grammarFunctor.mli 17 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
(* The functor [Make] transforms an abstract syntax tree for the grammar into a
   rich internal representation of the grammar. *)

(* The reason why this is now a functor, and the reason why its verbosity can
   be controlled, is that we may wish to invoke it several times, e.g. on the
   grammar before %inlining, and on the grammar after %inlining. 2015/11/10 *)

module Make (G : sig

  (* An abstract syntax tree for the grammar. *)
  val grammar: UnparameterizedSyntax.grammar

  (* This flag indicates whether it is OK to produce warnings, verbose
     information, etc., when this functor is invoked. If it is set to
     [false], then only serious errors can be signaled. *)
  val verbose: bool

end) : sig
19 20 21 22 23 24 25 26 27 28

(* ------------------------------------------------------------------------ *)
(* Nonterminals. *)

module Nonterminal : sig

  (* The type of nonterminals. *)

  type t

29 30 31 32
  (* Comparison. *)

  val compare: t -> t -> int

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55
  (* The number of nonterminals. This includes the extra nonterminals
     that are internally generated for the grammar's entry points. *)

  val n: int

  (* [lookup] maps an identifier to a nonterminal, or raises [Not_found]. *)

  val lookup : string -> t

  (* Nonterminals can be converted to integers. This feature is
     exploited in the table-based back-end. *)

  val n2i: t -> int

  (* This produces a string representation of a nonterminal. It should
     in principle never be applied to one of the internally generated
     nonterminals, as we do not wish users to become aware of the
     existence of these extra nonterminals. However, we do sometimes
     violate this rule when it is difficult to do otherwise.

     The Boolean parameter tells whether the string representation
     should be normalized, that is, whether parentheses and commas
     should be eliminated. This is necessary if the string is intended
56
     for use as a valid nonterminal name or as a valid OCaml
57 58 59 60
     identifier. *)

  val print: bool -> t -> string

61
  (* This is the OCaml type associated with a nonterminal
62 63 64 65 66 67
     symbol. It is known only if a %type declaration was provided.
     This function is not applicable to the internally generated
     nonterminals. *)

  val ocamltype: t -> Stretch.ocamltype option

68 69 70 71 72
  (* A start symbol always has a type. This allows us to define
     a simplified version of [ocamltype] for start symbols. *)

  val ocamltype_of_start_symbol: t -> Stretch.ocamltype

73 74 75 76 77 78 79 80 81 82 83
  (* Iteration over nonterminals. The order in which elements are
     examined, and the order of [map]'s output list, correspond to the
     numeric indices produced by [n2i] above. *)

  val iter: (t -> unit) -> unit
  val fold: (t -> 'a -> 'a) -> 'a -> 'a
  val map: (t -> 'a) -> 'a list

  (* Iteration over all nonterminals, except the start nonterminals. *)

  val iterx: (t -> unit) -> unit
POTTIER Francois's avatar
POTTIER Francois committed
84
  val foldx: (t -> 'a -> 'a) -> 'a -> 'a
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102

  (* Tabulation of a function over nonterminals. *)

  val tabulate: (t -> 'a) -> (t -> 'a)

  (* [positions nt] is a list of the positions associated with the
     definition of [nt]. There can be more than one position because
     definitions can be split over multiple files. *)

  val positions: t -> Positions.t list

  (* This tells whether a non-terminal symbol is one of the start
     symbols. *)

  val is_start: t -> bool

end

103 104 105 106 107 108 109
(* ------------------------------------------------------------------------ *)
(* Sets of nonterminals. *)

module NonterminalMap : GMap.S with type key = Nonterminal.t

module NonterminalSet = NonterminalMap.Domain

110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
(* ------------------------------------------------------------------------ *)
(* Terminals. *)

module Terminal : sig

  (* The type of terminals. *)

  type t

  (* The number of terminals. This includes the two pseudo-tokens
     [#] and [error]. *)

  val n: int

  (* Comparison. *)

  val equal: t -> t -> bool
127
  val compare: t -> t -> int
128 129 130 131 132

  (* [lookup] maps an identifier to a terminal, or raises [Not_found]. *)

  val lookup : string -> t

133 134 135
  (* Terminals can be converted to integers. This feature is exploited in the
     table-based back-end and in [LRijkstra]. The reverse conversion, [i2t],
     is unsafe and should not be used. [LRijkstra] uses it :-) *)
136 137

  val t2i: t -> int
138
  val i2t: int -> t (* unsafe! *)
139 140 141 142 143

  (* This produces a string representation of a terminal. *)

  val print: t -> string

144
  (* This is the OCaml type associated with a terminal
145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
     symbol. It is known only if the %token declaration was
     accompanied with a type. *)

  val ocamltype: t -> Stretch.ocamltype option

  (* These are the two pseudo-tokens [#] and [error]. The former is
     used to denote the end of the token stream. The latter is
     accessible to the user and is used for handling errors. *)

  val sharp: t
  val error: t

  (* This is the programmer-defined [EOF] token, if there is one. It
     is recognized based solely on its name, which is fragile, but
     this behavior is documented. This token is assumed to represent
160 161
     [ocamllex]'s [eof] pattern. It is used only by the reference
     interpreter, and in a rather non-essential way. *)
162 163 164

  val eof: t option

POTTIER Francois's avatar
POTTIER Francois committed
165
  (* A terminal symbol is pseudo if it is [#] or [error]. It is real otherwise. *)
166 167

  val pseudo: t -> bool
POTTIER Francois's avatar
POTTIER Francois committed
168
  val real: t -> bool
169 170 171

  (* Iteration over terminals. The order in which elements are
     examined, and the order of [map]'s output list, correspond to the
POTTIER Francois's avatar
POTTIER Francois committed
172
     numeric indices produced by [t2i] above. *)
173 174 175 176

  val iter: (t -> unit) -> unit
  val fold: (t -> 'a -> 'a) -> 'a -> 'a
  val map: (t -> 'a) -> 'a list
POTTIER Francois's avatar
POTTIER Francois committed
177 178 179

  (* [mapx] offers iteration over all terminals except [#]. *)

180 181
  val mapx: (t -> 'a) -> 'a list

POTTIER Francois's avatar
POTTIER Francois committed
182
  (* [iter_real] offers iteration over all real terminals. *)
POTTIER Francois's avatar
POTTIER Francois committed
183 184 185

  val iter_real: (t -> unit) -> unit

186 187 188 189 190
  (* The sub-module [Word] offers an implementation of words (that is,
     sequences) of terminal symbols. It is used by [LRijkstra]. We
     make it a functor, because it has internal state (a hash table)
     and a side effect (failure if there are more than 256 terminal
     symbols). *)
POTTIER Francois's avatar
POTTIER Francois committed
191

192 193 194 195 196 197
  (* The type [word] should be treated, as much as possible, as an
     abstract type. In fact, for efficiency reasons, we represent a
     word as a unique integer codes, and we allocate these integer
     codes sequentially, from 0 upwards. The conversion from [int]
     to [word] is of course unsafe and should be used wisely. *)

POTTIER Francois's avatar
POTTIER Francois committed
198 199
  module Word (X : sig end) : sig

200
    type word = int
POTTIER Francois's avatar
POTTIER Francois committed
201 202 203 204 205 206 207 208 209 210 211
    val epsilon: word
    val singleton: t -> word
    val append: word -> word -> word
    val length: word -> int
    (* [first w z] returns the first symbol of the word [w.z]. *)
    val first: word -> t -> t
    val elements: word -> t list
    val print: word -> string
    (* [verbose()] prints statistics about the use of the internal
       hash-consing table so far. *)
    val verbose: unit -> unit
POTTIER Francois's avatar
POTTIER Francois committed
212 213
    (* Lexicographic ordering. *)
    val compare: word -> word -> int
POTTIER Francois's avatar
POTTIER Francois committed
214
  end
215

216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
end

(* ------------------------------------------------------------------------ *)
(* Sets and maps over terminals. *)

module TerminalSet : sig

  (* All of the operations documented in [GSet] are available. *)

  include GSet.S with type element = Terminal.t

  (* This offers a string representation of a set of terminals. The
     symbols are simply listed one after the other and separated with
     spaces. *)

  val print: t -> string

  (* This is the set of all terminal symbols except the pseudo-tokens
     [#] and [error]. *)

  val universe: t

end

(* All of the operations documented in [GMap] are available. *)

module TerminalMap : GMap.S with type key = Terminal.t

(* ------------------------------------------------------------------------ *)
(* Symbols. *)

module Symbol : sig

  (* A symbol is either a nonterminal or a terminal. *)

  type t =
    | N of Nonterminal.t
    | T of Terminal.t

255 256 257 258
  (* [lookup] maps an identifier to a symbol, or raises [Not_found]. *)

  val lookup : string -> t

259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311
  (* Comparison. *)

  val equal: t -> t -> bool
  val lequal: t list -> t list -> bool

  (* These produce a string representation of a symbol, of a list of
     symbols, or of an array of symbols. The symbols are simply listed
     one after the other and separated with spaces. [printao] prints
     an array of symbols, starting at a particular offset. [printaod]
     is analogous, but can also print a single dot at a particular
     position between two symbols. *)

  val print: t -> string
  val printl: t list -> string
  val printa: t array -> string
  val printao: int -> t array -> string
  val printaod: int -> int -> t array -> string

end

(* ------------------------------------------------------------------------ *)
(* Sets and maps over symbols. *)

(* All of the operations documented in [Set] are available. *)

module SymbolSet : Set.S with type elt = Symbol.t

module SymbolMap : sig

  (* All of the operations documented in [Map] are available. *)

  include Map.S with type key = Symbol.t

  val domain: 'a t -> key list

  (* This returns [true] if and only if all of the symbols in
     the domain of the map at hand are nonterminals. *)

  val purelynonterminal: 'a t -> bool

end

(* ------------------------------------------------------------------------ *)
(* Productions. *)

module Production : sig

  (* This is the type of productions. This includes user-defined
     productions as well as the internally generated productions
     associated with the start symbols. *)

  type index

312 313 314 315
  (* Comparison. *)

  val compare: index -> index -> int

316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361
  (* Productions can be converted to integers and back. This is unsafe
     and should be avoided as much as possible. This feature is
     exploited, for efficiency, in the encoding of items. *)

  val p2i: index -> int
  val i2p: int -> index

  (* The number of productions. *)

  val n: int

  (* These map a production index to the production's definition, that
     is, a nonterminal (the left-hand side) and an array of symbols
     (the right-hand side). *)

  val def: index -> Nonterminal.t * Symbol.t array
  val nt: index -> Nonterminal.t
  val rhs: index -> Symbol.t array
  val length: index -> int

  (* This maps a production index to an array of the identifiers that
     should be used for naming the semantic values of the symbols in
     the right-hand side. *)

  val identifiers: index -> Syntax.identifier array

  (* This maps a production index to the production's semantic action.
     This function is not applicable to a start production. *)

  val action: index -> Syntax.action

  (* [positions prod] is a list of the positions associated with
     production [prod]. This is usually a singleton list, but there
     can be more than one position for start productions when the
     definition of the corresponding start symbol is split over
     multiple files. *)

  val positions: index -> Positions.t list

  (* Iteration over all productions. The order in which elements
     are examined, and the order of [map]'s output list, correspond
     to the numeric indices produced by [p2i] above. *)

  val iter: (index -> unit) -> unit
  val fold: (index -> 'a -> 'a) -> 'a -> 'a
  val map: (index -> 'a) -> 'a list
POTTIER Francois's avatar
POTTIER Francois committed
362
  val amap: (index -> 'a) -> 'a array
363 364 365 366 367

  (* Iteration over all productions, except the start productions. *)

  val iterx: (index -> unit) -> unit
  val foldx: (index -> 'a -> 'a) -> 'a -> 'a
368
  val mapx: (index -> 'a) -> 'a list
369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392

  (* This maps a (user) non-terminal start symbol to the corresponding
     start production. *)

  val startsymbol2startprod: Nonterminal.t -> index

  (* Iteration over the productions associated with a specific
     nonterminal. *)

  val iternt: Nonterminal.t -> (index -> unit) -> unit
  val foldnt: Nonterminal.t -> 'a -> (index -> 'a -> 'a) -> 'a

  (* This allows determining whether a production is a start
     production. If it is a start production, the start symbol that it
     is associated with is returned. If it is a regular production,
     nothing is returned. *)

  val classify: index -> Nonterminal.t option

  (* [is_start] is easier to use than [classify] when the start symbol
     is not needed. *)

  val is_start: index -> bool

393 394 395 396 397
  (* The integer [start] is published so as to allow the table back-end
     to produce code for [is_start]. It should not be used otherwise. *)

  val start: int

398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
  (* This produces a string representation of a production. It should
     never be applied to a start production, as we do not wish users
     to become aware of the existence of these extra productions. *)

  val print: index -> string

  (* Tabulation of a Boolean function over productions. [tabulateb f]
     returns a tabulated version of [f] as well as the number of
     productions where [f] is true. *)

  val tabulate: (index -> 'a) -> (index -> 'a)
  val tabulateb: (index -> bool) -> (index -> bool) * int

end

(* ------------------------------------------------------------------------ *)
(* Maps over productions. *)

module ProductionMap : sig

  include GMap.S with type key = Production.index

  (* Iteration over the start productions only. *)

  val start: (Production.index -> 'a) -> 'a t

end

(* ------------------------------------------------------------------------ *)
(* Analysis of the grammar. *)

module Analysis : sig

431 432 433 434 435
  (* [nullable nt] is the NULLABLE flag of the non-terminal symbol [nt].
     That is, it is true if and only if this symbol produces the empty
     word [epsilon]. *)

  val nullable: Nonterminal.t -> bool
436
  val nullable_symbol: Symbol.t -> bool
437

POTTIER Francois's avatar
POTTIER Francois committed
438 439 440
  (* [first nt] is the FIRST set of the non-terminal symbol [nt]. *)

  val first: Nonterminal.t -> TerminalSet.t
POTTIER Francois's avatar
POTTIER Francois committed
441
  val first_symbol: Symbol.t -> TerminalSet.t
POTTIER Francois's avatar
POTTIER Francois committed
442

POTTIER Francois's avatar
POTTIER Francois committed
443
  (* [nullable_first_prod prod i] considers the suffix of the production
444
     [prod] defined by offset [i]. It returns its NULLABLE flag as well
445
     as its FIRST set. The offset [i] must be contained between [0] and
446
     [n], inclusive, where [n] is the length of production [prod]. *)
447

448
  val nullable_first_prod: Production.index -> int -> bool * TerminalSet.t
449

450 451 452 453 454 455 456
  (* [first_prod_lookahead prod i t] computes [FIRST(alpha.t)], where [alpha]
     is the suffix of the production defined by offset [i], and [t] is a
     terminal symbol. The offset [i] must be contained between [0] and [n],
     inclusive, where [n] is the length of production [prod]. *)

  val first_prod_lookahead: Production.index -> int -> Terminal.t -> TerminalSet.t

457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516
  (* [explain_first_rhs tok rhs i] explains why the token [tok] appears
     in the FIRST set for the string of symbols found at offset [i] in
     the array [rhs]. *)

  val explain_first_rhs: Terminal.t -> Symbol.t array -> int -> string

  (* [follow nt] is the FOLLOW set of the non-terminal symbol [nt], that
     is, the set of terminal symbols that could follow an expansion of
     [nt] in a valid sentence. *)

  val follow: Nonterminal.t -> TerminalSet.t

end

(* ------------------------------------------------------------------------ *)
(* Conflict resolution via precedences. *)

module Precedence : sig

  (* Shift/reduce conflicts require making a choice between shifting a
     token and reducing a production. How these choices are made is of
     no concern to the back-end, but here is a rough explanation.

     Shifting is preferred when the token has higher precedence than
     the production, or they have same precedence and the token is
     right-associative.

     Reducing is preferred when the token has lower precedence than
     the production, or they have same precedence and the token is
     left-associative.

     Neither is allowed when the token and the production have same
     precedence and the token is non-associative.

     No preference is explicitly specified when the token or the
     production has undefined precedence. In that case, the default
     choice is to prefer shifting, but a conflict will be reported. *)

  type choice =
    | ChooseShift
    | ChooseReduce
    | ChooseNeither
    | DontKnow

  val shift_reduce: Terminal.t -> Production.index -> choice

  (* Reduce/reduce conflicts require making a choice between reducing
     two distinct productions. This is done by exploiting a partial
     order on productions.

     For compatibility with ocamlyacc, this order should be total and
     should correspond to textual order when the two productions
     originate in the same source file. When they originate in
     different source files, the two productions should be
     incomparable. *)

  val reduce_reduce: Production.index -> Production.index -> Production.index option

end

517 518 519 520 521
(* ------------------------------------------------------------------------ *)
(* %on_error_reduce declarations. *)

module OnErrorReduce : sig

522 523 524 525
  (* This is the set of %on_error_reduce declarations. Each declaration
     carries a level, which is used when several declarations are
     applicable in a single state. *)
  val declarations: Syntax.on_error_reduce_level StringMap.t
526 527 528

end

529 530 531
(* ------------------------------------------------------------------------ *)
(* Diagnostics. *)

POTTIER Francois's avatar
POTTIER Francois committed
532 533 534
(* This function prints warnings about useless precedence declarations for
   terminal symbols (%left, %right, %nonassoc) and productions (%prec). It
   should be invoked after only the automaton has been constructed. *)
535 536 537

val diagnostics: unit -> unit

538 539 540
(* ------------------------------------------------------------------------ *)

end (* module Make *)