| 1 |
package Whatpm::CSS::Tokenizer; |
| 2 |
use strict; |
| 3 |
|
| 4 |
sub BEFORE_TOKEN_STATE () { 0 } |
| 5 |
sub BEFORE_NMSTART_STATE () { 1 } |
| 6 |
sub NAME_STATE () { 2 } |
| 7 |
sub ESCAPE_OPEN_STATE () { 3 } |
| 8 |
sub STRING_STATE () { 4 } |
| 9 |
sub HASH_OPEN_STATE () { 5 } |
| 10 |
sub NUMBER_STATE () { 6 } |
| 11 |
sub NUMBER_FRACTION_STATE () { 7 } |
| 12 |
sub AFTER_NUMBER_STATE () { 8 } |
| 13 |
sub URI_BEFORE_WSP_STATE () { 9 } |
| 14 |
sub ESCAPE_STATE () { 10 } |
| 15 |
sub ESCAPE_BEFORE_LF_STATE () { 11 } |
| 16 |
sub ESCAPE_BEFORE_NL_STATE () { 12 } |
| 17 |
sub NUMBER_DOT_STATE () { 13 } |
| 18 |
sub NUMBER_DOT_NUMBER_STATE () { 14 } |
| 19 |
sub DELIM_STATE () { 15 } |
| 20 |
sub URI_UNQUOTED_STATE () { 16 } |
| 21 |
sub URI_AFTER_WSP_STATE () { 17 } |
| 22 |
sub AFTER_AT_STATE () { 18 } |
| 23 |
sub AFTER_AT_HYPHEN_STATE () { 19 } |
| 24 |
|
| 25 |
sub IDENT_TOKEN () { 1 } |
| 26 |
sub ATKEYWORD_TOKEN () { 2 } |
| 27 |
sub HASH_TOKEN () { 3 } |
| 28 |
sub FUNCTION_TOKEN () { 4 } |
| 29 |
sub URI_TOKEN () { 5 } |
| 30 |
sub URI_INVALID_TOKEN () { 6 } |
| 31 |
sub URI_PREFIX_TOKEN () { 7 } |
| 32 |
sub URI_PREFIX_INVALID_TOKEN () { 8 } |
| 33 |
sub STRING_TOKEN () { 9 } |
| 34 |
sub INVALID_TOKEN () { 10 } |
| 35 |
sub NUMBER_TOKEN () { 11 } |
| 36 |
sub DIMENSION_TOKEN () { 12 } |
| 37 |
sub PERCENTAGE_TOKEN () { 13 } |
| 38 |
sub UNICODE_RANGE_TOKEN () { 14 } |
| 39 |
sub DELIM_TOKEN () { 16 } |
| 40 |
sub PLUS_TOKEN () { 17 } |
| 41 |
sub GREATER_TOKEN () { 18 } |
| 42 |
sub COMMA_TOKEN () { 19 } |
| 43 |
sub TILDE_TOKEN () { 20 } |
| 44 |
sub DASHMATCH_TOKEN () { 21 } |
| 45 |
sub PREFIXMATCH_TOKEN () { 22 } |
| 46 |
sub SUFFIXMATCH_TOKEN () { 23 } |
| 47 |
sub SUBSTRINGMATCH_TOKEN () { 24 } |
| 48 |
sub INCLUDES_TOKEN () { 25 } |
| 49 |
sub SEMICOLON_TOKEN () { 26 } |
| 50 |
sub LBRACE_TOKEN () { 27 } |
| 51 |
sub RBRACE_TOKEN () { 28 } |
| 52 |
sub LPAREN_TOKEN () { 29 } |
| 53 |
sub RPAREN_TOKEN () { 30 } |
| 54 |
sub LBRACKET_TOKEN () { 31 } |
| 55 |
sub RBRACKET_TOKEN () { 32 } |
| 56 |
sub S_TOKEN () { 33 } |
| 57 |
sub CDO_TOKEN () { 34 } |
| 58 |
sub CDC_TOKEN () { 35 } |
| 59 |
sub COMMENT_TOKEN () { 36 } |
| 60 |
sub COMMENT_INVALID_TOKEN () { 37 } |
| 61 |
sub EOF_TOKEN () { 38 } |
| 62 |
|
| 63 |
our @TokenName = qw( |
| 64 |
0 IDENT ATKEYWORD HASH FUNCTION URI URI_INVALID URI_PREFIX URI_PREFIX_INVALID |
| 65 |
STRING INVALID NUMBER DIMENSION PERCENTAGE UNICODE_RANGE |
| 66 |
0 DELIM PLUS GREATER COMMA TILDE DASHMATCH |
| 67 |
PREFIXMATCH SUFFIXMATCH SUBSTRINGMATCH INCLUDES SEMICOLON |
| 68 |
LBRACE RBRACE LPAREN RPAREN LBRACKET RBRACKET S CDO CDC COMMENT |
| 69 |
COMMENT_INVALID EOF |
| 70 |
); |
| 71 |
|
| 72 |
sub new ($) { |
| 73 |
my $self = bless {token => [], get_char => sub { -1 }, |
| 74 |
onerror => sub { }}, shift; |
| 75 |
return $self; |
| 76 |
} # new |
| 77 |
|
| 78 |
sub init ($) { |
| 79 |
my $self = shift; |
| 80 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 81 |
$self->{c} = $self->{get_char}->(); |
| 82 |
#$self->{t} = {type => token-type, value => value, number => number}; |
| 83 |
} # init |
| 84 |
|
| 85 |
sub get_next_token ($) { |
| 86 |
my $self = shift; |
| 87 |
if (@{$self->{token}}) { |
| 88 |
return shift @{$self->{token}}; |
| 89 |
} |
| 90 |
|
| 91 |
my $char; |
| 92 |
my $num; # |{num}|, if any. |
| 93 |
my $i; # |$i + 1|th character in |unicode| in |escape|. |
| 94 |
my $q; |
| 95 |
## NOTE: |
| 96 |
## 0: in |ident|. |
| 97 |
## 1: in |URI| outside of |string|. |
| 98 |
## 0x0022: in |string1| or |invalid1|. |
| 99 |
## 0x0027: in |string2| or |invalid2|. |
| 100 |
|
| 101 |
A: { |
| 102 |
if ($self->{state} == BEFORE_TOKEN_STATE) { |
| 103 |
if ($self->{c} == 0x002D) { # - |
| 104 |
## NOTE: |-| in |ident| in |IDENT| |
| 105 |
$self->{t} = {type => IDENT_TOKEN, value => '-'}; |
| 106 |
$self->{state} = BEFORE_NMSTART_STATE; |
| 107 |
$self->{c} = $self->{get_char}->(); |
| 108 |
redo A; |
| 109 |
} elsif ($self->{c} == 0x0055 or $self->{c} == 0x0075) { # U or u |
| 110 |
$self->{t} = {type => IDENT_TOKEN, value => chr $self->{c}}; |
| 111 |
$self->{c} = $self->{get_char}->(); |
| 112 |
if ($self->{c} == 0x002B) { # + |
| 113 |
$self->{c} = $self->{get_char}->(); |
| 114 |
if ((0x0030 <= $self->{c} and $self->{c} <= 0x0039) or # 0..9 |
| 115 |
(0x0041 <= $self->{c} and $self->{c} <= 0x0046) or # A..F |
| 116 |
(0x0061 <= $self->{c} and $self->{c} <= 0x0066) or # a..f |
| 117 |
$self->{c} == 0x003F) { # ? |
| 118 |
$self->{t}->{value} .= '+' . chr $self->{c}; |
| 119 |
$self->{t}->{type} = UNICODE_RANGE_TOKEN; |
| 120 |
$self->{c} = $self->{get_char}->(); |
| 121 |
C: for (2..6) { |
| 122 |
if ((0x0030 <= $self->{c} and $self->{c} <= 0x0039) or # 0..9 |
| 123 |
(0x0041 <= $self->{c} and $self->{c} <= 0x0046) or # A..F |
| 124 |
(0x0061 <= $self->{c} and $self->{c} <= 0x0066) or # a..f |
| 125 |
$self->{c} == 0x003F) { # ? |
| 126 |
$self->{t}->{value} .= chr $self->{c}; |
| 127 |
$self->{c} = $self->{get_char}->(); |
| 128 |
} else { |
| 129 |
last C; |
| 130 |
} |
| 131 |
} # C |
| 132 |
|
| 133 |
if ($self->{c} == 0x002D) { # - |
| 134 |
$self->{c} = $self->{get_char}->(); |
| 135 |
if ((0x0030 <= $self->{c} and $self->{c} <= 0x0039) or # 0..9 |
| 136 |
(0x0041 <= $self->{c} and $self->{c} <= 0x0046) or # A..F |
| 137 |
(0x0061 <= $self->{c} and $self->{c} <= 0x0066)) { # a..f |
| 138 |
$self->{t}->{value} .= '-' . chr $self->{c}; |
| 139 |
$self->{c} = $self->{get_char}->(); |
| 140 |
C: for (2..6) { |
| 141 |
if ((0x0030 <= $self->{c} and $self->{c} <= 0x0039) or # 0..9 |
| 142 |
(0x0041 <= $self->{c} and $self->{c} <= 0x0046) or # A..F |
| 143 |
(0x0061 <= $self->{c} and $self->{c} <= 0x0066)) { # a..f |
| 144 |
$self->{t}->{value} .= chr $self->{c}; |
| 145 |
$self->{c} = $self->{get_char}->(); |
| 146 |
} else { |
| 147 |
last C; |
| 148 |
} |
| 149 |
} # C |
| 150 |
|
| 151 |
# |
| 152 |
} else { |
| 153 |
my $token = $self->{t}; |
| 154 |
$self->{t} = {type => IDENT_TOKEN, value => '-'}; |
| 155 |
$self->{state} = BEFORE_NMSTART_STATE; |
| 156 |
# reprocess |
| 157 |
return $token; |
| 158 |
#redo A; |
| 159 |
} |
| 160 |
} |
| 161 |
|
| 162 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 163 |
# reprocess |
| 164 |
return $self->{t}; |
| 165 |
#redo A; |
| 166 |
} else { |
| 167 |
unshift @{$self->{token}}, {type => PLUS_TOKEN}; |
| 168 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 169 |
# reprocess |
| 170 |
return $self->{t}; |
| 171 |
#redo A; |
| 172 |
} |
| 173 |
} else { |
| 174 |
$self->{state} = NAME_STATE; |
| 175 |
# reprocess |
| 176 |
redo A; |
| 177 |
} |
| 178 |
} elsif ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 179 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 180 |
$self->{c} == 0x005F or # _ |
| 181 |
$self->{c} > 0x007F) { # nonascii |
| 182 |
## NOTE: |nmstart| in |ident| in |IDENT| |
| 183 |
$self->{t} = {type => IDENT_TOKEN, value => chr $self->{c}}; |
| 184 |
$self->{state} = NAME_STATE; |
| 185 |
$self->{c} = $self->{get_char}->(); |
| 186 |
redo A; |
| 187 |
} elsif ($self->{c} == 0x005C) { # \ |
| 188 |
## NOTE: |nmstart| in |ident| in |IDENT| |
| 189 |
$self->{t} = {type => IDENT_TOKEN, value => ''}; |
| 190 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 191 |
$self->{c} = $self->{get_char}->(); |
| 192 |
redo A; |
| 193 |
} elsif ($self->{c} == 0x0040) { # @ |
| 194 |
## NOTE: |@| in |ATKEYWORD| |
| 195 |
$self->{t} = {type => ATKEYWORD_TOKEN, value => ''}; |
| 196 |
$self->{state} = AFTER_AT_STATE; |
| 197 |
$self->{c} = $self->{get_char}->(); |
| 198 |
redo A; |
| 199 |
} elsif ($self->{c} == 0x0022 or $self->{c} == 0x0027) { # " or ' |
| 200 |
$self->{t} = {type => STRING_TOKEN, value => ''}; |
| 201 |
$self->{state} = STRING_STATE; $q = $self->{c}; |
| 202 |
$self->{c} = $self->{get_char}->(); |
| 203 |
redo A; |
| 204 |
} elsif ($self->{c} == 0x0023) { # # |
| 205 |
## NOTE: |#| in |HASH|. |
| 206 |
$self->{t} = {type => HASH_TOKEN, value => ''}; |
| 207 |
$self->{state} = HASH_OPEN_STATE; |
| 208 |
$self->{c} = $self->{get_char}->(); |
| 209 |
redo A; |
| 210 |
} elsif (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { # 0..9 |
| 211 |
## NOTE: |num|. |
| 212 |
$self->{t} = {type => NUMBER_TOKEN, value => chr $self->{c}}; |
| 213 |
$self->{state} = NUMBER_STATE; |
| 214 |
$self->{c} = $self->{get_char}->(); |
| 215 |
redo A; |
| 216 |
} elsif ($self->{c} == 0x002E) { # . |
| 217 |
## NOTE: |num|. |
| 218 |
$self->{t} = {type => NUMBER_TOKEN, value => '0'}; |
| 219 |
$self->{state} = NUMBER_FRACTION_STATE; |
| 220 |
$self->{c} = $self->{get_char}->(); |
| 221 |
redo A; |
| 222 |
} elsif ($self->{c} == 0x002F) { # / |
| 223 |
$self->{c} = $self->{get_char}->(); |
| 224 |
if ($self->{c} == 0x002A) { # * |
| 225 |
C: { |
| 226 |
$self->{c} = $self->{get_char}->(); |
| 227 |
if ($self->{c} == 0x002A) { # * |
| 228 |
D: { |
| 229 |
$self->{c} = $self->{get_char}->(); |
| 230 |
if ($self->{c} == 0x002F) { # / |
| 231 |
# |
| 232 |
} elsif ($self->{c} == 0x002A) { # * |
| 233 |
redo D; |
| 234 |
} else { |
| 235 |
redo C; |
| 236 |
} |
| 237 |
} # D |
| 238 |
} elsif ($self->{c} == -1) { |
| 239 |
# stay in the state |
| 240 |
# reprocess |
| 241 |
return {type => COMMENT_INVALID_TOKEN}; |
| 242 |
#redo A; |
| 243 |
} else { |
| 244 |
redo C; |
| 245 |
} |
| 246 |
} # C |
| 247 |
|
| 248 |
# stay in the state. |
| 249 |
$self->{c} = $self->{get_char}->(); |
| 250 |
redo A; |
| 251 |
} else { |
| 252 |
# stay in the state. |
| 253 |
# reprocess |
| 254 |
return {type => DELIM_STATE, value => '/'}; |
| 255 |
#redo A; |
| 256 |
} |
| 257 |
} elsif ($self->{c} == 0x003C) { # < |
| 258 |
## NOTE: |CDO| |
| 259 |
$self->{c} = $self->{get_char}->(); |
| 260 |
if ($self->{c} == 0x0021) { # ! |
| 261 |
$self->{c} = $self->{get_char}->(); |
| 262 |
if ($self->{c} == 0x002C) { # - |
| 263 |
$self->{c} = $self->{get_char}->(); |
| 264 |
if ($self->{c} == 0x002C) { # - |
| 265 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 266 |
$self->{c} = $self->{get_char}->(); |
| 267 |
return {type => CDO_TOKEN}; |
| 268 |
#redo A; |
| 269 |
} else { |
| 270 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '!'}; |
| 271 |
## NOTE: |-| in |ident| in |IDENT| |
| 272 |
$self->{t} = {type => IDENT_TOKEN, value => '-'}; |
| 273 |
$self->{state} = BEFORE_NMSTART_STATE; |
| 274 |
#reprocess |
| 275 |
return {type => DELIM_TOKEN, value => '<'}; |
| 276 |
#redo A; |
| 277 |
} |
| 278 |
} else { |
| 279 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '!'}; |
| 280 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 281 |
#reprocess |
| 282 |
return {type => DELIM_TOKEN, value => '<'}; |
| 283 |
#redo A; |
| 284 |
} |
| 285 |
} else { |
| 286 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 287 |
#reprocess |
| 288 |
return {type => DELIM_TOKEN, value => '<'}; |
| 289 |
#redo A; |
| 290 |
} |
| 291 |
} elsif (my $t = { |
| 292 |
0x003B => SEMICOLON_TOKEN, # ; |
| 293 |
0x007B => LBRACE_TOKEN, # { |
| 294 |
0x007D => RBRACE_TOKEN, # } |
| 295 |
0x0028 => LPAREN_TOKEN, # ( |
| 296 |
0x0029 => RPAREN_TOKEN, # ) |
| 297 |
0x005B => LBRACKET_TOKEN, # [ |
| 298 |
0x005D => RBRACKET_TOKEN, # ] |
| 299 |
}->{$self->{c}}) { |
| 300 |
# stay in the state |
| 301 |
$self->{c} = $self->{get_char}->(); |
| 302 |
return {type => $t}; |
| 303 |
# redo A; |
| 304 |
} elsif ({ |
| 305 |
0x0020 => 1, # SP |
| 306 |
0x0009 => 1, # \t |
| 307 |
0x000D => 1, # \r |
| 308 |
0x000A => 1, # \n |
| 309 |
0x000C => 1, # \f |
| 310 |
}->{$self->{c}}) { |
| 311 |
W: { |
| 312 |
$self->{c} = $self->{get_char}->(); |
| 313 |
if ({ |
| 314 |
0x0020 => 1, # SP |
| 315 |
0x0009 => 1, # \t |
| 316 |
0x000D => 1, # \r |
| 317 |
0x000A => 1, # \n |
| 318 |
0x000C => 1, # \f |
| 319 |
}->{$self->{c}}) { |
| 320 |
redo W; |
| 321 |
} elsif (my $v = { |
| 322 |
0x002B => PLUS_TOKEN, # + |
| 323 |
0x003E => GREATER_TOKEN, # > |
| 324 |
0x002C => COMMA_TOKEN, # , |
| 325 |
0x007E => TILDE_TOKEN, # ~ |
| 326 |
}->{$self->{c}}) { |
| 327 |
# stay in the state |
| 328 |
$self->{c} = $self->{get_char}->(); |
| 329 |
return {type => $v}; |
| 330 |
#redo A; |
| 331 |
} else { |
| 332 |
# stay in the state |
| 333 |
# reprocess |
| 334 |
return {type => S_TOKEN}; |
| 335 |
#redo A; |
| 336 |
} |
| 337 |
} # W |
| 338 |
} elsif (my $v = { |
| 339 |
0x007C => DASHMATCH_TOKEN, # | |
| 340 |
0x005E => PREFIXMATCH_TOKEN, # ^ |
| 341 |
0x0024 => SUFFIXMATCH_TOKEN, # $ |
| 342 |
0x002A => SUBSTRINGMATCH_TOKEN, # * |
| 343 |
}->{$self->{c}}) { |
| 344 |
my $c = $self->{c}; |
| 345 |
$self->{c} = $self->{get_char}->(); |
| 346 |
if ($self->{c} == 0x003D) { # = |
| 347 |
# stay in the state |
| 348 |
$self->{c} = $self->{get_char}->(); |
| 349 |
return {type => $v}; |
| 350 |
#redo A; |
| 351 |
} else { |
| 352 |
# stay in the state |
| 353 |
# reprocess |
| 354 |
return {type => DELIM_TOKEN, value => chr $c}; |
| 355 |
#redo A; |
| 356 |
} |
| 357 |
} elsif ($self->{c} == 0x002B) { # + |
| 358 |
# stay in the state |
| 359 |
$self->{c} = $self->{get_char}->(); |
| 360 |
return {type => PLUS_TOKEN}; |
| 361 |
#redo A; |
| 362 |
} elsif ($self->{c} == 0x003E) { # > |
| 363 |
# stay in the state |
| 364 |
$self->{c} = $self->{get_char}->(); |
| 365 |
return {type => GREATER_TOKEN}; |
| 366 |
#redo A; |
| 367 |
} elsif ($self->{c} == 0x002C) { # , |
| 368 |
# stay in the state |
| 369 |
$self->{c} = $self->{get_char}->(); |
| 370 |
return {type => COMMA_TOKEN}; |
| 371 |
#redo A; |
| 372 |
} elsif ($self->{c} == 0x007E) { # ~ |
| 373 |
$self->{c} = $self->{get_char}->(); |
| 374 |
if ($self->{c} == 0x003D) { # = |
| 375 |
# stay in the state |
| 376 |
$self->{c} = $self->{get_char}->(); |
| 377 |
return {type => INCLUDES_TOKEN}; |
| 378 |
#redo A; |
| 379 |
} else { |
| 380 |
# stay in the state |
| 381 |
# reprocess |
| 382 |
return {type => TILDE_TOKEN}; |
| 383 |
#redo A; |
| 384 |
} |
| 385 |
} elsif ($self->{c} == -1) { |
| 386 |
# stay in the state |
| 387 |
$self->{c} = $self->{get_char}->(); |
| 388 |
return {type => EOF_TOKEN}; |
| 389 |
#redo A; |
| 390 |
} else { |
| 391 |
# stay in the state |
| 392 |
$self->{t} = {type => DELIM_TOKEN, value => chr $self->{c}}; |
| 393 |
$self->{c} = $self->{get_char}->(); |
| 394 |
return $self->{t}; |
| 395 |
#redo A; |
| 396 |
} |
| 397 |
} elsif ($self->{state} == BEFORE_NMSTART_STATE) { |
| 398 |
## NOTE: |nmstart| in |ident| in (|IDENT|, |DIMENSION|, or |
| 399 |
## |FUNCTION|) |
| 400 |
if ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 401 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 402 |
$self->{c} == 0x005F or # _ |
| 403 |
$self->{c} > 0x007F) { # nonascii |
| 404 |
$self->{t}->{value} .= chr $self->{c}; |
| 405 |
$self->{t}->{type} = DIMENSION_TOKEN |
| 406 |
if $self->{t}->{type} == NUMBER_TOKEN; |
| 407 |
$self->{state} = NAME_STATE; |
| 408 |
$self->{c} = $self->{get_char}->(); |
| 409 |
redo A; |
| 410 |
} elsif ($self->{c} == 0x005C) { # \ |
| 411 |
## TODO: 12-\X, 12-\{nl} |
| 412 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 413 |
$self->{c} = $self->{get_char}->(); |
| 414 |
redo A; |
| 415 |
} elsif ($self->{c} == 0x002D and # - |
| 416 |
$self->{t}->{type} == IDENT_TOKEN) { |
| 417 |
$self->{c} = $self->{get_char}->(); |
| 418 |
if ($self->{c} == 0x003E) { # > |
| 419 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 420 |
$self->{c} = $self->{get_char}->(); |
| 421 |
return {type => CDC_TOKEN}; |
| 422 |
#redo A; |
| 423 |
} else { |
| 424 |
## NOTE: |-|, |-|, $self->{c} |
| 425 |
#$self->{t} = {type => IDENT_TOKEN, value => '-'}; |
| 426 |
# stay in the state |
| 427 |
# reconsume |
| 428 |
return {type => DELIM_TOKEN, value => '-'}; |
| 429 |
#redo A; |
| 430 |
} |
| 431 |
} else { |
| 432 |
if ($self->{t}->{type} == NUMBER_TOKEN) { |
| 433 |
## NOTE: |-| after |NUMBER|. |
| 434 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '-'}; |
| 435 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 436 |
# reconsume |
| 437 |
$self->{t}->{value} = $self->{t}->{number}; |
| 438 |
delete $self->{t}->{number}; |
| 439 |
return $self->{t}; |
| 440 |
} else { |
| 441 |
## NOTE: |-| not followed by |nmstart|. |
| 442 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 443 |
$self->{c} = $self->{get_char}->(); |
| 444 |
return {type => DELIM_TOKEN, value => '-'}; |
| 445 |
} |
| 446 |
} |
| 447 |
} elsif ($self->{state} == AFTER_AT_STATE) { |
| 448 |
if ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 449 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 450 |
$self->{c} == 0x005F or # _ |
| 451 |
$self->{c} > 0x007F) { # nonascii |
| 452 |
$self->{t}->{value} .= chr $self->{c}; |
| 453 |
$self->{state} = NAME_STATE; |
| 454 |
$self->{c} = $self->{get_char}->(); |
| 455 |
redo A; |
| 456 |
} elsif ($self->{c} == 0x002D) { # - |
| 457 |
$self->{t}->{value} .= '-'; |
| 458 |
$self->{state} = AFTER_AT_HYPHEN_STATE; |
| 459 |
$self->{c} = $self->{get_char}->(); |
| 460 |
redo A; |
| 461 |
} elsif ($self->{c} == 0x005C) { # \ |
| 462 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 463 |
$self->{c} = $self->{get_char}->(); |
| 464 |
redo A; |
| 465 |
} else { |
| 466 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 467 |
# reprocess |
| 468 |
return {type => DELIM_TOKEN, value => '@'}; |
| 469 |
} |
| 470 |
} elsif ($self->{state} == AFTER_AT_HYPHEN_STATE) { |
| 471 |
if ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 472 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 473 |
$self->{c} == 0x005F or # _ |
| 474 |
$self->{c} > 0x007F) { # nonascii |
| 475 |
$self->{t}->{value} .= chr $self->{c}; |
| 476 |
$self->{state} = NAME_STATE; |
| 477 |
$self->{c} = $self->{get_char}->(); |
| 478 |
redo A; |
| 479 |
} elsif ($self->{c} == 0x002D) { # - |
| 480 |
$self->{c} = $self->{get_char}->(); |
| 481 |
if ($self->{c} == 0x003E) { # > |
| 482 |
unshift @{$self->{token}}, {type => CDC_TOKEN}; |
| 483 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 484 |
$self->{c} = $self->{get_char}->(); |
| 485 |
return {type => DELIM_TOKEN, value => '@'}; |
| 486 |
#redo A; |
| 487 |
} else { |
| 488 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '-'}; |
| 489 |
$self->{t} = {type => IDENT_TOKEN, value => '-'}; |
| 490 |
$self->{state} = BEFORE_NMSTART_STATE; |
| 491 |
# reprocess |
| 492 |
return {type => DELIM_TOKEN, value => '@'}; |
| 493 |
#redo A; |
| 494 |
} |
| 495 |
} elsif ($self->{c} == 0x005C) { # \ |
| 496 |
## TODO: @-\{nl} |
| 497 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 498 |
$self->{c} = $self->{get_char}->(); |
| 499 |
redo A; |
| 500 |
} else { |
| 501 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '-'}; |
| 502 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 503 |
# reprocess |
| 504 |
return {type => DELIM_TOKEN, value => '@'}; |
| 505 |
} |
| 506 |
} elsif ($self->{state} == AFTER_NUMBER_STATE) { |
| 507 |
if ($self->{c} == 0x002D) { # - |
| 508 |
## NOTE: |-| in |ident|. |
| 509 |
$self->{t}->{value} = '-'; |
| 510 |
$self->{state} = BEFORE_NMSTART_STATE; |
| 511 |
$self->{c} = $self->{get_char}->(); |
| 512 |
redo A; |
| 513 |
} elsif ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 514 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 515 |
$self->{c} == 0x005F or # _ |
| 516 |
$self->{c} > 0x007F) { # nonascii |
| 517 |
## NOTE: |nmstart| in |ident|. |
| 518 |
$self->{t}->{value} = chr $self->{c}; |
| 519 |
$self->{t}->{type} = DIMENSION_TOKEN; |
| 520 |
$self->{state} = NAME_STATE; |
| 521 |
$self->{c} = $self->{get_char}->(); |
| 522 |
redo A; |
| 523 |
} elsif ($self->{c} == 0x005C) { # \ |
| 524 |
## NOTE: |nmstart| in |ident| in |IDENT| |
| 525 |
$self->{t}->{value} = ''; |
| 526 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 527 |
$self->{c} = $self->{get_char}->(); |
| 528 |
redo A; |
| 529 |
} elsif ($self->{c} == 0x0025) { # % |
| 530 |
$self->{t}->{type} = PERCENTAGE_TOKEN; |
| 531 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 532 |
$self->{c} = $self->{get_char}->(); |
| 533 |
return $self->{t}; |
| 534 |
#redo A; |
| 535 |
} else { |
| 536 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 537 |
# reprocess |
| 538 |
return $self->{t}; |
| 539 |
#redo A; |
| 540 |
} |
| 541 |
} elsif ($self->{state} == HASH_OPEN_STATE) { |
| 542 |
## NOTE: The first |nmchar| in |name| in |HASH|. |
| 543 |
if ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 544 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 545 |
(0x0030 <= $self->{c} and $self->{c} <= 0x0039) or # 0..9 |
| 546 |
$self->{c} == 0x002D or # - |
| 547 |
$self->{c} == 0x005F or # _ |
| 548 |
$self->{c} > 0x007F) { # nonascii |
| 549 |
$self->{t}->{value} .= chr $self->{c}; |
| 550 |
$self->{state} = NAME_STATE; |
| 551 |
$self->{c} = $self->{get_char}->(); |
| 552 |
redo A; |
| 553 |
} elsif ($self->{c} == 0x005C) { # \ |
| 554 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 555 |
$self->{c} = $self->{get_char}->(); |
| 556 |
redo A; |
| 557 |
} else { |
| 558 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 559 |
$self->{c} = $self->{get_char}->(); |
| 560 |
return {type => DELIM_TOKEN, value => '#'}; |
| 561 |
#redo A; |
| 562 |
} |
| 563 |
} elsif ($self->{state} == NAME_STATE) { |
| 564 |
## NOTE: |nmchar| in (|ident| or |name|). |
| 565 |
if ((0x0041 <= $self->{c} and $self->{c} <= 0x005A) or # A..Z |
| 566 |
(0x0061 <= $self->{c} and $self->{c} <= 0x007A) or # a..z |
| 567 |
(0x0030 <= $self->{c} and $self->{c} <= 0x0039) or # 0..9 |
| 568 |
$self->{c} == 0x005F or # _ |
| 569 |
$self->{c} == 0x002D or # - |
| 570 |
$self->{c} > 0x007F) { # nonascii |
| 571 |
$self->{t}->{value} .= chr $self->{c}; |
| 572 |
# stay in the state |
| 573 |
$self->{c} = $self->{get_char}->(); |
| 574 |
redo A; |
| 575 |
} elsif ($self->{c} == 0x005C) { # \ |
| 576 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 0; |
| 577 |
$self->{c} = $self->{get_char}->(); |
| 578 |
redo A; |
| 579 |
} elsif ($self->{c} == 0x0028 and # ( |
| 580 |
$self->{t}->{type} == IDENT_TOKEN) { # ( |
| 581 |
my $func_name = $self->{t}->{value}; |
| 582 |
$func_name =~ tr/A-Z/a-z/; ## TODO: Unicode or ASCII case-insensitive? |
| 583 |
if ($func_name eq 'url' or $func_name eq 'url-prefix') { |
| 584 |
if ($self->{t}->{has_escape}) { |
| 585 |
## TODO: warn |
| 586 |
} |
| 587 |
$self->{t}->{type} |
| 588 |
= $func_name eq 'url' ? URI_TOKEN : URI_PREFIX_TOKEN; |
| 589 |
$self->{t}->{value} = ''; |
| 590 |
$self->{state} = URI_BEFORE_WSP_STATE; |
| 591 |
$self->{c} = $self->{get_char}->(); |
| 592 |
redo A; |
| 593 |
} else { |
| 594 |
$self->{t}->{type} = FUNCTION_TOKEN; |
| 595 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 596 |
$self->{c} = $self->{get_char}->(); |
| 597 |
return $self->{t}; |
| 598 |
#redo A; |
| 599 |
} |
| 600 |
} else { |
| 601 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 602 |
# reconsume |
| 603 |
return $self->{t}; |
| 604 |
#redo A; |
| 605 |
} |
| 606 |
} elsif ($self->{state} == URI_BEFORE_WSP_STATE) { |
| 607 |
while ({ |
| 608 |
0x0020 => 1, # SP |
| 609 |
0x0009 => 1, # \t |
| 610 |
0x000D => 1, # \r |
| 611 |
0x000A => 1, # \n |
| 612 |
0x000C => 1, # \f |
| 613 |
}->{$self->{c}}) { |
| 614 |
$self->{c} = $self->{get_char}->(); |
| 615 |
} |
| 616 |
if ($self->{c} == -1) { |
| 617 |
$self->{t}->{type} = { |
| 618 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 619 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 620 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 621 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 622 |
}->{$self->{t}->{type}}; |
| 623 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 624 |
$self->{c} = $self->{get_char}->(); |
| 625 |
return $self->{t}; |
| 626 |
#redo A; |
| 627 |
} elsif ($self->{c} < 0x0020 or $self->{c} == 0x0028) { # C0 or ( |
| 628 |
## TODO: Should we consider matches of "(" and ")"? |
| 629 |
$self->{t}->{type} = { |
| 630 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 631 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 632 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 633 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 634 |
}->{$self->{t}->{type}}; |
| 635 |
$self->{state} = URI_UNQUOTED_STATE; |
| 636 |
$self->{c} = $self->{get_char}->(); |
| 637 |
redo A; |
| 638 |
} elsif ($self->{c} == 0x0022 or $self->{c} == 0x0027) { # " or ' |
| 639 |
$self->{state} = STRING_STATE; $q = $self->{c}; |
| 640 |
$self->{c} = $self->{get_char}->(); |
| 641 |
redo A; |
| 642 |
} elsif ($self->{c} == 0x0029) { # ) |
| 643 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 644 |
$self->{c} = $self->{get_char}->(); |
| 645 |
return $self->{t}; |
| 646 |
#redo A; |
| 647 |
} elsif ($self->{c} == 0x005C) { # \ |
| 648 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 1; |
| 649 |
$self->{c} = $self->{get_char}->(); |
| 650 |
redo A; |
| 651 |
} else { |
| 652 |
$self->{t}->{value} .= chr $self->{c}; |
| 653 |
$self->{state} = URI_UNQUOTED_STATE; |
| 654 |
$self->{c} = $self->{get_char}->(); |
| 655 |
redo A; |
| 656 |
} |
| 657 |
} elsif ($self->{state} == URI_UNQUOTED_STATE) { |
| 658 |
if ({ |
| 659 |
0x0020 => 1, # SP |
| 660 |
0x0009 => 1, # \t |
| 661 |
0x000D => 1, # \r |
| 662 |
0x000A => 1, # \n |
| 663 |
0x000C => 1, # \f |
| 664 |
}->{$self->{c}}) { |
| 665 |
$self->{state} = URI_AFTER_WSP_STATE; |
| 666 |
$self->{c} = $self->{get_char}->(); |
| 667 |
redo A; |
| 668 |
} elsif ($self->{c} == -1) { |
| 669 |
$self->{t}->{type} = { |
| 670 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 671 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 672 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 673 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 674 |
}->{$self->{t}->{type}}; |
| 675 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 676 |
$self->{c} = $self->{get_char}->(); |
| 677 |
return $self->{t}; |
| 678 |
#redo A; |
| 679 |
} elsif ($self->{c} < 0x0020 or { |
| 680 |
0x0022 => 1, # " |
| 681 |
0x0027 => 1, # ' |
| 682 |
0x0028 => 1, # ( |
| 683 |
}->{$self->{c}}) { # C0 or ( |
| 684 |
## TODO: Should we consider matches of "(" and ")", '"', or "'"? |
| 685 |
$self->{t}->{type} = { |
| 686 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 687 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 688 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 689 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 690 |
}->{$self->{t}->{type}}; |
| 691 |
# stay in the state. |
| 692 |
$self->{c} = $self->{get_char}->(); |
| 693 |
redo A; |
| 694 |
} elsif ($self->{c} == 0x0029) { # ) |
| 695 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 696 |
$self->{c} = $self->{get_char}->(); |
| 697 |
return $self->{t}; |
| 698 |
#redo A; |
| 699 |
} elsif ($self->{c} == 0x005C) { # \ |
| 700 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 1; |
| 701 |
$self->{c} = $self->{get_char}->(); |
| 702 |
redo A; |
| 703 |
} else { |
| 704 |
$self->{t}->{value} .= chr $self->{c}; |
| 705 |
# stay in the state. |
| 706 |
$self->{c} = $self->{get_char}->(); |
| 707 |
redo A; |
| 708 |
} |
| 709 |
} elsif ($self->{state} == URI_AFTER_WSP_STATE) { |
| 710 |
if ({ |
| 711 |
0x0020 => 1, # SP |
| 712 |
0x0009 => 1, # \t |
| 713 |
0x000D => 1, # \r |
| 714 |
0x000A => 1, # \n |
| 715 |
0x000C => 1, # \f |
| 716 |
}->{$self->{c}}) { |
| 717 |
# stay in the state. |
| 718 |
$self->{c} = $self->{get_char}->(); |
| 719 |
redo A; |
| 720 |
} elsif ($self->{c} == -1) { |
| 721 |
$self->{t}->{type} = { |
| 722 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 723 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 724 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 725 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 726 |
}->{$self->{t}->{type}}; |
| 727 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 728 |
$self->{c} = $self->{get_char}->(); |
| 729 |
return $self->{t}; |
| 730 |
#redo A; |
| 731 |
} elsif ($self->{c} == 0x0029) { # ) |
| 732 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 733 |
$self->{c} = $self->{get_char}->(); |
| 734 |
return $self->{t}; |
| 735 |
#redo A; |
| 736 |
} elsif ($self->{c} == 0x005C) { # \ |
| 737 |
$self->{state} = ESCAPE_OPEN_STATE; $q = 1; |
| 738 |
$self->{c} = $self->{get_char}->(); |
| 739 |
redo A; |
| 740 |
} else { |
| 741 |
## TODO: Should we consider matches of "(" and ")", '"', or "'"? |
| 742 |
$self->{t}->{type} = { |
| 743 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 744 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 745 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 746 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 747 |
}->{$self->{t}->{type}}; |
| 748 |
# stay in the state. |
| 749 |
$self->{c} = $self->{get_char}->(); |
| 750 |
redo A; |
| 751 |
} |
| 752 |
} elsif ($self->{state} == ESCAPE_OPEN_STATE) { |
| 753 |
$self->{t}->{has_escape} = 1; |
| 754 |
if (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { # 0..9 |
| 755 |
## NOTE: second character of |unicode| in |escape|. |
| 756 |
$char = $self->{c} - 0x0030; |
| 757 |
$self->{state} = ESCAPE_STATE; $i = 2; |
| 758 |
$self->{c} = $self->{get_char}->(); |
| 759 |
redo A; |
| 760 |
} elsif (0x0041 <= $self->{c} and $self->{c} <= 0x0046) { # A..F |
| 761 |
## NOTE: second character of |unicode| in |escape|. |
| 762 |
$char = $self->{c} - 0x0041 + 0xA; |
| 763 |
$self->{state} = ESCAPE_STATE; $i = 2; |
| 764 |
$self->{c} = $self->{get_char}->(); |
| 765 |
redo A; |
| 766 |
} elsif (0x0061 <= $self->{c} and $self->{c} <= 0x0066) { # a..f |
| 767 |
## NOTE: second character of |unicode| in |escape|. |
| 768 |
$char = $self->{c} - 0x0061 - 0xA; |
| 769 |
$self->{state} = ESCAPE_STATE; $i = 2; |
| 770 |
$self->{c} = $self->{get_char}->(); |
| 771 |
redo A; |
| 772 |
} elsif ($self->{c} == 0x000A or # \n |
| 773 |
$self->{c} == 0x000C) { # \f |
| 774 |
if ($q == 0) { |
| 775 |
## NOTE: In |escape| in ... in |ident|. |
| 776 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 777 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '\\'}; |
| 778 |
return $self->{t}; |
| 779 |
# reconsume |
| 780 |
#redo A; |
| 781 |
} elsif ($q == 1) { |
| 782 |
## NOTE: In |escape| in |URI|. |
| 783 |
$self->{t}->{type} = { |
| 784 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 785 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 786 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 787 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 788 |
}->{$self->{t}->{type}}; |
| 789 |
$self->{t}->{value} .= chr $self->{c}; |
| 790 |
$self->{state} = URI_UNQUOTED_STATE; |
| 791 |
$self->{c} = $self->{get_char}->(); |
| 792 |
redo A; |
| 793 |
} else { |
| 794 |
## Note: In |nl| in ... in |string| or |ident|. |
| 795 |
$self->{t}->{value} .= chr $self->{c}; |
| 796 |
$self->{state} = STRING_STATE; |
| 797 |
$self->{c} = $self->{get_char}->(); |
| 798 |
redo A; |
| 799 |
} |
| 800 |
} elsif ($self->{c} == 0x000D) { # \r |
| 801 |
if ($q == 0) { |
| 802 |
## NOTE: In |escape| in ... in |ident|. |
| 803 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 804 |
unshift @{$self->{token}}, {type => DELIM_TOKEN, value => '\\'}; |
| 805 |
return $self->{t}; |
| 806 |
# reconsume |
| 807 |
#redo A; |
| 808 |
} elsif ($q == 1) { |
| 809 |
$self->{t}->{type} = { |
| 810 |
URI_TOKEN, URI_INVALID_TOKEN, |
| 811 |
URI_INVALID_TOKEN, URI_INVALID_TOKEN, |
| 812 |
URI_PREFIX_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 813 |
URI_PREFIX_INVALID_TOKEN, URI_PREFIX_INVALID_TOKEN, |
| 814 |
}->{$self->{t}->{type}}; |
| 815 |
$self->{t}->{value} .= "\x0D\x0A"; |
| 816 |
$self->{state} = URI_UNQUOTED_STATE; |
| 817 |
$self->{c} = $self->{get_char}->(); |
| 818 |
redo A; |
| 819 |
} else { |
| 820 |
## Note: In |nl| in ... in |string| or |ident|. |
| 821 |
$self->{t}->{value} .= "\x0D\x0A"; |
| 822 |
$self->{state} = ESCAPE_BEFORE_LF_STATE; |
| 823 |
$self->{c} = $self->{get_char}->(); |
| 824 |
redo A; |
| 825 |
} |
| 826 |
} else { |
| 827 |
## NOTE: second character of |escape|. |
| 828 |
$self->{t}->{value} .= chr $self->{c}; |
| 829 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 830 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 831 |
$self->{c} = $self->{get_char}->(); |
| 832 |
redo A; |
| 833 |
} |
| 834 |
} elsif ($self->{state} == ESCAPE_STATE) { |
| 835 |
## NOTE: third..seventh character of |unicode| in |escape|. |
| 836 |
if (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { # 0..9 |
| 837 |
$char = $char * 0x10 + $self->{c} - 0x0030; |
| 838 |
$self->{state} = ++$i == 7 ? ESCAPE_BEFORE_NL_STATE : ESCAPE_STATE; |
| 839 |
$self->{c} = $self->{get_char}->(); |
| 840 |
redo A; |
| 841 |
} elsif (0x0041 <= $self->{c} and $self->{c} <= 0x0046) { # A..F |
| 842 |
$char = $char * 0x10 + $self->{c} - 0x0041 + 0xA; |
| 843 |
$self->{state} = ++$i == 7 ? ESCAPE_BEFORE_NL_STATE : ESCAPE_STATE; |
| 844 |
$self->{c} = $self->{get_char}->(); |
| 845 |
redo A; |
| 846 |
} elsif (0x0061 <= $self->{c} and $self->{c} <= 0x0066) { # a..f |
| 847 |
$char = $char * 0x10 + $self->{c} - 0x0061 - 0xA; |
| 848 |
$self->{state} = ++$i == 7 ? ESCAPE_BEFORE_NL_STATE : ESCAPE_STATE; |
| 849 |
$self->{c} = $self->{get_char}->(); |
| 850 |
redo A; |
| 851 |
} elsif ($self->{c} == 0x0020 or # SP |
| 852 |
$self->{c} == 0x000A or # \n |
| 853 |
$self->{c} == 0x0009 or # \t |
| 854 |
$self->{c} == 0x000C) { # \f |
| 855 |
$self->{t}->{value} .= chr $char; |
| 856 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 857 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 858 |
$self->{c} = $self->{get_char}->(); |
| 859 |
redo A; |
| 860 |
} elsif ($self->{c} == 0x000D) { # \r |
| 861 |
$self->{state} = ESCAPE_BEFORE_LF_STATE; |
| 862 |
$self->{c} = $self->{get_char}->(); |
| 863 |
redo A; |
| 864 |
} else { |
| 865 |
$self->{t}->{value} .= chr $char; |
| 866 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 867 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 868 |
# reconsume |
| 869 |
redo A; |
| 870 |
} |
| 871 |
} elsif ($self->{state} == ESCAPE_BEFORE_NL_STATE) { |
| 872 |
## NOTE: eightth character of |unicode| in |escape|. |
| 873 |
if ($self->{c} == 0x0020 or # SP |
| 874 |
$self->{c} == 0x000A or # \n |
| 875 |
$self->{c} == 0x0009 or # \t |
| 876 |
$self->{c} == 0x000C) { # \f |
| 877 |
$self->{t}->{value} .= chr $char; |
| 878 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 879 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 880 |
$self->{c} = $self->{get_char}->(); |
| 881 |
redo A; |
| 882 |
} elsif ($self->{c} == 0x000D) { # \r |
| 883 |
$self->{state} = ESCAPE_BEFORE_NL_STATE; |
| 884 |
$self->{c} = $self->{get_char}->(); |
| 885 |
redo A; |
| 886 |
} else { |
| 887 |
$self->{t}->{value} .= chr $char; |
| 888 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 889 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 890 |
# reconsume |
| 891 |
redo A; |
| 892 |
} |
| 893 |
} elsif ($self->{state} == ESCAPE_BEFORE_LF_STATE) { |
| 894 |
## NOTE: |\n| in |\r\n| in |unicode| in |escape|. |
| 895 |
if ($self->{c} == 0x000A) { # \n |
| 896 |
$self->{t}->{value} .= chr $char; |
| 897 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 898 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 899 |
$self->{c} = $self->{get_char}->(); |
| 900 |
redo A; |
| 901 |
} else { |
| 902 |
$self->{t}->{value} .= chr $char; |
| 903 |
$self->{state} = $q == 0 ? NAME_STATE : |
| 904 |
$q == 1 ? URI_UNQUOTED_STATE : STRING_STATE; |
| 905 |
# reconsume |
| 906 |
redo A; |
| 907 |
} |
| 908 |
} elsif ($self->{state} == STRING_STATE) { |
| 909 |
## NOTE: A character in |string$Q| in |string| in |STRING|, or |
| 910 |
## a character in |invalid$Q| in |invalid| in |INVALID|, |
| 911 |
## where |$Q = $q == 0x0022 ? 1 : 2|. |
| 912 |
## Or, in |URI|. |
| 913 |
if ($self->{c} == 0x005C) { # \ |
| 914 |
$self->{state} = ESCAPE_OPEN_STATE; |
| 915 |
$self->{c} = $self->{get_char}->(); |
| 916 |
redo A; |
| 917 |
} elsif ($self->{c} == $q) { # " | ' |
| 918 |
if ($self->{t}->{type} == STRING_TOKEN) { |
| 919 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 920 |
$self->{c} = $self->{get_char}->(); |
| 921 |
return $self->{t}; |
| 922 |
#redo A; |
| 923 |
} else { |
| 924 |
$self->{state} = URI_AFTER_WSP_STATE; |
| 925 |
$self->{c} = $self->{get_char}->(); |
| 926 |
redo A; |
| 927 |
} |
| 928 |
} elsif ($self->{c} == 0x000A or # \n |
| 929 |
$self->{c} == 0x000D or # \r |
| 930 |
$self->{c} == 0x000C or # \f |
| 931 |
$self->{c} == -1) { |
| 932 |
$self->{t}->{type} = INVALID_TOKEN; |
| 933 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 934 |
# reconsume |
| 935 |
return $self->{t}; |
| 936 |
#redo A; |
| 937 |
} else { |
| 938 |
$self->{t}->{value} .= chr $self->{c}; |
| 939 |
# stay in the state |
| 940 |
$self->{c} = $self->{get_char}->(); |
| 941 |
redo A; |
| 942 |
} |
| 943 |
} elsif ($self->{state} == NUMBER_STATE) { |
| 944 |
## NOTE: 2nd, 3rd, or ... character in |num| before |.|. |
| 945 |
if (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { |
| 946 |
$self->{t}->{value} .= chr $self->{c}; |
| 947 |
# stay in the state |
| 948 |
$self->{c} = $self->{get_char}->(); |
| 949 |
redo A; |
| 950 |
} elsif ($self->{c} == 0x002E) { # . |
| 951 |
$self->{state} = NUMBER_DOT_STATE; |
| 952 |
$self->{c} = $self->{get_char}->(); |
| 953 |
redo A; |
| 954 |
} else { |
| 955 |
$self->{t}->{number} = $self->{t}->{value}; |
| 956 |
$self->{t}->{value} = ''; |
| 957 |
$self->{state} = AFTER_NUMBER_STATE; |
| 958 |
# reprocess |
| 959 |
redo A; |
| 960 |
} |
| 961 |
} elsif ($self->{state} == NUMBER_DOT_STATE) { |
| 962 |
## NOTE: The character immediately following |.| in |num|. |
| 963 |
if (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { |
| 964 |
$self->{t}->{value} .= '.' . chr $self->{c}; |
| 965 |
$self->{state} = NUMBER_DOT_NUMBER_STATE; |
| 966 |
$self->{c} = $self->{get_char}->(); |
| 967 |
redo A; |
| 968 |
} else { |
| 969 |
unshift @{$self->{token}}, {type => DELIM_STATE, value => '.'}; |
| 970 |
$self->{t}->{number} = $self->{t}->{value}; |
| 971 |
$self->{t}->{value} = ''; |
| 972 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 973 |
# reprocess |
| 974 |
return $self->{t}; |
| 975 |
#redo A; |
| 976 |
} |
| 977 |
} elsif ($self->{state} == NUMBER_FRACTION_STATE) { |
| 978 |
## NOTE: The character immediately following |.| at the beginning of |num|. |
| 979 |
if (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { |
| 980 |
$self->{t}->{value} .= '.' . chr $self->{c}; |
| 981 |
$self->{state} = NUMBER_DOT_NUMBER_STATE; |
| 982 |
$self->{c} = $self->{get_char}->(); |
| 983 |
redo A; |
| 984 |
} else { |
| 985 |
$self->{state} = BEFORE_TOKEN_STATE; |
| 986 |
$self->{c} = $self->{get_char}->(); |
| 987 |
return {type => DELIM_TOKEN, value => '.'}; |
| 988 |
#redo A; |
| 989 |
} |
| 990 |
} elsif ($self->{state} == NUMBER_DOT_NUMBER_STATE) { |
| 991 |
## NOTE: |[0-9]| in |num| after |.|. |
| 992 |
if (0x0030 <= $self->{c} and $self->{c} <= 0x0039) { |
| 993 |
$self->{t}->{value} .= chr $self->{c}; |
| 994 |
# stay in the state |
| 995 |
$self->{c} = $self->{get_char}->(); |
| 996 |
redo A; |
| 997 |
} else { |
| 998 |
$self->{t}->{number} = $self->{t}->{value}; |
| 999 |
$self->{t}->{value} = ''; |
| 1000 |
$self->{state} = AFTER_NUMBER_STATE; |
| 1001 |
# reprocess |
| 1002 |
redo A; |
| 1003 |
} |
| 1004 |
} else { |
| 1005 |
die "$0: Unknown state |$self->{state}|"; |
| 1006 |
} |
| 1007 |
} # A |
| 1008 |
} # get_next_token |
| 1009 |
|
| 1010 |
1; |
| 1011 |
# $Date: 2007/09/08 03:25:05 $ |
Parent Directory
| 
Revision Log