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
|
(in-package "ACL2")
#|
We introduce functions for loading the datastructure of the
stobj-based pathfinding program and prove that the loading and
calculating of this optimized program works just like the original
implementation.
See also "fpst.lisp".
Matt Wilding and David Greve
Updated February, 2003
|#
;; Assumes the "find path - stobj" program is loaded
(include-book "fpst")
(set-verify-guards-eagerness 2)
;; We want to prove the guards of the functions we need from J's
;; proof. However, some of the functions are not guard-provable
;; because the guards weren't in place. In this case, we add a
;; function with the same body and the needed guards, and use that
;; instead.
(defun myall-nodes (g)
(declare (xargs :guard (alistp g)))
(cond ((endp g) nil)
(t (cons (car (car g))
(myall-nodes (cdr g))))))
(defthm myall-nodes-is-all-nodes
(equal
(myall-nodes g)
(all-nodes g)))
; mygraph1p is just like J's graph1p, except that the node and the
; children are in-range naturals.
(defun mygraph1p (g nodes)
(declare (xargs :guard (and (true-listp nodes) (alistp g))))
(cond ((endp g) t)
(t (and (consp (car g))
(true-listp (cdr (car g)))
(numberlistp (car g) (maxnode)) ; needed for stobj version
(subsetp (cdr (car g)) nodes)
(no-duplicatesp (cdr (car g)))
(mygraph1p (cdr g) nodes)))))
(defthm mygraph1p-is-graph1p
(implies
(mygraph1p g nodes)
(graph1p g nodes)))
(defun mygraphp (g)
(declare (xargs :guard (alistp g)))
(and (alistp g)
(eqlable-listp (myall-nodes g))
(no-duplicatesp (myall-nodes g))
(mygraph1p g (myall-nodes g))))
(defthm mygraphp-is-graphp
(implies
(mygraphp g)
(graphp g)))
(defun myneighbors (node g)
(declare (xargs :guard (alistp g)))
(cond ((endp g) nil)
((equal node (car (car g)))
(cdr (car g)))
(t (myneighbors node (cdr g)))))
(defthm myneighbors-is-neighbors
(equal
(myneighbors n g)
(neighbors n g)))
(defthm consp-neighbors
(implies
(mygraph1p g l)
(iff
(consp (neighbors i g))
(neighbors i g))))
(defthm gp-update-nth
(implies
(and
(gp g)
(listp v)
(bounded-natp i max))
(gp (update-nth i v g)))
:hints (("goal" :in-theory (enable update-nth))))
;; Now, some functions that allow us to load the stobj with
;; values in the datastructures used by J's implementation.
(defun load-graph1 (g i st)
(declare (xargs :stobjs st
:guard (and (stp st) (alistp g)
(bounded-natp i (1+ (maxnode)))
(mygraphp g))
:measure (max 0 (nfix (- (maxnode) i)))))
(if (or (not (integerp i)) (not (< i (maxnode))))
st
(let
((st (update-gi i (myneighbors i g) st)))
(load-graph1 g (1+ i) st))))
(defun load-graph (g st)
(declare (xargs :stobjs st
:guard (and (stp st) (alistp g) (mygraphp g))))
(load-graph1 g 0 st))
(defun init-marks1 (i st)
(declare (xargs :stobjs st
:guard (and (stp st) (bounded-natp i (1+ (maxnode))))
:measure (max 0 (nfix (- (maxnode) i)))))
(if (or (not (integerp i)) (not (< i (maxnode))))
st
(let
((st (update-marksi i 0 st)))
(init-marks1 (1+ i) st))))
;; Some rules about our loading functions
(defthm stp-load-graph1
(implies
(and
(stp st)
(mygraphp g)
(integerp i)
(<= 0 i))
(stp (load-graph1 g i st))))
(defthm stp-init-marks1
(implies
(and
(stp st)
(integerp i)
(<= 0 i))
(stp (init-marks1 i st))))
(defthm nth-init-marks1-other
(implies
(not (equal i (marksindex)))
(equal (nth i (init-marks1 j st))
(nth i st))))
(defun init-marks (st)
(declare (xargs :stobjs st :guard (stp st)))
(init-marks1 0 st))
(defun load-st (g st)
(declare (xargs :stobjs st
:guard (and (stp st) (alistp g) (mygraphp g))))
(let ((st (load-graph g st)))
(let ((st (init-marks st)))
(let ((st (update-status 1 st)))
(let ((st (update-stack nil st)))
st)))))
(defthm graph-equivp-only-on-g
(implies
(equal (nth (gindex) st1) (nth (gindex) st2))
(equal
(graph-equivp1 g st1 i)
(graph-equivp1 g st2 i)))
:rule-classes nil)
(defthm graph-equivp1-init-marks
(equal
(graph-equivp1 g (init-marks1 i st) i)
(graph-equivp1 g st i))
:hints (("goal" :use (:instance graph-equivp-only-on-g
(st1 (init-marks1 i st)) (st2 st)))))
(defthm graphp1-equal-graphs
(implies
(and
(equal (nth (gindex) st1) (nth (gindex) st2))
(stp st1)
(stp st2))
(iff
(graphp1-st st1 i)
(graphp1-st st2 i)))
:hints (("goal" :in-theory (enable graphp1-st)))
:rule-classes nil)
(defthm graphp1-st-lesser
(implies
(and
(graphp1-st st i)
(bounded-natp i j))
(graphp1-st st j))
:hints (("goal" :in-theory (enable graphp1-st))))
(defthm graphp1-st-init-marks
(implies
(and
(stp st)
(integerp i)
(<= 0 i))
(equal
(graphp1-st (init-marks1 i st) i)
(graphp1-st st i)))
:hints (("goal" :use (:instance graphp1-equal-graphs
(st1 (init-marks1 i st)) (st2 st))
:in-theory (disable stp))))
(defthm nth-0-load-graph1-above
(implies
(and
(bounded-natp i j)
(integerp j))
(equal
(nth i (nth (gindex) (load-graph1 g j st)))
(nth i (nth (gindex) st)))))
(defthm graph-equivp1-load-graph1
(implies
(bounded-natp i (maxnode))
(graph-equivp1 g (load-graph1 g i st) i))
:hints (("Subgoal *1/3'" :expand (:free (x) (LOAD-GRAPH1 G x ST)))
("Subgoal *1/3''" :expand (:free (x n) (GRAPH-EQUIVP1 G x n)))))
(defthm nth-1-init-marks-above
(implies
(and
(bounded-natp i j)
(integerp j))
(equal
(nth i (nth (marksindex) (init-marks1 j st)))
(nth i (nth (marksindex) st)))))
(defthm mark-equivp1-init-marks1
(implies
(and
(integerp i)
(<= 0 i))
(mark-equivp1 nil (init-marks1 i st) i)))
(defthm equiv-load-st
(equiv nil g nil (load-st g st))
:hints (("goal" :in-theory (set-difference-theories
(enable graph-equivp mark-equivp)
'(graph-equivp1 mark-equivp1 init-marks1)))))
(defthm numberlistp-neighbors
(implies
(mygraph1p g n)
(numberlistp (neighbors i g) (maxnode))))
(defthm graph1p-st-load-graph1
(implies
(and
(mygraphp g)
(integerp i)
(<= 0 i))
(graphp1-st (load-graph1 g i st) i))
:hints (("goal" :in-theory (enable load-graph1 graphp1-st))))
(defun linear-find-st (a b g st)
(declare (xargs :stobjs st
:guard (and (stp st)
(bounded-natp a (maxnode))
(bounded-natp b (maxnode))
(alistp g)
(mygraphp g))))
(let ((st (load-st g st)))
(let ((st (linear-find-next-step-st-mbe (list a) b st)))
(if (not (equal (status st) 0))
(mv 'failure st)
(mv (stack st) st)))))
(defthm nth-init-marks1
(implies
(and
(integerp j)
(integerp i)
(<= 0 i)
(<= j i)
(< i (maxnode)))
(equal (nth i (nth (marksindex) (init-marks1 j st)))
0))
:hints (("goal" :expand ((INIT-MARKS1 I ST)))))
(defthm linear-find-next-step-st-mbe-base
(implies
(and
(equal (nth i (nth (marksindex) st)) 0)
(bounded-natp i (maxnode))
(equal (stack st) nil))
(equal (nth (stackindex) (linear-find-next-step-st-mbe (list i) i st)) (list i)))
:hints (("goal" :expand (linear-find-next-step-st-mbe (list i) i st))))
(defthm nth-load-graph1
(implies
(and
(integerp i)
(<= 0 i)
(not (equal i (gindex))))
(equal (nth i (load-graph1 g j st))
(nth i st))))
;; ****************
;; Main lemma
;; ****************
;; Our implementation returns the same value as the original
;; list-based one when we load the stobj using the functions
;; of this file.
(defthm linear-find-st-linear-find-path
(implies
(and
(bounded-natp a (maxnode))
(bounded-natp b (maxnode))
(mygraphp g)
(stp st))
(equal
(car (linear-find-st a b g st))
(linear-find-path a b g)))
:hints (("goal" :use ((:instance implementations-same (stack nil) (mt nil) (c (list a))
(st (load-st g st)))
equiv-load-st)
:in-theory (disable linear-find-path-is-find-path equiv-load-st stp))))
#|
ACL2 !>(assign g '((0 1 2) (1) (2 0 1 2 3) (3 1)))
((0 1 2) (1) (2 0 1 2 3) (3 1))
ACL2 !>(mygraphp (@ g))
T
ACL2 !>(stp st)
T
ACL2 !>(linear-find-st 0 3 (@ g) st)
((0 2 3) <st>)
ACL2 !>(linear-find-path 0 3 (@ g))
(0 2 3)
|#
;; Some functions for building graphs
;; Generate a graph with nodes numbered curr though last and edges
;; from each node to the list all
(defun completeg-helper (curr last all)
(declare (xargs :verify-guards t))
(declare (xargs :measure (nfix (- (1+ (nfix last)) (nfix curr)))))
(if (<= (nfix curr) (nfix last))
(cons
(cons curr all)
(completeg-helper (1+ (nfix curr)) last all))
nil))
;; Generate a list of naturals from curr to last
(defun listofnats (curr last)
(declare (xargs :verify-guards t
:measure (nfix (- (1+ (nfix last)) (nfix curr)))))
(if (<= (nfix curr) (nfix last))
(cons
curr
(listofnats (1+ (nfix curr)) last))
nil))
;; Generate a complete graph with nodes 0 to size-1
(defun completeg (size)
(declare (xargs :verify-guards t))
(completeg-helper 0 (1- (nfix size)) (listofnats 0 (1- (nfix size)))))
;; Generate a "bad" graph with n nodes. Nodes 0..n-1 are a complete
;; graph, and node n is disconnected
(defun badg (size)
(declare (xargs :verify-guards t))
(cons
(list size)
(completeg size)))
#|
The pathfinder can be run from the ACL2 read-eval-print loop.
First, load this book
ACL2 !>(include-book
"run-fpst")
Loading /accts/dagreve/local/src/acl2-2.8a/books/arithmetic/equalities.o
start address -T 1827ecc Finished loading /accts/dagreve/local/src/acl2-2.8a/books/arithmetic/equalities.o
Loading /accts/dagreve/local/src/acl2-2.8a/books/arithmetic/rational-listp.o
...
Summary
Form: ( INCLUDE-BOOK ; manual editing by Matt K. to avoid Makefile-deps dependency
"run-fpst" ...)
Rules: NIL
Warnings: None
Time: 1.49 seconds (prove: 0.00, print: 0.00, other: 1.49)
"/accts/dagreve/ACL/challenges/graph/run-fpst.lisp"
Next, load the datastructure with a graph. In this example, we load a
graph with 1,000 nodes and 1,000,000 edges, which takes about a minute.
ACL2 !>(load-st (badg 1000) st)
<st>
We try to find a non-existent path so that the program traverses all
the edges. This requires less than a second.
ACL2 !>(linear-find-next-step-st-mbe (list 0) 1000 st)
<st>
ACL2 !>(status st)
1
We reset the marks array and search for an existing path. Note that
it's not guaranteed to be the shortest one, only a valid one.
ACL2 !>(init-marks st)
<st>
ACL2 !>(linear-find-next-step-st-mbe (list 0) 6 st)
<st>
ACL2 !>(status st)
0
ACL2 !>(stack st)
(0 1 2 3 4 5 6)
|#
|