IncrementalEngine.ml 12.8 KB
Newer Older
1 2
open General

3 4 5 6 7 8 9 10 11
(* This signature describes the incremental LR engine. *)

(* In this mode, the user controls the lexer, and the parser suspends
   itself when it needs to read a new token. *)

module type INCREMENTAL_ENGINE = sig

  type token

12 13
  (* The type ['a checkpoint] represents an intermediate or final state of the
     parser. An intermediate checkpoint is a suspension: it records the parser's
14 15 16 17
     current state, and allows parsing to be resumed. The parameter ['a] is
     the type of the semantic value that will eventually be produced if the
     parser succeeds. *)

18
  (* [Accepted] and [Rejected] are final checkpoints. [Accepted] carries a
19 20
     semantic value. *)

21
  (* [InputNeeded] is an intermediate checkpoint. It means that the parser wishes
22 23
     to read one token before continuing. *)

24
  (* [Shifting] is an intermediate checkpoint. It means that the parser is taking
25 26 27 28 29
     a shift transition. It exposes the state of the parser before and after
     the transition. The Boolean parameter tells whether the parser intends to
     request a new token after this transition. (It always does, except when
     it is about to accept.) *)

30
  (* [AboutToReduce] is an intermediate checkpoint. It means that the parser is
31 32
     about to perform a reduction step. It exposes the parser's current
     state as well as the production that is about to be reduced. *)
33

34
  (* [HandlingError] is an intermediate checkpoint. It means that the parser has
35
     detected an error and is currently handling it, in several steps. *)
36

37
  type env
38

39 40
  type production

41
  type 'a checkpoint = private
42
    | InputNeeded of env
43
    | Shifting of env * env * bool
44 45
    | AboutToReduce of env * production
    | HandlingError of env
46 47 48 49
    | Accepted of 'a
    | Rejected

  (* [offer] allows the user to resume the parser after it has suspended
50 51
     itself with a checkpoint of the form [InputNeeded env]. [offer] expects the
     old checkpoint as well as a new token and produces a new checkpoint. It does not
52
     raise any exception. *)
53 54

  val offer:
55
    'a checkpoint ->
56
    token * Lexing.position * Lexing.position ->
57
    'a checkpoint
58

59
  (* [resume] allows the user to resume the parser after it has suspended
60 61 62
     itself with a checkpoint of the form [AboutToReduce (env, prod)] or
     [HandlingError env]. [resume] expects the old checkpoint and produces a new
     checkpoint. It does not raise any exception. *)
63

64
  val resume:
65 66
    'a checkpoint ->
    'a checkpoint
67

68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
  (* A token supplier is a function of no arguments which delivers a new token
     (together with its start and end positions) every time it is called. *)

  type supplier =
    unit -> token * Lexing.position * Lexing.position

  (* A pair of a lexer and a lexing buffer can be easily turned into a supplier. *)

  val lexer_lexbuf_to_supplier:
    (Lexing.lexbuf -> token) ->
    Lexing.lexbuf ->
    supplier

  (* The functions [offer] and [resume] are sufficient to write a parser loop.
     One can imagine many variations (which is why we expose these functions
     in the first place!). Here, we expose a few variations of the main loop,
     ready for use. *)

  (* [loop supplier checkpoint] begins parsing from [checkpoint], reading
     tokens from [supplier]. It continues parsing until it reaches a
     checkpoint of the form [Accepted v] or [Rejected]. In the former case, it
POTTIER Francois's avatar
POTTIER Francois committed
89
     returns [v]. In the latter case, it raises the exception [Error]. *)
90 91 92 93 94 95 96

  val loop: supplier -> 'a checkpoint -> 'a

  (* [loop_handle succeed fail supplier checkpoint] begins parsing from
     [checkpoint], reading tokens from [supplier]. It continues parsing until
     it reaches a checkpoint of the form [Accepted v] or [HandlingError env]
     (or [Rejected], but that should not happen, as [HandlingError _] will be
POTTIER Francois's avatar
POTTIER Francois committed
97 98
     observed first). In the former case, it calls [succeed v]. In the latter
     case, it calls [fail] with this checkpoint. It cannot raise [Error].
99 100 101 102

     This means that Menhir's traditional error-handling procedure (which pops
     the stack until a state that can act on the [error] token is found) does
     not get a chance to run. Instead, the user can implement her own error
POTTIER Francois's avatar
POTTIER Francois committed
103
     handling code, in the [fail] continuation. *)
104 105 106 107 108

  val loop_handle:
    ('a -> 'answer) ->
    ('a checkpoint -> 'answer) ->
    supplier -> 'a checkpoint -> 'answer
109

POTTIER Francois's avatar
POTTIER Francois committed
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
  (* [loop_handle_undo] is analogous to [loop_handle], except it passes a pair
     of checkpoints to the failure continuation.

     The first (and oldest) checkpoint is the last [InputNeeded] checkpoint that
     was encountered before the error was detected. The second (and newest)
     checkpoint is where the error was detected, as in [loop_handle]. Going back
     to the first checkpoint can be thought of as undoing any reductions that
     were performed after seeing the problematic token. (These reductions must
     be default reductions or spurious reductions.)

     [loop_handle_undo] must initially be applied to an [InputNeeded] checkpoint.
     The parser's initial checkpoints satisfy this constraint. *)

  val loop_handle_undo:
    ('a -> 'answer) ->
    ('a checkpoint -> 'a checkpoint -> 'answer) ->
    supplier -> 'a checkpoint -> 'answer

POTTIER Francois's avatar
POTTIER Francois committed
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
  (* [loop_test f checkpoint accu] assumes that [checkpoint] has been obtained
     by submitting a token to the parser. It runs the parser from [checkpoint],
     through an arbitrary number of reductions, until the parser either accepts
     this token (i.e., shifts) or rejects it (i.e., signals an error). If the
     parser decides to shift, then the accumulator is updated by applying the
     user function [f] to the [env] just before shifting and to the old [accu].
     Otherwise, the accumulator is not updated, i.e., [accu] is returned. *)

  (* It is desirable that the semantic actions be side-effect free, or that
     their side-effects be harmless (replayable). *)

  val loop_test:
    (env -> 'accu -> 'accu) ->
    'a checkpoint -> 'accu -> 'accu

POTTIER Francois's avatar
POTTIER Francois committed
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
  (* The function [loop_test] can be used, after an error has been detected, to
     dynamically test which tokens would have been accepted at this point. We
     provide this test, ready for use. *)

  (* For completeness, one must undo any spurious reductions before carrying out
     this test -- that is, one must apply [acceptable] to the FIRST checkpoint
     that is passed by [loop_handle_undo] to its failure continuation. *)

  (* This test causes some semantic actions to be run! The semantic actions
     should be side-effect free, or their side-effects should be harmless. *)

  (* The position [pos] is used as the start and end positions of the
     hypothetical token, and may be picked up by the semantic actions. We
     suggest using the position where the error was detected. *)

  val acceptable: 'a checkpoint -> token -> Lexing.position -> bool

160 161 162 163 164 165
  (* The abstract type ['a lr1state] describes the non-initial states of the
     LR(1) automaton. The index ['a] represents the type of the semantic value
     associated with this state's incoming symbol. *)

  type 'a lr1state

166 167 168 169
  (* The states of the LR(1) automaton are numbered (from 0 and up). *)

  val number: _ lr1state -> int

170 171 172
  (* An element is a pair of a non-initial state [s] and a semantic value [v]
     associated with the incoming symbol of this state. The idea is, the value
     [v] was pushed onto the stack just before the state [s] was entered. Thus,
173
     for some type ['a], the state [s] has type ['a lr1state] and the value [v]
174 175 176 177 178 179
     has type ['a]. In other words, the type [element] is an existential type. *)

  type element =
    | Element: 'a lr1state * 'a * Lexing.position * Lexing.position -> element

  (* The parser's stack is (or, more precisely, can be viewed as) a stream of
POTTIER Francois's avatar
POTTIER Francois committed
180
     elements. The type [stream] is defined by the module [General]. *)
181 182 183 184

  type stack =
    element stream

185 186 187 188
  (* This is the parser's stack, a stream of elements. This stream is empty if
     the parser is in an initial state; otherwise, it is non-empty.  The LR(1)
     automaton's current state is the one found in the top element of the
     stack. *)
189

190
  val stack: env -> stack
191

192 193 194 195 196 197 198
  (* These are the start and end positions of the current lookahead token. It
     is legal to invoke this function only after at least one token has been
     offered to the parser via [offer]. In other words, it is illegal to
     invoke it in an initial state. *)

  val positions: env -> Lexing.position * Lexing.position

199 200 201 202 203 204 205
  (* This tells whether the parser is about to perform a default reduction.
     In particular, when applied to an environment taken from a result of
     the form [AboutToReduce (env, prod)], this tells whether the reduction
     that is about to take place is a default reduction. *)

  val has_default_reduction: env -> bool

206
end
207

208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237
(* This signature is a fragment of the inspection API that is made available
   to the user when [--inspection] is used. This fragment contains type
   definitions for symbols. *)

module type SYMBOLS = sig

  (* The type ['a terminal] represents a terminal symbol. The type ['a
     nonterminal] represents a nonterminal symbol. In both cases, the index
     ['a] represents the type of the semantic values associated with this
     symbol. The concrete definitions of these types are generated. *)

  type 'a terminal
  type 'a nonterminal

  (* The type ['a symbol] represents a terminal or nonterminal symbol. It is
     the disjoint union of the types ['a terminal] and ['a nonterminal]. *)

  type 'a symbol =
    | T : 'a terminal -> 'a symbol
    | N : 'a nonterminal -> 'a symbol

  (* The type [xsymbol] is an existentially quantified version of the type
     ['a symbol]. This type is useful in situations where the index ['a]
     is not statically known. *)

  type xsymbol = 
    | X : 'a symbol -> xsymbol

end

238 239
(* This signature describes the inspection API that is made available to the
   user when [--inspection] is used. *)
240 241 242

module type INSPECTION = sig

243 244 245 246
  (* The types of symbols are described above. *)

  include SYMBOLS

247 248
  (* The type ['a lr1state] is meant to be the same as in [INCREMENTAL_ENGINE]. *)

249
  type 'a lr1state
250

251 252 253 254
  (* The type [production] is meant to be the same as in [INCREMENTAL_ENGINE].
     It represents a production of the grammar. A production can be examined
     via the functions [lhs] and [rhs] below. *)

255 256
  type production

257 258 259 260
  (* An LR(0) item is a pair of a production [prod] and a valid index [i] into
     this production. That is, if the length of [rhs prod] is [n], then [i] is
     comprised between 0 and [n], inclusive. *)

261
  type item =
262 263
      production * int

264 265 266 267 268 269 270 271
  (* Ordering functions. *)

  val compare_terminals: _ terminal -> _ terminal -> int
  val compare_nonterminals: _ nonterminal -> _ nonterminal -> int
  val compare_symbols: xsymbol -> xsymbol -> int
  val compare_productions: production -> production -> int
  val compare_items: item -> item -> int

272 273 274 275 276 277
  (* [incoming_symbol s] is the incoming symbol of the state [s], that is,
     the symbol that the parser must recognize before (has recognized when)
     it enters the state [s]. This function gives access to the semantic
     value [v] stored in a stack element [Element (s, v, _, _)]. Indeed,
     by case analysis on the symbol [incoming_symbol s], one discovers the
     type ['a] of the value [v]. *)
278

POTTIER Francois's avatar
POTTIER Francois committed
279
  val incoming_symbol: 'a lr1state -> 'a symbol
280

POTTIER Francois's avatar
POTTIER Francois committed
281 282 283 284 285 286
  (* [items s] is the set of the LR(0) items in the LR(0) core of the LR(1)
     state [s]. This set is not epsilon-closed. This set is presented as a
     list, in an arbitrary order. *)

  val items: _ lr1state -> item list

287 288 289
  (* [lhs prod] is the left-hand side of the production [prod]. This is
     always a non-terminal symbol. *)

290 291
  val lhs: production -> xsymbol

292 293 294
  (* [rhs prod] is the right-hand side of the production [prod]. This is
     a (possibly empty) sequence of (terminal or nonterminal) symbols. *)

295 296
  val rhs: production -> xsymbol list

297 298 299 300
  (* [nullable nt] tells whether the non-terminal symbol [nt] is nullable.
     That is, it is true if and only if this symbol produces the empty
     word [epsilon]. *)

301
  val nullable: _ nonterminal -> bool
302

303 304 305 306
  (* [first nt t] tells whether the FIRST set of the nonterminal symbol [nt]
     contains the terminal symbol [t]. That is, it is true if and only if
     [nt] produces a word that begins with [t]. *)

307 308 309 310 311 312
  val first: _ nonterminal -> _ terminal -> bool

  (* [xfirst] is analogous to [first], but expects a first argument of type
     [xsymbol] instead of [_ terminal]. *)

  val xfirst: xsymbol -> _ terminal -> bool
313

314 315 316 317 318 319 320
  (* [foreach_terminal] enumerates the terminal symbols, including [error].
     [foreach_terminal_but_error] enumerates the terminal symbols, excluding
     [error]. *)

  val foreach_terminal:           (xsymbol -> 'a -> 'a) -> 'a -> 'a
  val foreach_terminal_but_error: (xsymbol -> 'a -> 'a) -> 'a -> 'a

321 322
end

323 324 325 326 327 328 329 330 331 332 333 334
(* This signature combines the incremental API and the inspection API. *)

module type EVERYTHING = sig

  include INCREMENTAL_ENGINE

  include INSPECTION
    with type 'a lr1state := 'a lr1state
    with type production := production

end