| Brian Silverman | 70325d6 | 2015-09-20 17:00:43 -0400 | [diff] [blame] | 1 | # Copyright (c) 2008, Google Inc. |
| 2 | # All rights reserved. |
| 3 | # |
| 4 | # Redistribution and use in source and binary forms, with or without |
| 5 | # modification, are permitted provided that the following conditions are |
| 6 | # met: |
| 7 | # |
| 8 | # * Redistributions of source code must retain the above copyright |
| 9 | # notice, this list of conditions and the following disclaimer. |
| 10 | # * Redistributions in binary form must reproduce the above |
| 11 | # copyright notice, this list of conditions and the following disclaimer |
| 12 | # in the documentation and/or other materials provided with the |
| 13 | # distribution. |
| 14 | # * Neither the name of Google Inc. nor the names of its |
| 15 | # contributors may be used to endorse or promote products derived from |
| 16 | # this software without specific prior written permission. |
| 17 | # |
| 18 | # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
| 19 | # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
| 20 | # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR |
| 21 | # A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT |
| 22 | # OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, |
| 23 | # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
| 24 | # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
| 25 | # DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
| 26 | # THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 27 | # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 28 | # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 29 | # --- |
| 30 | # |
| 31 | # Author: falmeida@google.com (Filipe Almeida) |
| 32 | |
| 33 | # TODO(falmeida): Add more descriptive names to the states and drop the |
| 34 | # abbreviations. |
| 35 | # TODO(falmeida): Reorder the states so that it's easier to read. |
| 36 | # TODO(falmeida): Support CDATA blocks in the form: <![CDATA[. |
| 37 | |
| 38 | name = 'htmlparser' |
| 39 | |
| 40 | comment = 'Definition of a finite state machine for a subset of HTTP 4.1' |
| 41 | |
| 42 | condition('<', '<') |
| 43 | condition('>', '>') |
| 44 | condition('=', '=') |
| 45 | |
| 46 | # TODO(falmeida): This is not the correct expression. tag and attribute names |
| 47 | # can only consist of alpha character. |
| 48 | condition('id', 'A-Za-z0-9_:.-') |
| 49 | condition('idtag', 'A-Za-z0-9/_:.-') |
| 50 | |
| 51 | # Whitespace according to: http://www.w3.org/TR/html401/struct/text.html#h-9.1 |
| 52 | condition('space', ' \t\n\r') |
| 53 | condition('!', '!') |
| 54 | condition('q', '\'') |
| 55 | condition('dq', '\"') |
| 56 | condition('/', '/') |
| 57 | condition('*', '*') |
| 58 | condition('-', '-') |
| 59 | condition('?', '?') |
| 60 | condition('lf', '\n') |
| 61 | condition('quote', '\\') |
| 62 | |
| 63 | # TODO(falmeida): This default rule is a hack and shouldn't be here. |
| 64 | condition('default', '[:default:]') |
| 65 | |
| 66 | state(name = 'text', |
| 67 | external = 'text', |
| 68 | transitions = [ |
| 69 | ['<', 'tag_start'], |
| 70 | ['default', 'text'] |
| 71 | ]) |
| 72 | |
| 73 | # When we found the < character in text. |
| 74 | # Tag opening is defined in the HTML5 draft here: |
| 75 | # http://www.whatwg.org/specs/web-apps/current-work/#tag-open-state |
| 76 | # We don't exactly follow this and are much more loose in order to mimic the way |
| 77 | # the major browsers behave. |
| 78 | state(name = 'tag_start', |
| 79 | external = 'tag', |
| 80 | transitions = [ |
| 81 | ['idtag', 'tag_name'], |
| 82 | ['?', 'pi'], |
| 83 | ['!', 'declaration_start'], |
| 84 | ['<', 'tag_start'], |
| 85 | ['default', 'text'] |
| 86 | ]) |
| 87 | |
| 88 | # Name of the tag. Includes the closing tag character '/'. |
| 89 | state(name = 'tag_name', |
| 90 | external = 'tag', |
| 91 | transitions = [ |
| 92 | ['idtag', 'tag_name'], |
| 93 | ['space', 'tag_space'], |
| 94 | ['>', 'tag_close'] |
| 95 | ]) |
| 96 | |
| 97 | # HTML declaration and comment parsing |
| 98 | # |
| 99 | # We don't expose declaration state because at this point we only want to |
| 100 | # ensure that we are parsing them correctly so we don't get out of sync. |
| 101 | # This is specifically made for DOCTYPE declarations and won't work if DTD's |
| 102 | # are defined inside the declaration. |
| 103 | # The HTML5 spec says we should specificly look for the string '<!DOCTYPE HTML' |
| 104 | # but that will add a lot of unecessary states, and unless we build a simple |
| 105 | # declarative way to unfold a string match into multiple states, I don't |
| 106 | # think it's worth worrying about for now. |
| 107 | |
| 108 | # Got '<!'. The next character will decide if we open a declaration or a |
| 109 | # comment. |
| 110 | state(name = 'declaration_start', |
| 111 | external = 'text', |
| 112 | transitions = [ |
| 113 | ['-', 'comment_open'], |
| 114 | ['>', 'text'], |
| 115 | ['default', 'declaration_body'] |
| 116 | ]) |
| 117 | |
| 118 | # Inside a declaration. Ie: <!DOCTYPE. We close when we see a '>' |
| 119 | state(name = 'declaration_body', |
| 120 | external = 'text', |
| 121 | transitions = [ |
| 122 | ['>', 'text'], |
| 123 | ['default', 'declaration_body'] |
| 124 | ]) |
| 125 | |
| 126 | # Got '<!-'. |
| 127 | state(name = 'comment_open', |
| 128 | external = 'text', |
| 129 | transitions = [ |
| 130 | ['-', 'comment_body'], |
| 131 | ['default', 'text'] |
| 132 | ]) |
| 133 | |
| 134 | # Inside a comment. We only close when we see '-->' |
| 135 | state(name = 'comment_body', |
| 136 | external = 'comment', |
| 137 | transitions = [ |
| 138 | ['-', 'comment_dash'], |
| 139 | ['default', 'comment_body'] |
| 140 | ]) |
| 141 | |
| 142 | # Got '-' inside a comment. |
| 143 | state(name = 'comment_dash', |
| 144 | external = 'comment', |
| 145 | transitions = [ |
| 146 | ['-', 'comment_dash_dash'], |
| 147 | ['default', 'comment_body'] |
| 148 | ]) |
| 149 | |
| 150 | # Got '--' inside a comment. |
| 151 | state(name = 'comment_dash_dash', |
| 152 | external = 'comment', |
| 153 | transitions = [ |
| 154 | ['-', 'comment_dash_dash'], |
| 155 | ['>', 'text'], |
| 156 | ['default', 'comment_body'] |
| 157 | ]) |
| 158 | |
| 159 | # XML Processing instruction parsing according to: |
| 160 | # http://www.w3.org/TR/REC-xml/#sec-pi |
| 161 | # |
| 162 | # Everything between the characters <? and ?> is considered to be part of the |
| 163 | # processing instruction. |
| 164 | state(name = 'pi', |
| 165 | external = 'text', |
| 166 | transitions = [ |
| 167 | ['?', 'pi_may_end'], |
| 168 | ['default', 'pi'] |
| 169 | ]) |
| 170 | |
| 171 | state(name = 'pi_may_end', |
| 172 | external = 'text', |
| 173 | transitions = [ |
| 174 | ['>', 'text'], |
| 175 | ['default', 'pi'] |
| 176 | ]) |
| 177 | |
| 178 | # Whitespace between tag name, attributes. |
| 179 | state(name = 'tag_space', |
| 180 | external = 'tag', |
| 181 | transitions = [ |
| 182 | ['>', 'tag_close'], |
| 183 | ['space', 'tag_space'], |
| 184 | ['id', 'attr'], |
| 185 | ['/', 'tag_space'] |
| 186 | ]) |
| 187 | |
| 188 | state(name = 'tag_close', |
| 189 | external = 'text', |
| 190 | transitions = [ |
| 191 | ['<', 'tag_start'], |
| 192 | ['default', 'text'] |
| 193 | ]) |
| 194 | |
| 195 | # Name of the attribute. |
| 196 | state(name = 'attr', |
| 197 | external = 'attr', |
| 198 | transitions = [ |
| 199 | ['id', 'attr'], |
| 200 | ['>', 'tag_close'], |
| 201 | ['/', 'tag_space'], |
| 202 | ['=', 'value'], |
| 203 | ['space', 'attr_space'] |
| 204 | ]) |
| 205 | |
| 206 | # After the attribute name. |
| 207 | state(name = 'attr_space', |
| 208 | external = 'attr', |
| 209 | transitions = [ |
| 210 | ['>', 'tag_close'], |
| 211 | ['space', 'attr_space'], |
| 212 | ['id', 'attr'], |
| 213 | ['/', 'tag_space'], |
| 214 | ['=', 'value'] |
| 215 | ]) |
| 216 | |
| 217 | # Expecting a value, after attribute= |
| 218 | state(name = 'value', |
| 219 | external = 'value', |
| 220 | transitions = [ |
| 221 | ['q', 'value_q_start'], |
| 222 | ['dq', 'value_dq_start'], |
| 223 | ['space', 'value'], |
| 224 | ['>', 'tag_close'], |
| 225 | ['default', 'value_text'] |
| 226 | ]) |
| 227 | |
| 228 | # Unquoted attribute value. |
| 229 | state(name = 'value_text', |
| 230 | external = 'value', |
| 231 | transitions = [ |
| 232 | ['>', 'tag_close'], |
| 233 | ['space', 'tag_space'], |
| 234 | ['default', 'value_text'] |
| 235 | ]) |
| 236 | |
| 237 | # First character of a single quoted attribute value. |
| 238 | state(name = 'value_q_start', |
| 239 | external = 'value', |
| 240 | transitions = [ |
| 241 | ['q', 'tag_space'], |
| 242 | ['default', 'value_q'] |
| 243 | ]) |
| 244 | |
| 245 | # In the middle of a single quoted attribute value. |
| 246 | state(name = 'value_q', |
| 247 | external = 'value', |
| 248 | transitions = [ |
| 249 | ['q', 'tag_space'], |
| 250 | ['default', 'value_q'] |
| 251 | ]) |
| 252 | |
| 253 | # First character of a double quoted attribute value. |
| 254 | state(name = 'value_dq_start', |
| 255 | external = 'value', |
| 256 | transitions = [ |
| 257 | ['dq', 'tag_space'], |
| 258 | ['default', 'value_dq'] |
| 259 | ]) |
| 260 | |
| 261 | # In the middle of a double quoted attribute value. |
| 262 | state(name = 'value_dq', |
| 263 | external = 'value', |
| 264 | transitions = [ |
| 265 | ['dq', 'tag_space'], |
| 266 | ['default', 'value_dq'] |
| 267 | ]) |
| 268 | |
| 269 | # CDATA escaping text spans. |
| 270 | # TODO(falmeida): These states should go after cdata_text. |
| 271 | |
| 272 | # Got '<!' |
| 273 | state(name = 'cdata_comment_start', |
| 274 | external = 'text', |
| 275 | transitions = [ |
| 276 | ['-', 'cdata_comment_start_dash'], |
| 277 | ['default', 'cdata_text'], |
| 278 | ]) |
| 279 | |
| 280 | # Got '<!-'. |
| 281 | state(name = 'cdata_comment_start_dash', |
| 282 | external = 'text', |
| 283 | transitions = [ |
| 284 | ['-', 'cdata_comment_body'], |
| 285 | ['default', 'cdata_text'] |
| 286 | ]) |
| 287 | |
| 288 | # Inside a comment |
| 289 | state(name = 'cdata_comment_body', |
| 290 | external = 'text', |
| 291 | transitions = [ |
| 292 | ['-', 'cdata_comment_dash'], |
| 293 | ['default', 'cdata_comment_body'] |
| 294 | ]) |
| 295 | |
| 296 | # Got '-' inside a comment. |
| 297 | state(name = 'cdata_comment_dash', |
| 298 | external = 'text', |
| 299 | transitions = [ |
| 300 | ['-', 'cdata_comment_dash_dash'], |
| 301 | ['default', 'cdata_comment_body'] |
| 302 | ]) |
| 303 | |
| 304 | # Got '--' inside a comment. |
| 305 | state(name = 'cdata_comment_dash_dash', |
| 306 | external = 'text', |
| 307 | transitions = [ |
| 308 | ['-', 'cdata_comment_dash_dash'], |
| 309 | ['>', 'cdata_text'], |
| 310 | ['default', 'cdata_comment_body'] |
| 311 | ]) |
| 312 | |
| 313 | # CDATA processing |
| 314 | # |
| 315 | # To simplify the code, we treat RCDATA and CDATA sections the same since the |
| 316 | # differences between them don't affect the context we are in. |
| 317 | state(name = 'cdata_text', |
| 318 | external = 'text', |
| 319 | transitions = [ |
| 320 | ['<', 'cdata_lt'], |
| 321 | ['default', 'cdata_text'] |
| 322 | ]) |
| 323 | |
| 324 | # Possible beginning of the closing tag. |
| 325 | state(name = 'cdata_lt', |
| 326 | external = 'text', |
| 327 | transitions = [ |
| 328 | ['/', 'cdata_may_close'], |
| 329 | ['!', 'cdata_comment_start'], |
| 330 | ['default', 'cdata_text'] |
| 331 | ]) |
| 332 | |
| 333 | # If we encounter </tag where tag matches the last opened tag, we exit the |
| 334 | # CDATA section. Part of this logic is handled in the code. |
| 335 | state(name = 'cdata_may_close', |
| 336 | external = 'text', |
| 337 | transitions = [ |
| 338 | ['idtag', 'cdata_may_close'], |
| 339 | ['>', 'text'], |
| 340 | ['space', 'tag_space'], |
| 341 | ['default', 'cdata_text'] |
| 342 | ]) |
| 343 | |
| 344 | # The next states are used for specialized parser modes. |
| 345 | state(name = 'js_file', |
| 346 | external = 'js_file', |
| 347 | transitions = [ |
| 348 | ['default', 'js_file'] |
| 349 | ]) |
| 350 | |
| 351 | # TODO(falmeida): Having css_file and js_file as the external name doesn't make |
| 352 | # sense. This should instead be text and the js/css state be |
| 353 | # returned by # in_js() and in_css(). |
| 354 | state(name = 'css_file', |
| 355 | external = 'css_file', |
| 356 | transitions = [ |
| 357 | ['default', 'css_file'] |
| 358 | ]) |
| 359 | |
| 360 | state(name = 'null', |
| 361 | external = 'text', |
| 362 | transitions = [ |
| 363 | ['default', 'null'] |
| 364 | ]) |
| 365 | |