1 (* Title: HOL/TPTP/TPTP_Parser/tptp_syntax.ML
2 Author: Nik Sultana, Cambridge University Computer Laboratory
4 TPTP abstract syntax and parser-related definitions.
7 signature TPTP_SYNTAX =
9 exception TPTP_SYNTAX of string
10 val debug: ('a -> unit) -> 'a -> unit
12 (*Note that in THF "^ [X] : ^ [Y] : f @ g" should parse
13 as "(^ [X] : (^ [Y] : f)) @ g"
16 datatype number_kind = Int_num | Real_num | Rat_num
18 datatype status_value =
19 Suc | Unp | Sap | Esa | Sat | Fsa
20 | Thm | Eqv | Tac | Wec | Eth | Tau
21 | Wtc | Wth | Cax | Sca | Tca | Wca
22 | Cup | Csp | Ecs | Csa | Cth | Ceq
23 | Unc | Wcc | Ect | Fun | Uns | Wuc
24 | Wct | Scc | Uca | Noc
27 type atomic_word = string
28 type inference_rule = atomic_word
29 type file_info = name option
30 type single_quoted = string
31 type file_name = single_quoted
32 type creator_name = atomic_word
33 type variable = string
34 type upper_word = string
36 datatype language = FOF | CNF | TFF | THF | FOT | TFF_with_arithmetic
38 Role_Axiom | Role_Hypothesis | Role_Definition | Role_Assumption |
39 Role_Lemma | Role_Theorem | Role_Conjecture | Role_Negated_Conjecture |
40 Role_Plain | Role_Fi_Domain | Role_Fi_Functors | Role_Fi_Predicates |
41 Role_Type | Role_Unknown
43 and general_data = (*Bind of variable * formula_data*)
45 | Application of string * general_term list (*general_function*)
46 | V of upper_word (*variable*)
47 | Number of number_kind * string
48 | Distinct_Object of string
49 | (*formula_data*) Formula_Data of language * tptp_formula (* $thf(<thf_formula>) *)
50 | (*formula_data*) Term_Data of tptp_term
52 and interpreted_symbol =
53 UMinus | Sum | Difference | Product | Quotient | Quotient_E |
54 Quotient_T | Quotient_F | Remainder_E | Remainder_T | Remainder_F |
55 Floor | Ceiling | Truncate | Round | To_Int | To_Rat | To_Real |
56 (*FIXME these should be in defined_pred, but that's not being used in TPTP*)
57 Less | LessEq | Greater | GreaterEq | EvalEq | Is_Int | Is_Rat |
60 and logic_symbol = Equals | NEquals | Or | And | Iff | If | Fi | Xor |
61 Nor | Nand | Not | Op_Forall | Op_Exists |
62 (*FIXME these should be in defined_pred, but that's not being used in TPTP*)
65 and quantifier = (*interpreted binders*)
66 Forall | Exists | Epsilon | Iota | Lambda | Dep_Prod | Dep_Sum
69 Type_Ind | Type_Bool | Type_Type | Type_Int | Type_Rat | Type_Real
72 Uninterpreted of string
73 | Interpreted_ExtraLogic of interpreted_symbol
74 | Interpreted_Logic of logic_symbol
75 | TypeSymbol of tptp_base_type
79 General_Data of general_data (*general_data*)
80 | General_Term of general_data * general_term (*general_data : general_term*)
81 | General_List of general_term list
84 Term_Func of symbol * tptp_term list
86 | Term_Conditional of tptp_formula * tptp_term * tptp_term
87 | Term_Num of number_kind * string
88 | Term_Distinct_Object of string
89 | Term_Let of tptp_let list * tptp_term (*FIXME remove list?*)
92 TFF_Typed_Atom of symbol * tptp_type option (*only TFF*)
93 | THF_Atom_term of tptp_term (*from here on, only THF*)
94 | THF_Atom_conn_term of symbol
97 Pred of symbol * tptp_term list
98 | Fmla of symbol * tptp_formula list
99 | Sequent of tptp_formula list * tptp_formula list
100 | Quant of quantifier * (string * tptp_type option) list * tptp_formula
101 | Conditional of tptp_formula * tptp_formula * tptp_formula
102 | Let of tptp_let list * tptp_formula (*FIXME remove list?*)
104 | Type_fmla of tptp_type
105 | THF_typing of tptp_formula * tptp_type (*only THF*)
108 Let_fmla of (string * tptp_type option) * tptp_formula
109 | Let_term of (string * tptp_type option) * tptp_term (*only TFF*)
112 Prod_type of tptp_type * tptp_type
113 | Fn_type of tptp_type * tptp_type
114 | Atom_type of string
115 | Defined_type of tptp_base_type
116 | Sum_type of tptp_type * tptp_type (*only THF*)
117 | Fmla_type of tptp_formula
118 | Subtype of symbol * symbol (*only THF*)
120 type general_list = general_term list
121 type parent_details = general_list
122 type useful_info = general_term list
123 type info = useful_info
125 type annotation = general_term * general_term list
127 exception DEQUOTE of string
129 type position = string * int * int
132 Annotated_Formula of position * language * string * role *
133 tptp_formula * annotation option
134 | Include of string * string list
136 type tptp_problem = tptp_line list
138 val dequote : single_quoted -> single_quoted
140 val role_to_string : role -> string
142 val status_to_string : status_value -> string
144 val nameof_tff_atom_type : tptp_type -> string
146 (*Returns the list of all files included in a directory and its
147 subdirectories. This is only used for testing the parser/interpreter against
149 val get_file_list : Path.T -> Path.T list
151 val string_of_tptp_term : tptp_term -> string
152 val string_of_tptp_formula : tptp_formula -> string
157 structure TPTP_Syntax : TPTP_SYNTAX =
160 exception TPTP_SYNTAX of string
162 datatype number_kind = Int_num | Real_num | Rat_num
164 datatype status_value =
165 Suc | Unp | Sap | Esa | Sat | Fsa
166 | Thm | Eqv | Tac | Wec | Eth | Tau
167 | Wtc | Wth | Cax | Sca | Tca | Wca
168 | Cup | Csp | Ecs | Csa | Cth | Ceq
169 | Unc | Wcc | Ect | Fun | Uns | Wuc
170 | Wct | Scc | Uca | Noc
173 type atomic_word = string
174 type inference_rule = atomic_word
175 type file_info = name option
176 type single_quoted = string
177 type file_name = single_quoted
178 type creator_name = atomic_word
179 type variable = string
180 type upper_word = string
182 datatype language = FOF | CNF | TFF | THF | FOT | TFF_with_arithmetic
184 Role_Axiom | Role_Hypothesis | Role_Definition | Role_Assumption |
185 Role_Lemma | Role_Theorem | Role_Conjecture | Role_Negated_Conjecture |
186 Role_Plain | Role_Fi_Domain | Role_Fi_Functors | Role_Fi_Predicates |
187 Role_Type | Role_Unknown
188 and general_data = (*Bind of variable * formula_data*)
189 Atomic_Word of string
190 | Application of string * (general_term list)
191 | V of upper_word (*variable*)
192 | Number of number_kind * string
193 | Distinct_Object of string
194 | (*formula_data*) Formula_Data of language * tptp_formula (* $thf(<thf_formula>) *)
195 | (*formula_data*) Term_Data of tptp_term
197 and interpreted_symbol =
198 UMinus | Sum | Difference | Product | Quotient | Quotient_E |
199 Quotient_T | Quotient_F | Remainder_E | Remainder_T | Remainder_F |
200 Floor | Ceiling | Truncate | Round | To_Int | To_Rat | To_Real |
201 Less | LessEq | Greater | GreaterEq | EvalEq | Is_Int | Is_Rat |
205 and logic_symbol = Equals | NEquals | Or | And | Iff | If | Fi | Xor |
206 Nor | Nand | Not | Op_Forall | Op_Exists |
209 and quantifier = (*interpreted binders*)
210 Forall | Exists | Epsilon | Iota | Lambda | Dep_Prod | Dep_Sum
213 Type_Ind | Type_Bool | Type_Type | Type_Int | Type_Rat | Type_Real
216 Uninterpreted of string
217 | Interpreted_ExtraLogic of interpreted_symbol
218 | Interpreted_Logic of logic_symbol
219 | TypeSymbol of tptp_base_type
223 General_Data of general_data (*general_data*)
224 | General_Term of general_data * general_term (*general_data : general_term*)
225 | General_List of general_term list
228 Term_Func of symbol * tptp_term list
230 | Term_Conditional of tptp_formula * tptp_term * tptp_term
231 | Term_Num of number_kind * string
232 | Term_Distinct_Object of string
233 | Term_Let of tptp_let list * tptp_term (*FIXME remove list?*)
236 TFF_Typed_Atom of symbol * tptp_type option (*only TFF*)
237 | THF_Atom_term of tptp_term (*from here on, only THF*)
238 | THF_Atom_conn_term of symbol
241 Pred of symbol * tptp_term list
242 | Fmla of symbol * tptp_formula list
243 | Sequent of tptp_formula list * tptp_formula list
244 | Quant of quantifier * (string * tptp_type option) list * tptp_formula
245 | Conditional of tptp_formula * tptp_formula * tptp_formula
246 | Let of tptp_let list * tptp_formula
248 | Type_fmla of tptp_type
249 | THF_typing of tptp_formula * tptp_type
252 Let_fmla of (string * tptp_type option) * tptp_formula
253 | Let_term of (string * tptp_type option) * tptp_term
256 Prod_type of tptp_type * tptp_type
257 | Fn_type of tptp_type * tptp_type
258 | Atom_type of string
259 | Defined_type of tptp_base_type
260 | Sum_type of tptp_type * tptp_type
261 | Fmla_type of tptp_formula
262 | Subtype of symbol * symbol
264 type general_list = general_term list
265 type parent_details = general_list
266 type useful_info = general_term list
267 type info = useful_info
269 (*type annotation = (source * info option)*)
270 type annotation = general_term * general_term list
272 exception DEQUOTE of string
274 type position = string * int * int
277 Annotated_Formula of position * language * string * role * tptp_formula * annotation option
278 | Include of string * string list
280 type tptp_problem = tptp_line list
282 fun debug f x = if !Runtime.debug then (f x; ()) else ()
284 fun nameof_tff_atom_type (Atom_type str) = str
285 | nameof_tff_atom_type _ = raise TPTP_SYNTAX "nameof_tff_atom_type called on non-atom type"
287 (*Used for debugging. Returns all files contained within a directory or its
288 subdirectories. Follows symbolic links, filters away directories.*)
289 fun get_file_list path =
291 fun check_file_entry f rest =
293 (*NOTE needed since no File.is_link and File.read_link*)
294 val f_str = Path.implode f
296 if File.is_dir f then
297 rest @ get_file_list f
298 else if OS.FileSys.isLink f_str then
299 (*follow links -- NOTE this breaks if links are relative paths*)
300 check_file_entry (Path.explode (OS.FileSys.readLink f_str)) rest
308 |> (fn l => fold check_file_entry l [])
311 fun role_to_string role =
313 Role_Axiom => "axiom"
314 | Role_Hypothesis => "hypothesis"
315 | Role_Definition => "definition"
316 | Role_Assumption => "assumption"
317 | Role_Lemma => "lemma"
318 | Role_Theorem => "theorem"
319 | Role_Conjecture => "conjecture"
320 | Role_Negated_Conjecture => "negated_conjecture"
321 | Role_Plain => "plain"
322 | Role_Fi_Domain => "fi_domain"
323 | Role_Fi_Functors => "fi_functors"
324 | Role_Fi_Predicates => "fi_predicates"
325 | Role_Type => "type"
326 | Role_Unknown => "unknown"
328 (*accepts a string "'abc'" and returns "abc"*)
329 fun dequote str : single_quoted =
331 raise (DEQUOTE "empty string")
336 if str = "unprefix" then
337 raise DEQUOTE ("string doesn't open with quote:" ^ str)
338 else if str = "unsuffix" then
339 raise DEQUOTE ("string doesn't close with quote:" ^ str)
343 (* Printing parsed TPTP formulas *)
344 (*FIXME this is not pretty-printing, just printing*)
346 fun status_to_string status_value =
348 Suc => "suc" | Unp => "unp"
349 | Sap => "sap" | Esa => "esa"
350 | Sat => "sat" | Fsa => "fsa"
351 | Thm => "thm" | Wuc => "wuc"
352 | Eqv => "eqv" | Tac => "tac"
353 | Wec => "wec" | Eth => "eth"
354 | Tau => "tau" | Wtc => "wtc"
355 | Wth => "wth" | Cax => "cax"
356 | Sca => "sca" | Tca => "tca"
357 | Wca => "wca" | Cup => "cup"
358 | Csp => "csp" | Ecs => "ecs"
359 | Csa => "csa" | Cth => "cth"
360 | Ceq => "ceq" | Unc => "unc"
361 | Wcc => "wcc" | Ect => "ect"
362 | Fun => "fun" | Uns => "uns"
363 | Wct => "wct" | Scc => "scc"
364 | Uca => "uca" | Noc => "noc"
366 fun string_of_tptp_term x =
368 Term_Func (symbol, tptp_term_list) =>
369 "(" ^ string_of_symbol symbol ^ " " ^
370 String.concatWith " " (map string_of_tptp_term tptp_term_list) ^ ")"
371 | Term_Var str => str
372 | Term_Conditional (tptp_formula, tptp_term1, tptp_term2) => "" (*FIXME*)
373 | Term_Num (_, str) => str
374 | Term_Distinct_Object str => str
376 and string_of_symbol (Uninterpreted str) = str
377 | string_of_symbol (Interpreted_ExtraLogic interpreted_symbol) = string_of_interpreted_symbol interpreted_symbol
378 | string_of_symbol (Interpreted_Logic logic_symbol) = string_of_logic_symbol logic_symbol
379 | string_of_symbol (TypeSymbol tptp_base_type) = string_of_tptp_base_type tptp_base_type
380 | string_of_symbol (System str) = str
382 and string_of_tptp_base_type Type_Ind = "$i"
383 | string_of_tptp_base_type Type_Bool = "$o"
384 | string_of_tptp_base_type Type_Type = "$tType"
385 | string_of_tptp_base_type Type_Int = "$int"
386 | string_of_tptp_base_type Type_Rat = "$rat"
387 | string_of_tptp_base_type Type_Real = "$real"
389 and string_of_interpreted_symbol x =
393 | Difference => "$difference"
394 | Product => "$product"
395 | Quotient => "$quotient"
396 | Quotient_E => "$quotient_e"
397 | Quotient_T => "$quotient_t"
398 | Quotient_F => "$quotient_f"
399 | Remainder_E => "$remainder_e"
400 | Remainder_T => "$remainder_t"
401 | Remainder_F => "$remainder_f"
403 | Ceiling => "$ceiling"
404 | Truncate => "$truncate"
406 | To_Int => "$to_int"
407 | To_Rat => "$to_rat"
408 | To_Real => "$to_real"
410 | LessEq => "$lesseq"
411 | Greater => "$greater"
412 | GreaterEq => "$greatereq"
413 | EvalEq => "$evaleq"
414 | Is_Int => "$is_int"
415 | Is_Rat => "$is_rat"
418 and string_of_logic_symbol Equals = "="
419 | string_of_logic_symbol NEquals = "!="
420 | string_of_logic_symbol Or = "|"
421 | string_of_logic_symbol And = "&"
422 | string_of_logic_symbol Iff = "<=>"
423 | string_of_logic_symbol If = "=>"
424 | string_of_logic_symbol Fi = "<="
425 | string_of_logic_symbol Xor = "<~>"
426 | string_of_logic_symbol Nor = "~|"
427 | string_of_logic_symbol Nand = "~&"
428 | string_of_logic_symbol Not = "~"
429 | string_of_logic_symbol Op_Forall = "!!"
430 | string_of_logic_symbol Op_Exists = "??"
431 | string_of_logic_symbol True = "$true"
432 | string_of_logic_symbol False = "$false"
434 and string_of_quantifier Forall = "!"
435 | string_of_quantifier Exists = "?"
436 | string_of_quantifier Epsilon = "@+"
437 | string_of_quantifier Iota = "@-"
438 | string_of_quantifier Lambda = "^"
439 | string_of_quantifier Dep_Prod = "!>"
440 | string_of_quantifier Dep_Sum = "?*"
442 and string_of_tptp_atom (TFF_Typed_Atom (symbol, tptp_type_option)) =
443 (case tptp_type_option of
444 NONE => string_of_symbol symbol
446 string_of_symbol symbol ^ " : " ^ string_of_tptp_type tptp_type)
447 | string_of_tptp_atom (THF_Atom_term tptp_term) = string_of_tptp_term tptp_term
448 | string_of_tptp_atom (THF_Atom_conn_term symbol) = string_of_symbol symbol
450 and string_of_tptp_formula (Pred (symbol, tptp_term_list)) =
451 "(" ^ string_of_symbol symbol ^
452 String.concatWith " " (map string_of_tptp_term tptp_term_list) ^ ")"
453 | string_of_tptp_formula (Fmla (symbol, tptp_formula_list)) =
455 string_of_symbol symbol ^
456 String.concatWith " " (map string_of_tptp_formula tptp_formula_list) ^ ")"
457 | string_of_tptp_formula (Sequent (tptp_formula_list1, tptp_formula_list2)) = "" (*FIXME*)
458 | string_of_tptp_formula (Quant (quantifier, varlist, tptp_formula)) =
459 string_of_quantifier quantifier ^ "[" ^
460 String.concatWith ", " (map (fn (n, ty) =>
463 | SOME ty => n ^ " : " ^ string_of_tptp_type ty) varlist) ^ "] : (" ^
464 string_of_tptp_formula tptp_formula ^ ")"
465 | string_of_tptp_formula (Conditional _) = "" (*FIXME*)
466 | string_of_tptp_formula (Let _) = "" (*FIXME*)
467 | string_of_tptp_formula (Atom tptp_atom) = string_of_tptp_atom tptp_atom
468 | string_of_tptp_formula (Type_fmla tptp_type) = string_of_tptp_type tptp_type
469 | string_of_tptp_formula (THF_typing (tptp_formula, tptp_type)) =
470 string_of_tptp_formula tptp_formula ^ " : " ^ string_of_tptp_type tptp_type
472 and string_of_tptp_type (Prod_type (tptp_type1, tptp_type2)) =
473 string_of_tptp_type tptp_type1 ^ " * " ^ string_of_tptp_type tptp_type2
474 | string_of_tptp_type (Fn_type (tptp_type1, tptp_type2)) =
475 string_of_tptp_type tptp_type1 ^ " > " ^ string_of_tptp_type tptp_type2
476 | string_of_tptp_type (Atom_type str) = str
477 | string_of_tptp_type (Defined_type tptp_base_type) =
478 string_of_tptp_base_type tptp_base_type
479 | string_of_tptp_type (Sum_type (tptp_type1, tptp_type2)) = ""
480 | string_of_tptp_type (Fmla_type tptp_formula) = string_of_tptp_formula tptp_formula
481 | string_of_tptp_type (Subtype (symbol1, symbol2)) =
482 string_of_symbol symbol1 ^ " << " ^ string_of_symbol symbol2