icuSources/common/rbbirpt.txt

   1
   2 #*****************************************************************************
   3 #
   4 #   Copyright (C) 2002-2016, International Business Machines Corporation and others.
   5 #   All Rights Reserved.
   6 #
   7 #*****************************************************************************
   8 #
   9 #  file:  rbbirpt.txt
  10 #  ICU Break Iterator Rule Parser State Table
  11 #
  12 #     This state table is used when reading and parsing a set of RBBI rules
  13 #     The rule parser uses a state machine; the data in this file define the
  14 #     state transitions that occur for each input character.
  15 #
  16 #     *** This file defines the RBBI rule grammar.   This is it.
  17 #     *** The determination of what is accepted is here.
  18 #
  19 #     This file is processed by a perl script "rbbicst.pl" to produce initialized C arrays
  20 #     that are then built with the rule parser.
  21 #
  22 #    perl rbbicst.pl    < rbbirpt.txt > rbbirpt.h
  23
  24 #
  25 # Here is the syntax of the state definitions in this file:
  26 #
  27 #
  28 #StateName:
  29 #   input-char           n next-state           ^push-state     action
  30 #   input-char           n next-state           ^push-state     action
  31 #       |                |   |                      |             |
  32 #       |                |   |                      |             |--- action to be performed by state machine
  33 #       |                |   |                      |                  See function RBBIRuleScanner::doParseActions()
  34 #       |                |   |                      |
  35 #       |                |   |                      |--- Push this named state onto the state stack.
  36 #       |                |   |                           Later, when next state is specified as "pop",
  37 #       |                |   |                           the pushed state will become the current state.
  38 #       |                |   |
  39 #       |                |   |--- Transition to this state if the current input character matches the input
  40 #       |                |        character or char class in the left hand column.  "pop" causes the next
  41 #       |                |        state to be popped from the state stack.
  42 #       |                |
  43 #       |                |--- When making the state transition specified on this line, advance to the next
  44 #       |                     character from the input only if 'n' appears here.
  45 #       |
  46 #       |--- Character or named character classes to test for.  If the current character being scanned
  47 #            matches, peform the actions and go to the state specified on this line.
  48 #            The input character is tested sequentally, in the order written.  The characters and
  49 #            character classes tested for do not need to be mutually exclusive.  The first match wins.
  50 #
  51
  52
  53
  54
  55 #
  56 #  start state, scan position is at the beginning of the rules file, or in between two rules.
  57 #
  58 start:
  59     escaped                term                  ^break-rule-end    doExprStart
  60     white_space          n start
  61     '^'                  n start-after-caret     ^break-rule-end    doNoChain
  62     '$'                    scan-var-name         ^assign-or-rule    doExprStart
  63     '!'                  n rev-option
  64     ';'                  n start                                                  # ignore empty rules.
  65     eof                    exit
  66     default                term                  ^break-rule-end    doExprStart
  67
  68 #
  69 #  break-rule-end:  Returned from doing a break-rule expression.
  70 #
  71 break-rule-end:
  72     ';'                  n start                                    doEndOfRule
  73     white_space          n break-rule-end
  74     default                errorDeath                               doRuleError
  75
  76 #
  77 # start of a rule, after having seen a '^' (inhibits rule chain in).
  78 #     Similar to the main 'start' state in most respects, except
  79 #          - empty rule is an error.
  80 #          - A second '^' is an error.
  81 #
  82 start-after-caret:
  83     escaped                term                                     doExprStart
  84     white_space          n start-after-caret
  85     '^'                    errorDeath                               doRuleError    # two '^'s
  86     '$'                    scan-var-name         ^term-var-ref      doExprStart
  87     ';'                    errorDeath                               doRuleError    # ^ ;
  88     eof                    errorDeath                               doRuleError
  89     default                term                                     doExprStart
  90
  91 #
  92 #   !               We've just scanned a '!', indicating either a !!key word flag or a
  93 #                   !Reverse rule.
  94 #
  95 rev-option:
  96     '!'                  n option-scan1
  97     default                reverse-rule           ^break-rule-end   doReverseDir
  98
  99 option-scan1:
 100     name_start_char      n option-scan2                             doOptionStart
 101     default                errorDeath                               doRuleError
 102
 103 option-scan2:
 104     name_char            n option-scan2
 105     default                option-scan3                             doOptionEnd
 106
 107 option-scan3:
 108     ';'                  n start
 109     white_space          n option-scan3
 110     default                errorDeath                               doRuleError
 111
 112
 113 reverse-rule:
 114     default                term                   ^break-rule-end   doExprStart
 115
 116
 117 #
 118 #  term.  Eat through a single rule character, or a composite thing, which
 119 #         could be a parenthesized expression, a variable name, or a Unicode Set.
 120 #
 121 term:
 122     escaped              n expr-mod                                 doRuleChar
 123     white_space          n term
 124     rule_char            n expr-mod                                 doRuleChar
 125     '['                    scan-unicode-set      ^expr-mod
 126     '('                  n term                  ^expr-mod          doLParen
 127     '$'                    scan-var-name         ^term-var-ref
 128     '.'                  n expr-mod                                 doDotAny
 129     default                errorDeath                               doRuleError
 130
 131
 132
 133 #
 134 #  term-var-ref   We've just finished scanning a reference to a $variable.
 135 #                 Check that the variable was defined.
 136 #                 The variable name scanning is in common with assignment statements,
 137 #                 so the check can't be done there.
 138 term-var-ref:
 139     default                expr-mod                                 doCheckVarDef
 140
 141
 142 #
 143 #   expr-mod      We've just finished scanning a term, now look for the optional
 144 #                 trailing '*', '?', '+'
 145 #
 146 expr-mod:
 147     white_space          n  expr-mod
 148     '*'                  n  expr-cont                               doUnaryOpStar
 149     '+'                  n  expr-cont                               doUnaryOpPlus
 150     '?'                  n  expr-cont                               doUnaryOpQuestion
 151     default                 expr-cont
 152
 153
 154 #
 155 #  expr-cont      Expression, continuation.  At a point where additional terms are
 156 #                                            allowed, but not required.
 157 #
 158 expr-cont:
 159     escaped                 term                                    doExprCatOperator
 160     white_space          n  expr-cont
 161     rule_char               term                                    doExprCatOperator
 162     '['                     term                                    doExprCatOperator
 163     '('                     term                                    doExprCatOperator
 164     '$'                     term                                    doExprCatOperator
 165     '.'                     term                                    doExprCatOperator
 166     '/'                     look-ahead                              doExprCatOperator
 167     '{'                  n  tag-open                                doExprCatOperator
 168     '|'                  n  term                                    doExprOrOperator
 169     ')'                  n  pop                                     doExprRParen
 170     default                 pop                                     doExprFinished
 171
 172
 173 #
 174 #   look-ahead    Scanning a '/', which identifies a break point, assuming that the
 175 #                 remainder of the expression matches.
 176 #
 177 #                 Generate a parse tree as if this was a special kind of input symbol
 178 #                 appearing in an otherwise normal concatenation expression.
 179 #
 180 look-ahead:
 181     '/'                   n expr-cont-no-slash                      doSlash
 182     default                 errorDeath
 183
 184
 185 #
 186 #  expr-cont-no-slash    Expression, continuation.  At a point where additional terms are
 187 #                                            allowed, but not required.  Just like
 188 #                                            expr-cont, above, except that no '/'
 189 #                                            look-ahead symbol is permitted.
 190 #
 191 expr-cont-no-slash:
 192     escaped                 term                                    doExprCatOperator
 193     white_space          n  expr-cont
 194     rule_char               term                                    doExprCatOperator
 195     '['                     term                                    doExprCatOperator
 196     '('                     term                                    doExprCatOperator
 197     '$'                     term                                    doExprCatOperator
 198     '.'                     term                                    doExprCatOperator
 199     '|'                  n  term                                    doExprOrOperator
 200     ')'                  n  pop                                     doExprRParen
 201     default                 pop                                     doExprFinished
 202
 203
 204 #
 205 #   tags             scanning a '{', the opening delimiter for a tag that identifies
 206 #                    the kind of match.  Scan the whole {dddd} tag, where d=digit
 207 #
 208 tag-open:
 209     white_space          n  tag-open
 210     digit_char              tag-value                               doStartTagValue
 211     default                 errorDeath                              doTagExpectedError
 212
 213 tag-value:
 214     white_space          n  tag-close
 215     '}'                     tag-close
 216     digit_char           n  tag-value                               doTagDigit
 217     default                 errorDeath                              doTagExpectedError
 218
 219 tag-close:
 220     white_space          n  tag-close
 221     '}'                  n  expr-cont-no-tag                        doTagValue
 222     default                 errorDeath                              doTagExpectedError
 223
 224
 225
 226 #
 227 #  expr-cont-no-tag    Expression, continuation.  At a point where additional terms are
 228 #                                            allowed, but not required.  Just like
 229 #                                            expr-cont, above, except that no "{ddd}"
 230 #                                            tagging is permitted.
 231 #
 232 expr-cont-no-tag:
 233     escaped                 term                                    doExprCatOperator
 234     white_space          n  expr-cont-no-tag
 235     rule_char               term                                    doExprCatOperator
 236     '['                     term                                    doExprCatOperator
 237     '('                     term                                    doExprCatOperator
 238     '$'                     term                                    doExprCatOperator
 239     '.'                     term                                    doExprCatOperator
 240     '/'                     look-ahead                              doExprCatOperator
 241     '|'                  n  term                                    doExprOrOperator
 242     ')'                  n  pop                                     doExprRParen
 243     default                 pop                                     doExprFinished
 244
 245
 246
 247
 248 #
 249 #   Variable Name Scanning.
 250 #
 251 #                    The state that branched to here must have pushed a return state
 252 #                    to go to after completion of the variable name scanning.
 253 #
 254 #                    The current input character must be the $ that introduces the name.
 255 #                    The $ is consummed here rather than in the state that first detected it
 256 #                    so that the doStartVariableName action only needs to happen in one
 257 #                    place (here), and the other states don't need to worry about it.
 258 #
 259 scan-var-name:
 260    '$'                  n scan-var-start                            doStartVariableName
 261    default                errorDeath
 262
 263
 264 scan-var-start:
 265     name_start_char      n scan-var-body
 266     default                errorDeath                               doVariableNameExpectedErr
 267
 268 scan-var-body:
 269     name_char            n scan-var-body
 270     default                pop                                      doEndVariableName
 271
 272
 273
 274 #
 275 #  scan-unicode-set   Unicode Sets are parsed by the the UnicodeSet class.
 276 #                     Within the RBBI parser, after finding the first character
 277 #                     of a Unicode Set, we just hand the rule input at that
 278 #                     point of to the Unicode Set constructor, then pick
 279 #                     up parsing after the close of the set.
 280 #
 281 #                     The action for this state invokes the UnicodeSet parser.
 282 #
 283 scan-unicode-set:
 284     '['                   n pop                                      doScanUnicodeSet
 285     'p'                   n pop                                      doScanUnicodeSet
 286     'P'                   n pop                                      doScanUnicodeSet
 287     default                 errorDeath
 288
 289
 290
 291
 292
 293
 294
 295 #
 296 #  assign-or-rule.   A $variable was encountered at the start of something, could be
 297 #                    either an assignment statement or a rule, depending on whether an '='
 298 #                    follows the variable name.  We get to this state when the variable name
 299 #                    scanning does a return.
 300 #
 301 assign-or-rule:
 302     white_space          n assign-or-rule
 303     '='                  n term                  ^assign-end        doStartAssign   # variable was target of assignment
 304     default                term-var-ref          ^break-rule-end                    # variable was a term in a rule
 305
 306
 307
 308 #
 309 #  assign-end        This state is entered when the end of the expression on the
 310 #                    right hand side of an assignment is found.  We get here via
 311 #                    a pop; this state is pushed when the '=' in an assignment is found.
 312 #
 313 #                    The only thing allowed at this point is a ';'.  The RHS of an
 314 #                    assignment must look like a rule expression, and we come here
 315 #                    when what is being scanned no longer looks like an expression.
 316 #
 317 assign-end:
 318     ';'                  n start                                    doEndAssign
 319     default                errorDeath                               doRuleErrorAssignExpr
 320
 321
 322
 323 #
 324 # errorDeath.   This state is specified as the next state whenever a syntax error
 325 #               in the source rules is detected.  Barring bugs, the state machine will never
 326 #               actually get here, but will stop because of the action associated with the error.
 327 #               But, just in case, this state asks the state machine to exit.
 328 errorDeath:
 329     default              n errorDeath                               doExit
 330
 331