
"ins", |
| 257 |
"kbd", |
| 258 |
"keygen", |
| 259 |
"li", |
| 260 |
"link", |
| 261 |
"listing", |
| 262 |
"main", |
| 263 |
"marquee", |
| 264 |
"math", |
| 265 |
"menu", |
| 266 |
"meta", |
| 267 |
"nav", |
| 268 |
"nobr", |
| 269 |
"noembed", |
| 270 |
"noframes", |
| 271 |
"noscript", |
| 272 |
"object", |
| 273 |
"ol", |
| 274 |
"optgroup", |
| 275 |
"option", |
| 276 |
"p", |
| 277 |
"param", |
| 278 |
"plaintext", |
| 279 |
"pre", |
| 280 |
"rb", |
| 281 |
"rp", |
| 282 |
"rt", |
| 283 |
"rtc", |
| 284 |
"ruby", |
| 285 |
"s", |
| 286 |
"samp", |
| 287 |
"script", |
| 288 |
"search", |
| 289 |
"section", |
| 290 |
"select", |
| 291 |
"small", |
| 292 |
"source", |
| 293 |
"span", |
| 294 |
"strike", |
| 295 |
"strong", |
| 296 |
"style", |
| 297 |
"sub", |
| 298 |
"sup", |
| 299 |
"summary", |
| 300 |
"svg", |
| 301 |
"table", |
| 302 |
"tbody", |
| 303 |
"td", |
| 304 |
"template", |
| 305 |
"textarea", |
| 306 |
"tfoot", |
| 307 |
"th", |
| 308 |
"thead", |
| 309 |
"title", |
| 310 |
"tr", |
| 311 |
"track", |
| 312 |
"tt", |
| 313 |
"u", |
| 314 |
"ul", |
| 315 |
"var", |
| 316 |
"wbr", |
| 317 |
"xmp", |
| 318 |
|
| 319 |
NULL |
| 320 |
}; |
| 321 |
|
| 322 |
#endif /* HTML_ENABLE */ |