/usr/share/acl2-6.3/books/paco/database.lisp is in acl2-books-source 6.3-5.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 | (in-package "ACL2")
; This file defines the macro db so that (acl2::db acl2::state) is the
; Paco analogue of the current ACL2 logical world. This file can only
; be loaded after all the Paco files have been loaded, since it might
; conceptually use any Paco record or data structure.
; Does the ACL2 sometimes contain multiple entries for the same key
; and property? Yes. Every :logic mode function in axioms.lisp is
; defined twice, so the properties as ABSOLUTE-EVENT-NUMBER, FORMALS,
; GUARD, STOBJS-IN, STOBJS-OUT, SYMBOL-CLASS, and TYPE-PRESCRIPTIONS
; are all duplicated many times. In addition, even user-defined
; functions may redefine both SYMBOL-CLASS and TYPE-PRESCRIPTIONS, as
; well as all the properties, e.g., LEMMAS, associated with newly
; added rules. (Some stats, FYI: The boot strap world contains
; approximately 48K triples. Of these, approximately 27K have unique
; symbol/property pairs. There are about 3K symbol/property pairs
; that occur multiple times. It takes about a minute to sweep the
; world and find the duplicate entries in the completely naive way.
; The way we implement it here is about 30 times faster.)
; Seen is an alist pairing symbols with all the properties we have so
; far seen. We say a trip (sym prop . val) is ``in'' seen if prop is
; an element of the value associated with sym in seen.
; Some ACL2 record accessor forms expand into terms involving the
; :program mode function record-error. (In particular, the access
; forms for "non-cheap" records.) This means that the functions that
; copy these records over to the Paco world have to be in :program
; mode. Rather than make a careful determination of which records
; have this problem, I just do the entire development in :program
; mode.
(program)
(set-state-ok t)
(defun trip-seen (trip seen)
; We determine if trip is in seen.
(let ((temp (assoc-eq (car trip) seen)))
(cond
((null temp) nil)
(t (member-eq (cadr trip) (cdr temp))))))
(defun mark-as-seen (trip seen)
; Assuming trip is not yet in seen, we put it there.
(let ((temp (assoc-eq (car trip) seen)))
(cond
((null temp) (cons (cons (car trip) (list (cadr trip))) seen))
(t (put-assoc-eq (car trip)
(cons (cadr trip) (cdr temp))
seen)))))
(defun copy-recognizer-alist (alist)
(cond
((endp alist) nil)
(t (cons
(paco::make paco::recognizer-tuple
:nume (access recognizer-tuple (car alist) :nume)
:fn (access recognizer-tuple (car alist) :fn)
:true-ts (access recognizer-tuple (car alist) :true-ts)
:false-ts (access recognizer-tuple (car alist) :false-ts)
:strongp (access recognizer-tuple (car alist) :strongp))
(copy-recognizer-alist (cdr alist))))))
(defun copy-type-prescriptions (tp-lst)
(cond
((endp tp-lst) nil)
(t (cons
(paco::make paco::type-prescription
:nume (access type-prescription (car tp-lst) :nume)
:basic-ts (access type-prescription (car tp-lst) :basic-ts)
:term (access type-prescription (car tp-lst) :term)
:hyps (access type-prescription (car tp-lst) :hyps)
:vars (access type-prescription (car tp-lst) :vars))
(copy-type-prescriptions (cdr tp-lst))))))
(defun acceptable-rewrite-hyps (hyps)
(cond ((endp hyps) t)
((or (variablep (car hyps))
(fquotep (car hyps)))
(acceptable-rewrite-hyps (cdr hyps)))
((or (eq (ffn-symb (car hyps)) 'FORCE)
(eq (ffn-symb (car hyps)) 'CASE-SPLIT))
nil)
((and (eq (ffn-symb (car hyps)) 'SYNP)
(not (eq (car (cadr (fargn (car hyps) 2))) 'SYNTAXP)))
nil)
(t (acceptable-rewrite-hyps (cdr hyps)))))
(defun copy-lemmas (lemmas)
(cond
((endp lemmas) nil)
((acceptable-rewrite-hyps
(access rewrite-rule (car lemmas) :hyps))
(cons
(paco::make
paco::rewrite-rule
:nume
(access rewrite-rule (car lemmas) :nume)
:hyps
(access rewrite-rule (car lemmas) :hyps)
:equiv
(access rewrite-rule (car lemmas) :equiv)
:lhs
(access rewrite-rule (car lemmas) :lhs)
:rhs
(access rewrite-rule (car lemmas) :rhs)
:subclass
(let ((subclass (access rewrite-rule (car lemmas) :subclass)))
(case subclass
(abbreviation 'paco::backchain)
(meta 'paco::meta)
(definition 'paco::definition)
(otherwise 'paco::backchain)))
:heuristic-info
(cond ((eq (access rewrite-rule (car lemmas) :subclass) 'definition)
(cons (car (access rewrite-rule (car lemmas) :heuristic-info))
(list (cdr (access rewrite-rule (car lemmas) :heuristic-info)))))
(t (access rewrite-rule (car lemmas) :heuristic-info))))
(copy-lemmas (cdr lemmas))))
(t (copy-lemmas (cdr lemmas)))))
(defun copy-elim-rule (x)
(paco::make
paco::elim-rule
:nume (access elim-rule x :nume) ; nume
:crucial-position (access elim-rule x :crucial-position) ; number
:destructor-term (access elim-rule x :destructor-term) ; term
:destructor-terms (access elim-rule x :destructor-terms) ; terms
:hyps (access elim-rule x :hyps) ; terms
:equiv 'EQUAL ; fn symbol
:lhs (access elim-rule x :lhs) ; term
:rhs (access elim-rule x :rhs)) ; term
)
(defun copy-built-in-clauses1 (bic-lst)
(cond ((endp bic-lst) nil)
(t (cons (paco::make
paco::built-in-clause
:nume
(access built-in-clause (car bic-lst) :nume)
:all-fnnames
(access built-in-clause (car bic-lst) :all-fnnames)
:clause
(access built-in-clause (car bic-lst) :clause))
(copy-built-in-clauses1 (cdr bic-lst))))))
(defun copy-built-in-clauses (bic-alist)
(cond ((endp bic-alist) nil)
(t (cons (cons (caar bic-alist)
(copy-built-in-clauses1 (cdar bic-alist)))
(copy-built-in-clauses (cdr bic-alist))))))
(defun copy-induction-machine (lst)
; An ACL2 induction-machine is a list of tests-and-calls records.
; Both tests and calls are just lists of terms and so don't need to be
; changed. Does the record need to be changed? That would depend on
; whether both records are of type "cheap". Rather than answer that,
; I copy the record.
(cond ((endp lst) nil)
(t (cons (paco::make
paco::tests-and-calls
:tests (access tests-and-calls (car lst) :tests)
:calls (access tests-and-calls (car lst) :calls))
(copy-induction-machine (cdr lst))))))
; The quick-block-info for a function is a list of symbols as shown below.
; Paco could test against the ACL2 incarnation of those symbols rather than
; its own.
(defun copy-quick-block-info (lst)
(cond ((endp lst) nil)
(t (cons (case (car lst)
(self-reflexive 'paco::self-reflexive)
(unchanging 'paco::unchanging)
(otherwise 'paco::questionable))
(copy-quick-block-info (cdr lst))))))
(defun copy-justification (j)
(paco::make
paco::justification
:subset (access justification j :subset) ; list of vars
:mp (access justification j :mp) ; pred symbol (e.g. E0-ORDINALP)
:rel (access justification j :rel) ; rel symbol (e.g. E0-ORD-<)
:measure (access justification j :measure); term
))
(defun copy-induction-rules (lst)
(cond
((endp lst) nil)
(t (cons (paco::make
paco::induction-rule
:nume (access induction-rule (car lst) :nume) ; nume
:pattern (access induction-rule (car lst) :pattern) ; term
:condition (access induction-rule (car lst) :condition) ; term
:scheme (access induction-rule (car lst) :scheme) ; term
)
(copy-induction-rules (cdr lst))))))
(defun copy-type-set-inverter-rules (lst)
; This function exposes a lie in our bland reassurance that an ACL2
; term can be transferred to the Paco world without any changes.
; Type-set inverter rules contains terms in the variable X. But in
; ACL2 that variable is acl2::x while in Paco it is paco::x. So we
; have to rename it.
(cond
((endp lst) nil)
(t (cons
(paco::make
paco::type-set-inverter-rule
:nume (access type-set-inverter-rule (car lst) :nume) ; nume
:ts (access type-set-inverter-rule (car lst) :ts) ; type-set
:terms (subst-var-lst ; terms in X
'paco::x
'acl2::x
(access type-set-inverter-rule (car lst) :terms))
)
(copy-type-set-inverter-rules (cdr lst))))))
(defun copy-generalize-rules (lst)
(cond
((endp lst) nil)
(t (cons
(paco::make
paco::generalize-rule
:nume (access generalize-rule (car lst) :nume) ; nume
:formula (access generalize-rule (car lst) :formula) ; term
)
(copy-generalize-rules (cdr lst))))))
(defun copy-trips (w seen a w0 paco-w0 state)
; Here, w is the ACL2 world, seen is a structure recording which trips
; we've already seen, a is the emerging Paco world. It is in reverse
; order.
; Finally, w0 is another ACL2 world and paco-w0 is the Paco world
; corresponding to w0. We use it to short-circuit the construction.
; We return (mv flg paco-w), where flg is t or nil indicating that we
; encountered w0 as a tail of w, and paco-w is the Paco world
; corresponding to w.
; Note: The only reason this function takes state is that it must
; sometimes call fn-rune-nume or getprop and needs the current ACL2
; world to do so. I thought there were enough ``w's'' among the
; arguments above not to add yet another. So I get the world out of
; state when I need it. Nothing fancy is going on with state.
(cond
((equal w w0) (mv t (revappend a paco-w0)))
((endp w) (mv nil (revappend a nil)))
((trip-seen (car w) seen)
(copy-trips (cdr w) seen a w0 paco-w0 state))
(t (let* ((new-seen (mark-as-seen (car w) seen))
(sym (caar w))
(prop (cadar w))
(val (cddar w))
(new-a
(case prop
(GLOBAL-VALUE
(case sym
(RECOGNIZER-ALIST
(cons
(list* 'PACO::RECOGNIZER-ALIST
'PACO::GLOBAL-VALUE
(copy-recognizer-alist val))
a))
(BUILT-IN-CLAUSES
(cons
(list* 'PACO::BUILT-IN-CLAUSES
'PACO::GLOBAL-VALUE
(copy-built-in-clauses val))
a))
(TYPE-SET-INVERTER-RULES
(cons
(list* 'PACO::TYPE-SET-INVERTER-RULES
'PACO::GLOBAL-VALUE
(copy-type-set-inverter-rules val))
a))
(GENERALIZE-RULES
(cons (list* 'PACO::GENERALIZE-RULES
'PACO::GLOBAL-VALUE
(copy-generalize-rules val))
a))
(otherwise a)))
(FORMALS
; Every time we see a FORMALS property we know we are dealing with a function
; symbol, fn. Paco must have some properties for fn that ACL2 does not
; maintain. The first is a FN-NUMES property, giving the numes for
; (:DEFINITION fn), (:EXECUTABLE-COUNTERPART fn) and (:INDUCTION fn). (ACL2
; has the runic mapping pairs property from which it recovers the numes.) Paco
; must also have a BODY property, where ACL2 has the more modern def-bodies
; property. Paco must also have a CONTROLLER-ALISTS property, which is a list
; of controller-alists. Each controller-alist is an alist pairing the function
; symbols of a clique with a mask of ts and nils indicating the formals that
; are measured. ACL2, as of Version 3.3, no longer has such a property and
; instead records a single controller alist in each def-body. So, when this
; function sees a FORMALS property it takes the occasion to introduce FORMALS
; to Paco's world and also to introduce FN-NUMES and CONTROLLER-ALISTS.
(list*
(list* sym
'PACO::FORMALS
val)
(list* sym
'PACO::BODY
(access def-body
(car
(last
(getprop sym 'def-bodies nil
'current-acl2-world
(w state))))
:concl))
; If you change the layout of the FN-NUMES value below, change
; fn-nume!
(list* sym
'PACO::FN-NUMES
(list
(fnume (list :DEFINITION sym) (w state))
(fnume (list :EXECUTABLE-COUNTERPART sym) (w state))
(fnume (list :INDUCTION sym) (w state))))
; We just use the controller-alist for the first def-body of the
; function.
(list* sym
'PACO::CONTROLLER-ALISTS
(list (access def-body
(car
(last
(getprop sym 'def-bodies nil
'current-acl2-world
(w state))))
:controller-alist)))
a))
((BODY UNNORMALIZED-BODY RECURSIVEP CONTROLLER-ALISTS
PRIMITIVE-RECURSIVE-DEFUNP LEVEL-NO)
; These properties have values that are the same in ACL2 as in Paco.
; Generally speaking, the only time it is safe to make that assumption
; is if the value is composed of terms, function symbols, conses (not
; "records" but just ordinary pairs and true-lists) and T and NIL. If
; the value contains other symbols, like UNCHANGING, there is the
; possibility of confusion of packages and should be copied. Note that
; we convert the property name into the PACO package.
(cons
(list* sym
(intern-in-package-of-symbol (symbol-name prop)
'PACO::REWRITE)
val)
a))
; For each of the remaining properties of interest we must translate
; from the ACL2 world to the Paco world.
(TYPE-PRESCRIPTIONS
(cons
(list* sym
'PACO::TYPE-PRESCRIPTIONS
(copy-type-prescriptions val))
a))
(LEMMAS
(cons
(list* sym
'PACO::LEMMAS
(copy-lemmas val))
a))
(ELIMINATE-DESTRUCTORS-RULE
(cond
((equal (access elim-rule val :equiv) 'EQUAL)
(cons (list* sym
'PACO::ELIMINATE-DESTRUCTORS-RULE
(copy-elim-rule val))
a))
(t a)))
(INDUCTION-MACHINE
(cons (list* sym
'PACO::INDUCTION-MACHINE
(copy-induction-machine val))
a))
(QUICK-BLOCK-INFO
(cons (list* sym
'PACO::QUICK-BLOCK-INFO
(copy-quick-block-info val))
a))
(JUSTIFICATION
(cons (list* sym
'PACO::JUSTIFICATION
(copy-justification val))
a))
(INDUCTION-RULES
(cons (list* sym
'PACO::INDUCTION-RULES
(copy-induction-rules val))
a))
(otherwise a))))
(copy-trips (cdr w) new-seen new-a w0 paco-w0 state)))))
(defun transfer-paco-w1 (w w0 paco-w0 state)
; Logically, prog2$ is the identity function for its second argument,
; so the retract-world is irrelevant. Logically, extend-world is the
; identity function for its second argument. So, logically, this
; function is just a call of copy-trips.
; Actually, because of the hidden ACL2 effects of these functions,
; this installs the property list created by copy-trips, say paco-w,
; into the Common Lisp world, under the world name paco::paco. That
; is, if acl2::getprop is ever invoked on paco-w with world name
; paco::paco, the computation goes directly to the Common Lisp
; The PSIM simulator need not do all this. In an ideal world of
; memoized functions, it would be just as fast for this function just
; to return paco-w and for paco::getprop to be the slow version.
; Note: The only reason this function takes state is so that it can
; pass it to copy-trips. The only reason copy-trips takes state is so
; that it can get the world out of it to make the call to
; fn-rune-nume. This function doesn't do anything fancy with state.
(mv-let (flg paco-w)
(copy-trips w nil nil w0 paco-w0 state)
(prog2$ (retract-world 'paco::paco (if flg paco-w0 nil))
(extend-world 'paco::paco paco-w))))
; See transfer-environments for the basic invariants concerning
; paco::paco-w and acl2::paco-w. Roughly, the first is always the
; Paco world corresponding to the second. In the function below, we
; compute and set the first. But we do not set the second. That is
; because we do not want to set it until we've computed all the
; environment variables based on the ACL2 world.
(defun transfer-paco-w (state)
; Note: This function takes state because it deals in error triples
; and sets state globals.
(let ((paco-w (if (and (boundp-global 'paco::paco-w state)
(boundp-global 'acl2::paco-w state))
(transfer-paco-w1 (w state)
(@ acl2::paco-w)
(@ paco::paco-w)
state)
(transfer-paco-w1 (w state) nil nil state))))
(er-progn (assign paco::paco-w paco-w)
; You might expect us to do this:
; (assign acl2::paco-w (w state))
; but we don't because there are other Paco variables we want to transfer.
; If we were to set acl2::paco-w now and then get interrupted while
; computing another variable, that variable would not correspond to
; the new w even though paco-w and (w state) would be equal.
(value nil))))
; ---------------------------------------------------------------------------
; Section: Mapping ACL2's ENS to Paco's
(defun copy-ens1 (i max array-name array ans)
(declare (xargs :measure (nfix (+ 1 (- max i)))))
(cond ((not (and (integerp i)
(<= 0 i)
(integerp max)
(<= 0 max)))
nil)
((> i max) (revappend ans nil))
((aref1 array-name array i)
(copy-ens1 (+ 1 i) max array-name array ans))
(t (copy-ens1 (+ 1 i) max array-name array (cons i ans)))))
(defun copy-ens (ens)
; We construct a Paco btree containing all the numes that are disabled
; by ens. If a nume is greater than the :index-of-last-enabling it is
; enabled, so we only look at numes less than or equal to that index.
; The :theory-array of the ens is a bounded-integer-alistp: a true-list of
; pairs (n . x) where each n is either :HEADER or a natural less than
; the car of the :DIMENSIONS list of the :HEADER.
; Therefore, to determine all the disabled numes we map i from 0 to
; the :index-of-last-enabling and collect every i that is not bound in
; the :theory-array. Since the returned list is ordered ascending, we
; don't have to sort it and so use Paco's make-btree1 instead of
; make-btree.
; WARNING: Some integers in this btree may not actually be numes. It
; may be possible for ACL2 to skip a number while enumerating rules.
; (Actually, it doesn't as of this writing, but I don't recall that
; being guaranteed.) But since the structure is only probed with
; numes, this will do.
(paco::make-btree1
(copy-ens1 0
(access enabled-structure ens :index-of-last-enabling)
(access enabled-structure ens :array-name)
(access enabled-structure ens :theory-array)
nil)))
(defun transfer-paco-ens (state)
(cond ((and (boundp-global 'paco::paco-ens state)
(boundp-global 'acl2::paco-ens state)
(equal (@ acl2::paco-ens) (ens state)))
(value nil))
(t (er-progn
(assign paco::paco-ens (copy-ens (ens state)))
(assign acl2::paco-ens (ens state))
(value nil)))))
; Paco deals exclusively with numes. The user may wish to know the
; rune corresponding to a nume seen in a Paco trace. But ACL2
; provides such map. We provide one.
(defun transfer-nume-to-rune-map (state)
; This function makes the value of (@ nume-to-rune-map) be a structure
; that maps all numes to their runes. Actually, the structure is an
; enabled structure and we just use the universal-theory to compute
; it.
(let ((d (car
(cadr
(assoc-keyword
:dimensions
(cdr (assoc-eq :header
(access enabled-structure (ens state)
:theory-array)))))))
(name 'NUME-TO-RUNE-MAP-0))
; We re-use the same enabled structure over and over. We start by
; asking if we've got one and if not build an empty one. Then we load
; it, if the world has changed since the last loaded it. The world
; with which it was loaded is (@ acl2::paco-w).
(er-let*
((map
(cond
((boundp-global 'nume-to-rune-map state)
(value (@ nume-to-rune-map)))
(t (er-progn
(assign
nume-to-rune-map
(make enabled-structure
:index-of-last-enabling -1
:theory-array (compress1 name
(cons (list :header
:dimensions (list d)
:maximum-length (1+ d)
:default nil
:name name)
nil))
:array-name name
:array-length d
:array-name-root
(all-but-last (coerce (symbol-name name) 'list))
:array-name-suffix 0))
(value (@ nume-to-rune-map)))))))
(cond
((and (boundp-global 'acl2::paco-w state)
(equal (@ acl2::paco-w) (w state)))
(value nil))
(t
(er-let*
((map (load-theory-into-enabled-structure
'(universal-theory-fn :here (w state)) ;;; theory-expr
(universal-theory-fn :here (w state)) ;;; theory
nil ;;; augmented-p
map ;;; ens
nil ;;; incrmt-array-name-flg
nil ;;; index-of-last-enabling
(w state) ;;; wrld
'transfer-nume-to-rune-map ;;; ctx
state) ;;; state
))
(er-progn
(assign nume-to-rune-map map)
; We do not assign acl2::paco-w until we have transferred all environment
; variables.
(value nil))))))))
; The functions for using the map, nume-to-rune and numes-to-runes, are
; actually defined in the output-module because they are first used there.
; ---------------------------------------------------------------------------
; Section: Putting It All Together
(defun transfer-environments (state from-scratch-flg)
(pprogn
(fms "Computing and transferring the ACL2 environment to Paco..."
nil *standard-co* state nil)
(er-progn
(if from-scratch-flg
(er-progn
(assign paco::paco-w nil)
(assign acl2::paco-w nil))
(value nil))
(cond
((and (boundp-global 'acl2::paco-w state)
(equal (@ acl2::paco-w) (w state)))
(pprogn
(fms "~%Done.~%" nil *standard-co* state nil)
(value nil)))
(t (er-progn
(transfer-paco-w state) ; set paco::paco-w
(transfer-paco-ens state) ; set paco::paco-ens (and acl2::paco-ens)
(transfer-nume-to-rune-map state)
; set acl2::rune-to-nume-map
(assign acl2::paco-w (w state))
(pprogn
(fms "~%Done.~%" nil *standard-co* state nil)
(value nil))))))))
(defmacro paco::transfer-environments (&optional (from-scratch-flg 'nil))
; If this were defined as a function it would have to take state as an
; argument and since it is normally called from the PACO package we
; would have to write acl2::state which is inconvenient. Hence, we
; make it a macro.
`(transfer-environments state ,from-scratch-flg))
(defmacro paco::w nil
'(@ paco::paco-w))
(defmacro paco::ens nil
'(@ paco::paco-ens))
(defmacro paco::rune (nume)
`(nume-to-rune ,nume (@ nume-to-rune-map)))
; ---------------------------------------------------------------------------
; Section: Paco Hint Translation
(defun probable-paco-clause-id (lst)
(cond ((atom lst) (equal lst nil))
((and (integerp (car lst))
(< 0 (car lst)))
(probable-paco-clause-id (cdr lst)))
(t nil)))
(defun value-nil (key x ctx state)
(cond ((eq x nil) (value nil))
(t (er soft ctx "Paco ~x0 hints must end in nil." key))))
(defun translate-paco-use-hint (arg ctx wrld state)
(cond
((atom arg) (value-nil :USE arg ctx state))
((and (true-listp (car arg))
(eq (car (car arg)) :INSTANCE)
(< 0 (len (cdr (car arg))))
(symbolp (cadr (car arg))))
(let ((thm (formula (cadr (car arg)) nil wrld)))
(er-let*
((alist (translate-substitution (cddr (car arg)) ctx wrld state))
(thm (if thm
(value thm)
(er soft ctx "~x0 does not name a theorem." (cadr (car arg)))))
(rst (translate-paco-use-hint (cdr arg) ctx wrld state)))
(value (cons (sublis-var alist thm) rst)))))
(t (er soft ctx "Ill-formed :INSTANCE ~x0" (car arg)))))
(defun translate-paco-expand-hint (arg ctx wrld state)
(cond
((atom arg) (value-nil :EXPAND arg ctx state))
(t (er-let*
((term (translate (car arg) t t t ctx wrld state)))
(cond
((and (nvariablep term)
(not (fquotep term))
(or (flambda-applicationp term)
(body (ffn-symb term) t wrld)))
(er-let*
((rst (translate-paco-expand-hint (cdr arg) ctx wrld state)))
(value (cons term rst))))
(t (er soft ctx
"The term ~x0 is not expandable."
term)))))))
(defun runes-to-numes (lst ctx wrld state)
; This returns a list of numes, with no duplications, or else signals
; an error.
(cond
((atom lst) (value nil))
((runep (car lst) wrld)
(er-let* ((rst (runes-to-numes (cdr lst) ctx wrld state)))
(value (add-to-set-equal (fnume (car lst) wrld) rst))))
(t (er soft ctx "~x0 is not a rune." (car lst)))))
(defun translate-paco-in-theory-hint (arg ctx wrld state)
(cond
((and (true-listp arg)
(case (car arg)
(paco::e/d (and (equal (len arg) 3)
(true-listp (cadr arg))
(true-listp (caddr arg))))
(paco::enable t)
(paco::disable t)
(otherwise nil)))
(er-let* ((numes-to-enable
(runes-to-numes (case (car arg)
(paco::e/d (cadr arg))
(paco::enable (cdr arg))
(otherwise nil))
ctx wrld state))
(numes-to-disable
(runes-to-numes (case (car arg)
(paco::e/d (caddr arg))
(paco::enable nil)
(otherwise (cdr arg)))
ctx wrld state)))
(let ((currently-disabled-numes (paco::btree-contents (paco::ens))))
(value
(paco::make-btree
(union-equal
numes-to-disable
(set-difference-equal currently-disabled-numes
numes-to-enable)))))))
(t (er soft ctx "Ill-formed :IN-THEORY hint, ~x0." arg))))
(defun translate-paco-hands-off-hint (arg ctx wrld state)
(cond
((atom arg) (value-nil :HANDS-OFF arg ctx state))
((and (consp (car arg))
(eq (car (car arg)) 'lambda)
(consp (cdr (car arg)))
(true-listp (cadr (car arg))))
; At this point we know that the car of arg is of the form (lambda
; (...) . &) and we want to translate it. To do so, we create a term
; by applying it to a list of terms. Where do we get a list of the
; right number of terms? We use its own formals!
(er-let*
((term (translate (cons (car arg) (cadr (car arg)))
t t t ctx wrld state))
(rst (translate-paco-hands-off-hint (cdr arg) ctx wrld state)))
; Below we assume that if you give translate ((lambda ...) ...) and it
; does not cause an error, then it gives you back a function application.
(value (cons (ffn-symb term) rst))))
((and (symbolp (car arg))
(function-symbolp (car arg) wrld))
(er-let*
((rst (translate-paco-hands-off-hint (cdr arg) ctx wrld state)))
(value (cons (car arg) rst))))
(t (er soft ctx
"The object ~x0 is not a legal element of a :HANDS-OFF ~
hint."
(car arg)))))
(defun translate-paco-do-not-hint (arg ctx wrld state)
(declare (ignore wrld))
(cond ((and (true-listp arg)
(subsetp-equal arg paco::*waterfall*))
(value arg))
(t (er soft ctx "The :DO-NOT hint requires a subset of ~x0."
paco::*waterfall*))))
(defun translate-paco-cases-hint (arg ctx wrld state)
(cond
((atom arg) (value-nil :CASES arg ctx state))
(t (er-let*
((term (translate (car arg) t t t ctx wrld state))
(rst (translate-paco-cases-hint (cdr arg) ctx wrld state)))
(value (cons term rst))))))
(defun translate-paco-by-hint (arg ctx wrld state)
(cond ((symbolp arg)
(let ((thm (formula arg t wrld)))
(cond (thm (value thm))
(t (er soft ctx "~x0 does not name a theorem." arg)))))
(t (er soft ctx "~x0 does not name a theorem." arg))))
(defun translate-paco-induct-hint (arg ctx wrld state)
(er-let* ((term (translate arg t t t ctx wrld state)))
(value (cond ((equal term *nil*) :DO-NOT-INDUCT)
((equal term *t*) t)
(t term)))))
(defun translate-paco-x-hint (key val ctx wrld state)
(case key
(:USE (translate-paco-use-hint val ctx wrld state))
(:EXPAND (translate-paco-expand-hint val ctx wrld state))
(:IN-THEORY (translate-paco-in-theory-hint val ctx wrld state))
(:HANDS-OFF (translate-paco-hands-off-hint val ctx wrld state))
(:DO-NOT (translate-paco-do-not-hint val ctx wrld state))
(:CASES (translate-paco-cases-hint val ctx wrld state))
(:BY (translate-paco-by-hint val ctx wrld state))
(:INDUCT (translate-paco-induct-hint val ctx wrld state))
(otherwise (er soft ctx "Paco does not support ~x0 hints." key))))
(defun translate-paco-hint-key-val-lst (lst ctx wrld state)
(cond ((endp lst) (value nil))
((keywordp (car lst))
(er-let*
((tval (translate-paco-x-hint (car lst) (cadr lst) ctx wrld state))
(rst (translate-paco-hint-key-val-lst (cddr lst) ctx wrld state)))
(value (cons (cons (car lst) tval) rst))))
(t (er soft ctx
"Every other element of a hint specification must ~
be a keyword and ~x0 is not."
(car lst)))))
(defun translate-paco-hint (x ctx wrld state)
(cond
((and (true-listp x)
(probable-paco-clause-id (car x))
(evenp (len (cdr x))))
(er-let* ((alist (translate-paco-hint-key-val-lst (cdr x) ctx wrld state)))
(value (cons (car x) alist))))
(t (er soft ctx
"Each Paco hint must be of the form (id :key1 val1 ... :keyn valn),~
where id is a true-list of natural numbers. ~x0 is not of this ~
form."
x))))
(defun translate-paco-hints (x ctx wrld state)
; key input output
; :USE (... (:INSTANCE name . subst) ...) => (... thm ...)
; :EXPAND (... term ...) => (... term ...)
; :IN-THEORY (E/D (... rune ...) (... rune ...)) => ens
; :IN-THEORY (ENABLE ... rune ...) => ens
; :IN-THEORY (DISABLE ... rune ...) => ens
; :HANDS-OFF (... fn ...) => (... fn ...)
; :DO-NOT (... process ...) => (... process ...)
; :CASES (... term ...) => (... term ...)
; :BY name => thm
; :INDUCT term => t, :DO-NOT-INDUCT, or term
; where (output) thm and term are in translated form, ens is a Paco
; btree, fn is a function symbol or lambda expression, and process is
; member of the *waterfall*.
(cond ((atom x)
(if (null x)
(value nil)
(er soft ctx ":HINTS must be a true-list.")))
(t (er-let* ((pair (translate-paco-hint (car x) ctx wrld state))
(rst (translate-paco-hints (cdr x) ctx wrld state)))
(value (cons pair rst))))))
; ---------------------------------------------------------------------------
; Section: The Paco DEFTHM
(defun paco::defthm-fn (name term waterfall-depth state hints rule-classes)
(er-progn
(transfer-environments state nil)
(er-let* ((tterm (translate term t t t 'paco::defthm (w state) state))
(hint-settings
(translate-paco-hints hints (cons 'dthm name)
(w state) state)))
(let* ((p (paco::prove tterm (paco::ens) (paco::w)
hint-settings
waterfall-depth)))
(cond ((eq (paco::describe-proof-attempt p 0) :QED)
(er-progn (assign last-proof-attempt p)
(defaxiom-fn name tterm state
rule-classes
nil
`(paco::dthm ,name ,term
,@(if hint-settings
(list :hint-settings
hint-settings)
nil)
,@(if (and rule-classes
(not (equal rule-classes
'(:REWRITE))))
(list :rule-classes
rule-classes)
nil)))
(value :QED)))
(t (er-progn
(assign last-proof-attempt p)
(pprogn
(fms *proof-failure-string* nil *standard-co* state nil)
(mv t nil state)))))))))
(defmacro paco::dthm (name term &key
(hints 'nil)
(depth '10)
(rule-classes '(:REWRITE)))
`(paco::defthm-fn ',name ',term ,depth state
,(list 'quote hints)
,(list 'quote rule-classes)))
(defmacro paco::show-proof (d-level)
`(paco::describe-proof-attempt (@ last-proof-attempt) ,d-level))
; We define prf in the ACL2 package so that we can type :prf 1 after
; a Paco proof attempt and see the proof.
(defmacro prf (d-level)
`(paco::show-proof ,d-level))
; And we define paco::prf so we can write (prf 1).
(defmacro paco::prf (d-level)
`(paco::show-proof ,d-level))
|