Edinburgh IMP77 Compiler - Version 8.4 1 %begin ;! force an extra scope since top-level routines like printsymbol were clashing with perms version. 2 3 %reals %long 4 5 ! This program shows explicitly the results of write(m,n) for each significant 6 ! size of m and n, and tests the extreme cases of MIN and MAX INT. 7 8 ! First, we replace the standard output calls with local versions which 9 ! keep track of 'outpos', to help with formatting output into columns: 10 11 ! In order to do this for the in-built 'write' etc, we have reimplemented local copies 12 ! of the current library versions which have been tested and confirmed to be identical. 13 14 ! The output table is rather wide and may need to be viewed using a small font. 15 16 %own %integer outpos = 0 17 18 %routine intrinsic printsymbol(%integer ch) 19 printsymbol(ch) 20 %end 21 22 %routine printsymbol(%integer ch) 23 intrinsic printsymbol(ch) 24 %if ch = 'J'&31 %or ch = 'M'&31 %or ch = 'L'&31 %start ;! nl cr ff 25 outpos = 0 26 %else %if ch = 'H'&31 ;! bs 27 outpos = outpos - 1 %if outpos > 0 28 %else %if ch = 'I'&31 ;! tab 29 outpos = (outpos + 8) & (\7) ;! at least on linux 30 %else 31 outpos = outpos + 1 32 %finish 33 %end 34 35 %routine newline 36 printsymbol(nl) 37 %end 38 39 %routine newlines(%integer n) 40 %integer i 41 %return %if n <= 0 42 newline %for i = 1, 1, n 43 %end 44 45 %routine space 46 printsymbol(' ') 47 %end 48 49 %routine spaces(%integer n) 50 %integer i 51 %return %if n <= 0 52 space %for i = 1, 1, n 53 %end 54 55 %routine printstring(%string (255) s) 56 %integer i 57 printsymbol(charno(s, i)) %for i = 1, 1, length(s) 58 %end 59 60 %routine write(%integer m, n) 61 printstring("Please don't ever use plain 'write' in this program. Use write68k for now."); newline 62 printstring("The implementation in the current Imp77 is severely broken w.r.t. formatting."); newline 63 %stop 64 %end ?N unused ?M unused 65 66 !%begin ; ! Sorry, all original comments {in brackets} were lost in an unfortunate reformatting. 67 68 69 ! ----------------------------------------------------------------------------------------- 70 71 %routine write3l(%integer n, places);! Same as current Imp library 72 73 ! Write an integer on the current output stream, to 74 ! occupy PLACES character spaces on the page. 75 ! Nominally based on PrintString(I to S(n, places)) 76 ! This version doesn't use strings though, and is 77 ! therefore smaller and quicker. It builds the 78 ! characters "backwards" in an array 79 80 %routine decimal(%integer n, %integer %name places, digits, vec) 81 %integer p, sign, d, x 82 vec = vec + 15 83 sign = '-' 84 %if n >= 0 %start 85 n = -n 86 sign = ' ' 87 sign = 0 %if places <= 0 88 %finish 89 %if places <= 0 %then places = -places %else places = places + 1 90 p = vec 91 %cycle 92 x = n // 10 93 d = n - x * 10 94 n = x 95 vec = vec - 1 96 byteinteger(vec) = '0' - d 97 %repeat %until n = 0 98 %if sign # 0 %start 99 vec = vec - 1 100 byteinteger(vec) = sign 101 %finish 102 digits = p - vec 103 %end 104 %string (32) ch 105 ! Enough room for a 32 bit integer, plus sign 106 %integer new, digit, sign, i 107 length(ch) = 0 108 ! First collect the digits 109 %if n = 0 %start 110 length(ch) = length(ch) + 1 111 charno(ch, length(ch)) = '0' 112 sign = 0 113 %else 114 sign = 1 115 ! we make everything negative to avoid the -ve MaxInt wrap problem 116 %if n > 0 %start 117 sign = 0 118 n = -n 119 %finish 120 %while n # 0 %cycle 121 new = n // 10 122 digit = (new * 10) - n 123 length(ch) = length(ch) + 1 124 charno(ch, length(ch)) = digit + '0' 125 n = new 126 %repeat 127 %finish 128 ! Now deal with the sign, taking into account the slightly 129 ! idiosyncratic behaviour on positive numbers 130 %if sign # 0 %start 131 length(ch) = length(ch) + 1 132 charno(ch, length(ch)) = '-' 133 %else 134 %if places > 0 %start 135 length(ch) = length(ch) + 1 136 charno(ch, length(ch)) = ' ' 137 %finish 138 %finish 139 ! Now we adjust Places, also slightly mysteriously 140 %if places <= 0 %then places = -places %else places = places + 1 141 ! The array at CH now contains the characters (backwards) 142 ! and the character count is in NEXT 143 %while places > length(ch) %cycle 144 length(ch) = length(ch) + 1 145 charno(ch, length(ch)) = ' ' 146 places = places - 1 147 %repeat 148 %for i = length(ch), -1, 1 %cycle 149 printsymbol(charno(ch, i)) 150 %repeat 151 %end ?DECIMAL unused 152 153 ! ----------------------------------------------------------------------------------------- 154 155 %routine vax write(%integer n, p) 156 %string(63)%fn itos(%integer n, p) 157 %string(255) answer 158 %byteintegerarray a(0:64) 159 %integer sign, sym, pt, val 160 val = p 161 %if p > 0 %then p = p+1 %else p = -p 162 p = 63 %if p > 63 163 sign = ' ' 164 sign = 0 %if val <= 0 165 pt = 0 166 %if n < 0 %start 167 sign = '-' 168 %if n = 16_80000000 %start 169 string(addr(a(pt))) = "8463847412" 170 pt = pt + 10 171 -> set 172 %finish 173 n = -n 174 %finish 175 %cycle 176 sym = n-n//10*10 177 pt = pt+1 178 a(pt) = sym+'0' 179 n = n//10 180 %repeat %until n = 0 181 set: 182 %if sign # 0 %start 183 pt = pt+1; a(pt) = sign 184 %finish 185 %while pt < p %cycle 186 pt = pt+1; a(pt) = ' ' 187 %repeat 188 answer = "" 189 %cycle 190 answer = answer.tostring(a(pt)) 191 pt = pt-1 192 %repeat %until pt = 0 193 %result = answer 194 %end 195 space %and p = p-1 %while p > 62 196 printstring(itos(n, p)) 197 %end 198 199 ! ----------------------------------------------------------------------------------------- 200 201 %routine write68k(%integer v, p) 202 %integer vv, q, pos 203 %byte %integer %array store(0 : 15) 204 vv = v 205 vv = -vv %if vv > 0 206 pos = 15 207 %while vv <= -10 %cycle 208 q = vv // 10 209 store(pos) = q * 10 - vv + '0' 210 pos = pos - 1 211 vv = q 212 %repeat 213 store(pos) = '0' - vv 214 %if p <= 0 %start 215 spaces(pos - 16 - p) %if p < 0 216 %finish %else %start 217 spaces(pos - 16 + p) 218 printsymbol(' ') %if v >= 0 219 %finish 220 printsymbol('-') %if v < 0 221 printsymbol(store(pos)) %and pos = pos + 1 %until pos = 16 222 %end 223 224 ! ----------------------------------------------------------------------------------------- 225 226 %routine emaswrite(%integer value, places) 227 !*********************************************************************** 228 !* SIMPLE MINDED ALL IMP VERSION NOT USING STRINGS * 229 !*********************************************************************** 230 %string (15) %fn swrite %alias "S#SWRITE"(%integer value, places) 231 !*********************************************************************** 232 !* SIMPLE MINDED ALL IMP VERSION * 233 !*********************************************************************** 234 %string (1) sign 235 %string (255{15}) res 236 %integer work, ptr 237 %string (1) %array ch(0 : 15) 238 res = "" 239 sign = " " 240 %if value = X'80000000' %then %start 241 res = "-2147483548" 242 !res = " " . res %for ptr = 1, 1, places - 10 ;! causes run-time string overflow 243 %if places <= 0 %start 244 places = -15 %if places < -15 245 res = " " . res %while length(res) < |places| 246 %else 247 places = 14 %if places > 14 248 res = " " . res %while length(res) < places 249 res = " " . res 250 %finish 251 %result = res 252 %finish 253 %if value = X'80000000' %then %result = "-2147483648" 254 %if value < 0 %then sign = "-" %and value = -value 255 ptr = 0 256 %cycle 257 work = value // 10 258 ch(ptr) = tostring(value - 10 * work + '0') 259 value = work 260 ptr = ptr + 1 261 %repeat %until value = 0 262 res = res . " " %for work = ptr, 1, places - 1 263 work = ptr - 1 264 res = res . sign 265 res = res . ch(ptr) %for ptr = work, -1, 0 266 %result = res 267 %end 268 %integer sign, work, ptr 269 %byte %integer %array ch(0 : 15) 270 sign = ' ' 271 %if value = X'80000000' %then printstring(swrite(value, places)) %and %return 272 %if value < 0 %then sign = '-' %and value = -value 273 ptr = 0 274 %cycle 275 work = value // 10 276 ch(ptr) = value - 10 * work 277 value = work 278 ptr = ptr + 1 279 %repeat %until value = 0 280 %if places > ptr %then spaces(places - ptr) 281 work = ptr - 1 282 printsymbol(sign) 283 printsymbol(ch(ptr) + '0') %for ptr = work, -1, 0 284 %end 285 286 ! ----------------------------------------------------------------------------------------- 287 288 %routine emasi77write(%integer num, pl) 289 %string (63) %function itos(%integer n, places) 290 ! 291 ! ITOS0 292 ! 293 ! Returns a string representing the argument as 294 ! a decimal number, without any space justification 295 ! or padding. 296 ! 297 %const %integer maxdigits = 255 - 64 298 %const %integer maxint = 16_7FFFFFFF 299 %const %integer minint = 16_80000000 300 %const %integer minintlastdigit = '8' 301 %string (maxdigits) %function itos0(%integer n) 302 %string (maxdigits) res 303 %integer s, p, e 304 s = 0 305 %if n < 0 %start 306 s = -1 307 ! mark -ve 308 %if n = minint %start 309 ! special case 310 n = maxint 311 ! rely on "max int = (-min int) - 1" 312 s = 1 313 ! mark special 314 %finish %else %start 315 n = -n 316 %finish 317 %finish 318 p = addr(res) + maxdigits 319 e = p 320 ! end point 321 byteinteger(p) = rem(n, 10) + '0' %and n = n // 10 %and p = p - 1 %until n = 0 322 %if s # 0 %start 323 ! negative 324 byteinteger(p) = '-' 325 p = p - 1 326 %if s > 0 %start 327 ! min int 328 charno(res, maxdigits) = minintlastdigit 329 %finish 330 %finish 331 byteinteger(p) = e - p 332 ! length byte 333 %result = string(p) 334 %end 335 %string (63 + maxdigits) res 336 %integer p, slen 337 p = addr(res) + 62 338 ! digit string right justified 339 string(p) = itos0(n) 340 slen = byteinteger(p) 341 %if n > 0 %and places > 0 %start 342 byteinteger(p) = ' ' 343 p = p - 1 344 slen = slen + 1 345 %finish 346 %if places <= 0 %start 347 places = -places 348 %finish %else %start 349 places = places + 1 350 %finish 351 places = 63 %if places > 63 352 %while slen < places %cycle 353 byteinteger(p) = ' ' 354 p = p - 1 355 slen = slen + 1 356 %repeat 357 byteinteger(p) = slen 358 %result = string(p) 359 %end 360 %string (63) res 361 res = itos(num, pl) 362 pl = \pl %if pl < 0 363 spaces(pl - 62) 364 printstring(res) 365 %end 366 367 ! ----------------------------------------------------------------------------------------- 368 369 %routine mouseswrite(%integer n, p) 370 %string (72) %fn itos(%integer n, p) 371 %byte %array s(0 : 72) 372 %integer x, sign 373 x = 73 374 sign = 0 375 %if p > 0 %start 376 sign = ' ' 377 p = 72 - p 378 %else 379 p = p + 73 380 %finish 381 p = 1 %if p <= 0 382 %if n < 0 %start 383 sign = '-' 384 n = -n 385 %if n < 0 %start 386 !max int 387 n = n // (-10) 388 x = x - 1 389 s(x) = '8' 390 %finish 391 %finish 392 %cycle 393 x = x - 1 394 s(x) = rem(n, 10) + '0' 395 n = n // 10 396 %repeat %until n = 0 397 %if sign # 0 %start 398 x = x - 1 399 s(x) = sign 400 %finish 401 %while x > p %cycle 402 x = x - 1 403 s(x) = ' ' 404 %repeat 405 x = x - 1 406 s(x) = 72 - x 407 %result = string(addr(s(x))) 408 %end 409 printstring(itos(n, p)) 410 %end 411 412 ! ----------------------------------------------------------------------------------------- 413 414 %ROUTINE DEC10 WRITE(%INTEGER N, S) 415 %ROUTINESPEC P(%INTEGER N) 416 %INTEGER SIGN 417 %IF S<=0 %THEN SIGN=S %ELSE SIGN = ' '; !GIVE NO SIGN FOR WRITE(N,0) 418 S=|S|; S=63 %IF S>63 419 %IF N < 0 %START ;! doesn't use negative trick but could be easily modified to. 420 N = -N 421 SIGN = '-' 422 %FINISH 423 P(N) 424 %ROUTINE P(%INTEGER N) 425 S = S-1 426 P(N//10) %IF N >= 10 427 %IF SIGN > 0 %START 428 SPACES(S-1) 429 PRINTSYMBOL(SIGN); SIGN = 0 430 %FINISH %ELSE %IF SIGN<0 %THEN SPACES(S) %AND SIGN=0; !FOR -VE S 431 PRINTSYMBOL(REM(N,10)+'0') 432 %END 433 %END 434 435 ! ----------------------------------------------------------------------------------------- 436 437 ! ######################################################################################### 438 439 %ROUTINESPEC EMAS PRINTFL (%LONGREAL X, %INTEGER N) 440 441 %ROUTINE EMAS PRINT(%LONGREAL X, %INTEGER N,M) 442 !*********************************************************************** 443 !* PRINTS A REAL NUMBER (X) ALLOWING N PLACES BEFORE THE DECIMAL * 444 !* POINT AND M PLACES AFTER.IT REQUIRES (M+N+2) PRINT PLACES * 445 !* UNLESS (M=0) WHEN (N+1) PLACES ARE REQUIRED. * 446 !* * 447 !* A LITTLE CARE IS NEEDED TO AVOID UNNECESSARY LOSS OF ACCURACY * 448 !* AND TO AVOID OVERFLOW WHEN DEALING WITH VERY LARGE NUMBERS * 449 !*********************************************************************** 450 %CONSTLONGREAL DZ=0 451 %LONGREAL Y,Z,ROUND,FACTOR 452 %INTEGER I,J,L,MORIG 453 %BYTEINTEGER SIGN 454 M=M&63; ! DEAL WITH STUPID PARAMS 455 MORIG=M 456 %IF N<0 %THEN N=1; N=N&31; ! DEAL WITH STUPID PARAMS 457 X=X+DZ; ! NORMALISE 458 SIGN=' '; ! '+' IMPLIED 459 %IF X<0 %THEN SIGN='-' 460 Y=|X|; ! ALL WORK DONE WITH Y 461 ROUND=0.5/10^M; ! ROUNDING FACTOR 462 %IF Y>1.0*10^16 %OR N=0 %THENSTART; ! MEANINGLESS FIGURES GENERATED 463 %IF N>M %THEN M=N; ! FOR FIXED POINT PRINTING 464 EMAS PRINTFL(X,M); ! OF ENORMOUS NUMBERS 465 %RETURN; ! SO PRINT IN FLOATING FORM 466 %FINISH 467 I=0; Z=1; Y=Y+ROUND 468 %CYCLE; ! COUNT LEADING PLACES 469 I=I+1; Z=10*Z; ! NO DANGER OF OVERFLOW HERE 470 %REPEAT %UNTIL Z>Y 471 PRINTSYMBOL(SIGN) 472 J=I-1; Z=10^J 473 FACTOR=1/10 474 %CYCLE 475 %CYCLE 476 L=INT PT(Y/Z); ! OBTAIN NEXT DIGIT 477 Y=Y-L*Z; Z=Z*FACTOR; ! AND REDUCE TOTAL 478 PRINTSYMBOL(L+'0') 479 J=J-1 480 %REPEAT %UNTIL J<0 481 %IF M=0 %THEN %EXIT; ! NO DECIMAL PART TO BE O/P 482 PRINTSTRING(".") 483 J=M-1; Z=10^(J-1); M=0 484 Y=10*Y*Z 485 %REPEAT 486 %if MORIG>0 %start { Chop any redundant trailing 0s} 487 !opline_length=opline_length-1 %while %c 488 ! opline_l(opline_length-1)='0' %and opline_l(opline_length-2)#'.' 489 %finish 490 %END; ! OF ROUTINE PRINT 491 492 %ROUTINE EMAS PRINTFL (%LONGREAL X, %INTEGER N) 493 !*********************************************************************** 494 !* PRINTS IN FLOATING POINT FORMAT WITH N PLACES AFTER THE * 495 !* DECIMAL POINT. ALWAYS TAKES N+7 PRINTING POSITIONS. * 496 !* CARE REQUIRED TO AVOID OVERFLOW WITH LARGE X * 497 !*********************************************************************** 498 %LONGREAL SIGN,ROUND,FACTOR,LB,UB 499 %CONSTLONGREAL DZ=0 500 %INTEGER COUNT,INC 501 inc=integer(addr(x)) 502 %if inc>>20&x'7ff'=x'7ff' %start 503 printstring("NAN"); 504 !outhex(inc); 505 !outhex(integer(addr(x)+4)) 506 !outsym('}'); 507 %return 508 %finish 509 ROUND=0.5/10^N; ! TO ROUND SCALED NO 510 LB=1-ROUND; UB=10-ROUND 511 SIGN=1 512 X=X+DZ; ! NORMALISE 513 %IF X=0 %THEN COUNT=0 %ELSESTART 514 %IF X<0 %THEN X=-X %AND SIGN=-SIGN 515 INC=1; COUNT=0 516 FACTOR=1/10 517 %IF X<=1 %THEN FACTOR=10 %AND INC=-1 518 ! FORCE INTO RANGE 1->10 519 %WHILE X=UB %CYCLE 520 X=X*FACTOR; COUNT=COUNT+INC 521 %REPEAT 522 %FINISH 523 EMAS PRINT(SIGN*X,1,N) 524 %if count#0 %start 525 PRINTSTRING("@") ;! original had 'E' - was that correct? 526 WRITE68K(COUNT,0) 527 %finish 528 %END; ! OF ROUTINE PRINTFL 529 530 ! ----------------------------------------------------------------------------------------- 531 %ROUTINESPEC ERCC PRINTFL(%LONGREAL X, %INTEGER N) 532 %ROUTINE ERCC PRINT(%LONGREAL X, %INTEGER N,M) 533 !*********************************************************************** 534 !* PRINTS A REAL NUMBER (X) ALLOWING N PLACES BEFORE THE DECIMAL * 535 !* POINT AND M PLACES AFTER.IT REQUIRES (M+N+2) PRINT PLACES * 536 !* UNLESS (M=0) WHEN (N+1) PLACES ARE REQUIRED. * 537 !* * 538 !* A LITTLE CARE IS NEEDED TO AVOID UNNECESSARY LOSS OF ACCURACY * 539 !* AND TO AVOID OVERFLOW WHEN DEALING WITH VERY LARGE NUMBERS * 540 !*********************************************************************** 541 %CONSTLONGREAL PMAX= 1@16 542 %CONSTLONGREAL DZ=0 543 %LONGREAL Y,Z,ROUND,FACTOR 544 %INTEGER I,J,L 545 %BYTEINTEGER SIGN 546 M=M&63; ! DEAL WITH STUPID PARAMS 547 %IF N<0 %THEN N=1; N=N&31; ! DEAL WITH STUPID PARAMS 548 X=X+DZ; ! NORMALISE 549 SIGN=' '; ! '+' IMPLIED 550 %IF X<0 %THEN SIGN='-' 551 Y=|X|; ! ALL WORK DONE WITH Y 552 ROUND=0.5/10^M; ! ROUNDING FACTOR 553 %IF Y>PMAX %OR N=0 %THENSTART; ! MEANINGLESS FIGURES GENERATED 554 %IF N>M %THEN M=N; ! FOR FIXED POINT PRINTING 555 ERCC PRINTFL(X,M); ! OF ENORMOUS NUMBERS 556 %RETURN; ! SO PRINT IN FLOATING FORM 557 %FINISH 558 I=0; Z=1; Y=Y+ROUND 559 %CYCLE; ! COUNT LEADING PLACES 560 I=I+1; Z=10*Z; ! NO DANGER OF OVERFLOW HERE 561 %REPEAT %UNTIL Z>Y 562 SPACES(N-I); ! O.K FOR ZERO OR -VE SPACES 563 PRINT SYMBOL(SIGN) 564 J=I-1; Z=10^J 565 FACTOR=1/10 566 %CYCLE 567 %CYCLE 568 L=INT PT(Y/Z); ! OBTAIN NEXT DIGIT 569 Y=Y-L*Z; Z=Z*FACTOR; ! AND REDUCE TOTAL 570 PRINT SYMBOL(L+'0') 571 J=J-1 572 %REPEAT %UNTIL J<0 573 %IF M=0 %THENRETURN; ! NO DECIMAL PART TO BE O/P 574 PRINTSTRING(".") 575 J=M-1; Z=10^(J-1); M=0 576 Y=10*Y*Z 577 %REPEAT 578 %END; ! OF ROUTINE PRINT 579 580 %ROUTINE ERCC PRINTFL(%LONGREAL X, %INTEGER N) 581 !*********************************************************************** 582 !* PRINTS IN FLOATING POINT FORMAT WITH N PLACES AFTER THE * 583 !* DECIMAL POINT. ALWAYS TAKES N+7 PRINTING POSITIONS. * 584 !* CARE REQUIRED TO AVOID OVERFLOW WITH LARGE X * 585 !*********************************************************************** 586 %CONSTLONGREAL DZ=0 587 %LONGREAL SIGN,ROUND,FACTOR,LB,UB 588 %INTEGER COUNT,INC 589 ROUND=0.5/10^N; ! TO ROUND SCALED NO 590 LB=1-ROUND; UB=10-ROUND 591 SIGN=1 592 X=X+DZ; ! NORMALISE 593 %IF X=0 %THEN COUNT=-99 %ELSESTART 594 %IF X<0 %THEN X=-X %AND SIGN=-SIGN 595 INC=1; COUNT=0 596 FACTOR=1/10 597 %IF X<=1 %THEN FACTOR=10 %AND INC=-1 598 ! FORCE INTO RANGE 1->10 599 %WHILE X=UB %CYCLE 600 X=X*FACTOR; COUNT=COUNT+INC 601 %REPEAT 602 %FINISH 603 ERCC PRINT(SIGN*X,1,N) 604 PRINTSTRING("@") 605 emasWRITE(COUNT,2) 606 %END; ! OF ROUTINE PRINTFL 607 ! ----------------------------------------------------------------------------------------- 608 609 %routine vax printfl(%long %real r, %integer places) 610 %string (63) %fn ftos(%long %real r, %integer p) 611 %routine vax fraction(%long %real r, %integer places, %string (*) %name s) 612 %long %real %fn fracpt(%long %real x) 613 %long %real z 614 z = x - intpt(x) 615 z = z + 1.0 %if z < 0 616 %result = z 617 %end 618 %return %if places <= 0 619 s = s . "." 620 %while places > 0 %cycle 621 places = places - 1 622 r = fracpt(r) * 10 623 s = s . tostring(intpt(r) + '0') 624 %repeat 625 %end 626 %routine round(%long %real %name v, %integer p) 627 %long %real x 628 %return %if v = 0 629 %if v < 0 %then x = -0.5 %else x = 0.5 630 %cycle 631 p = p - 1 632 %exit %if p < 0 633 x = x / 10 634 %repeat 635 v = v + x 636 %end 637 {XOWN}%string (63) s = "" 638 %integer exp, sym 639 !p = 7 %if p > 7 ;! WHY?????????????????????????? p = 7 %if p > 7 640 ! removing the p<=7 constraint makes this behave the same as EMAS PRINTFL 641 %if r < 0 %then sym = '-' %and r = -r %else sym = ' ' 642 %if r = 0 %start 643 exp = -99 644 %else 645 exp = 0 646 exp = exp + 1 %and r = r / 10 %while r >= 10 647 exp = exp - 1 %and r = r * 10 %while r < 1 648 %finish 649 round(r, p) 650 exp = exp + 1 %and r = r / 10 %if r >= 10 651 s = s . tostring(intpt(r) + '0') 652 vax fraction(r, p, s) 653 %result = tostring(sym) . s . "@" . itos(exp, 0) 654 %end 655 places = \places %if places < 0 656 spaces(places - 62) 657 printstring(ftos(r, places)) 658 %end 659 660 ! ----------------------------------------------------------------------------------------- 661 662 %long %real %fn dec10 fracpt(%long %real x) 663 %long %real z 664 z = x - intpt(x) 665 z = z + 1.0 %if z < 0 666 %result = x - intpt(x) 667 668 %end 669 670 %ROUTINE DEC10 FRACTION(%LONG %REAL R, %INTEGER PLACES) 671 %RETURN %IF PLACES <= 0 672 PRINTSYMBOL('.') 673 %WHILE PLACES>0 %CYCLE 674 R = dec10 FRACPT(R)*10.0 ; ! first digit after the point of a 0.xxxxxxx positive number. 675 PLACES = PLACES-1 676 PRINTSYMBOL(INTPT(R)+'0') 677 %REPEAT 678 %END 679 680 %ROUTINE DEC10 PRINTFL(%LONG %REAL R, %INTEGER P) 681 %INTEGER EXP,SIGN, pp 682 %real roundup 683 SIGN=' ' 684 SIGN='-' %AND R = -R %IF R < 0 685 PRINTSYMBOL(SIGN) 686 %IF R = 0 %START 687 EXP = 0 ;! Was EXP = 99 for completely unknown reasons. 688 %ELSE 689 ! Normalise: 690 EXP = 0 691 EXP = EXP + 1 %AND R = R*0.1 %WHILE R >= 10.0 692 EXP = EXP - 1 %AND R = R*10.0 %WHILE R < 1.0 693 ! R should now be a single digit followed by a decimal part, with the exponent adjusted appropriately 694 ! Round to P places: 695 roundup = 0.5 696 %IF P <= 0 %START 697 !R = R+0.5 ;! round to an int? 698 pp = 0 699 %ELSE 700 ! This was just not working so I've replaced it with the 'pp' code below. I suspect faulty exponentiation? 701 !R = R + 0.5*10\(-P) 702 !R = R + 0.5*10^(-P) 703 pp = p 704 %FINISH 705 %while pp > 0 %then roundup = roundup / 10 %and pp = pp - 1 706 R = R + roundup 707 %FINISH 708 PRINTSYMBOL(INT PT(R)+'0') 709 DEC10 FRACTION(R,P) 710 PRINTSYMBOL('@'); WRITE68K(EXP,1) 711 %END 712 713 %ROUTINE DEC10 PRINT(%LONG %REAL R, %INTEGER B,A) 714 %LONG %REAL RM 715 %CONSTINTEGER MAX INT=16_7FFFFFFF 716 %INTEGER SIGN 717 %ROUTINE P(%INTEGER N) 718 B = B-1 719 P(N//10) %IF N >= 10 720 %IF SIGN # 0 %START 721 SPACES(B-1) 722 PRINTSYMBOL(SIGN); SIGN = 0 723 %FINISH 724 PRINTSYMBOL(REM(N,10)+'0') 725 %END 726 RM=|R| 727 DEC10 PRINTFL(R, B+A) %AND %RETURN %IF RM> MAX INT 728 %IF A<=0 %START 729 RM=RM+0.5 730 %ELSE 731 !RM=RM+0.5*10\(-A) 732 RM=RM+0.5*10^(-A) 733 %FINISH 734 %IF B<=0 %THEN SIGN=0 %ELSE SIGN=' ' 735 %IF R<0 %THEN SIGN='-' 736 P(INT PT(RM)); !THUS ALLOWING FOR -0.1 ETC. 737 DEC10 FRACTION(RM,A) 738 %END 739 740 ! ----------------------------------------------------------------------------------------- 741 742 %routine print 68k(%long %real x, %integer n,m) 743 %constreal pmax = 2147483647.0 744 %long %real y,z,POW 745 %integer i=0,l,count=0,sign 746 sign = ' ' 747 sign = '-' %if x < 0 748 !y = |x|+0.5/10.0\{^}m; !modulus, rounded 749 POW = 10.0^|m| 750 POW = 0.5/POW 751 y = |x| 752 y = y+POW; !modulus, rounded 753 %if y > pmax %start 754 count = count+1 %and y = y/10.0 %until y < 10.0 755 %finish 756 z = 1.0 757 %cycle 758 i = i+1; z = z*10.0 759 %repeat %until z > y 760 spaces(n-i) 761 printsymbol(sign) %unless sign = ' ' %and n <= 0 ;! n < 0 just makes a difference of one space character 762 %cycle 763 z = z/10.0 764 l = int pt(y/z) 765 y = y-l*z 766 printsymbol(l+'0') 767 i = i-1 768 %exit %if i+m <= 0 769 print symbol('.') %if i = 0 770 %repeat 771 printsymbol('@') %and write 68k(count,0) %if count # 0 772 %end; !print 773 774 %routine printfl 68k(%long %real x, %integer n) 775 %long %real y,round 776 %integer count=-99,sign=0 777 %if x # 0 %start 778 x = -x %and sign = 1 %if x < 0 779 !Adjust X so that 1.0 <= rounded(X) < 10.0 780 !count = 0; round = 0.5\{^}n 781 count = 0; round = 0.5^|n| 782 y = 1.0-round 783 %if x < y %start; !ie rounded(X) < 1.0 784 count = count-1 %and x = x*10.0 %until x >= y 785 %finish %else %start 786 y = 10.0-round 787 %while x >= y %cycle; !ie rounded(X) > 10.0 788 count = count+1; x = x/10.0 789 %repeat 790 %finish 791 x = -x %if sign # 0 792 %finish 793 print 68k(x,1,n) 794 printsymbol('@') 795 write 68k(count,0) 796 %end; !printfl 797 798 ! ----------------------------------------------------------------------------------------- 799 %routine mousesprintfl(%real r, %integer places) 800 %routine round(%real %name v, %integer p) 801 %real x 802 %return %if v = 0 803 %if v < 0 %then x = -0.5 %else x = 0.5 804 %cycle 805 p = p - 1 806 %exit %if p < 0 807 x = x / 10 808 %repeat 809 v = v + x 810 %end 811 %routine fraction(%real r, %integer places, %string (*) %name s) 812 %long %real %fn fracpt(%long %real x) 813 %long %real z 814 z = x - intpt(x) 815 z = z + 1.0 %if z < 0 816 %result = z 817 %end 818 !%realfn MOUSES FRACPT(%real r) ;! dependent on machine representation of floating point. 819 ! %integer m,x,z 820 ! %if |R| >= 1.0 %start 821 ! x = integer(addr(r)) 822 ! z = x>>24&b'01111111' 823 ! %result = 0 %if z > x'46' 824 ! m = -1 825 ! m = m<<4 %and z = z+1 %while z < x'46' 826 ! x = x&(m!x'FF000000') 827 ! r = r-real(addr(x)) 828 ! %finish 829 ! r = r+1 %if r < 0 830 ! %result = r 831 !%end 832 833 %return %if places <= 0 834 s = s . "." 835 %while places > 0 %cycle 836 places = places - 1 837 r = fracpt(r) * 10 838 s = s . tostring(intpt(r) + '0') 839 %repeat 840 %end 841 %string (64) %fn ftos(%real r, %integer p) 842 {XOWN}%string (64) s = "" 843 %integer exp, sym 844 p = 7 %if p > 7 845 %if r < 0 %then sym = '-' %and r = -r %else sym = ' ' 846 %if r = 0 %start 847 exp = -99 848 %else 849 exp = 0 850 exp = exp + 1 %and r = r / 10 %while r >= 10 851 exp = exp - 1 %and r = r * 10 %while r < 1 852 %finish 853 round(r, p) 854 exp = exp + 1 %and r = r / 10 %if r >= 10 855 s = s . tostring(intpt(r) + '0') 856 fraction(r, p, s) 857 s = s . "@" . itos(exp, 0) 858 %result = s 859 %end 860 printstring(ftos(r, places)) 861 %end 862 ! ----------------------------------------------------------------------------------------- 863 864 ! Convert a floating point number to a string, along the lines of 865 ! +/-nnn.nnn@+/-nn, to occupy Places character spaces 866 ! with sf significant figures. 867 ! Rounding/truncation will occur to ensure sf limits 868 ! 869 ! %longreal f = floating point number to convert to a string 870 ! %integer sf = number of significant figures required 871 ! places = number of characters in converted string 872 ! 873 ! N.B. sf must be in the range of accuracy of %longreal (<15) 874 ! places must be slightly greater than sf 875 ! (to allow for sign and scientific notation) 876 ! sign uses 1 character (+ is omitted) 877 ! decimal point uses 1 character 878 ! scientific notation could use : 879 ! 1 char for "@" ndicating scientific notation 880 ! 1 char for sign of exponent (sign always present) 881 ! 1+ char for integer exponent 882 ! 883 ! N.B. scientific notation is only used if the decimal string could 884 ! not fit in the places required 885 ! 886 ! eg. F to S (0.003999, 3, 7) -> " 0.004 " (note rounding creates zeros after the 4) 887 ! or 888 ! eg. F to S (0.003999, 3, 4) -> " 4.00@-3" (converted to scientific notation to fit) 889 ! eg. F to S (0.003999, 3, 6) -> " 0.004" (fits exactly) 890 ! 891 %string(255) %function imp77 F to S( %longreal f, %integer sf, places) 892 %string(255) s,sx 893 %string(15) science 894 %integer dc,carry 895 %integer sign,exponent, digit, point 896 897 %if (sf > 14) %then sf = 14 898 899 %begin 900 %integer i 901 %integerarray digits(1:sf+1) 902 903 sign = 0 904 905 %for i = 1,1,sf+1 %cycle 906 digits(i) = 0 907 %repeat 908 909 %if (f < 0) %then sign = 1 %and f = -f 910 911 ! clear the floating point string 912 length(s) = 0 913 914 %if (f = 0) %start 915 ! convert to positive zero 916 sign = 0 917 dc = 2 918 sf = 2 919 exponent = 0 920 point = 1 921 %finish %else %start 922 ! prepare the exponent, point values for (f # 0) 923 ! evaluate the exponent value 924 exponent = 0 925 %while f < 1 %cycle 926 f = f * 10 927 exponent = exponent - 1 928 %repeat 929 930 %while f >= 10 %cycle 931 f = f / 10 932 exponent = exponent + 1 933 %repeat 934 935 ! locate the decimal point 936 point = 1 937 ! zero the digit count 938 dc = 0 939 940 ! Loop to get the significant figure digits with an extra digit 941 ! so we can use it to round the the required sig figs 942 ! Rounding as we go through this loop can "oversize" the digit. 943 ! The next stage does the rounding to the correct sig fig. 944 %for i = 1,1,sf + 1 %cycle 945 digit = int pt(f) 946 %if (digit > 9) %then digit = 9 947 dc = dc + 1 948 digits(dc) = digit 949 f = (f - digit)*10 950 %repeat 951 952 ! do the rounding to the sf significant figures 953 %if (digits(dc) > 4) %start 954 ! we need to possibly round up the digits sequence 955 carry = 1 956 %for i = dc - 1,-1,1 %cycle 957 digits(i) = digits(i) + carry 958 %if (digits(i) > 9) %start 959 carry = 1 960 digits(i) = 0 961 %finish %else %start 962 carry = 0 963 %finish 964 %repeat 965 %finish 966 ! ignore the rounding digit 967 dc = dc - 1 968 969 ! Now eliminate any trailing zeros 970 ! We assume that rounding has created 971 ! a sequence of trailing zeros 972 %while (digits(dc) = 0) %and (dc > 1) %cycle 973 dc = dc - 1 974 %repeat 975 sf = dc 976 %finish 977 978 length(s) = 0 979 point = 1 980 %for i = 1,1,sf %cycle 981 s = s.tostring(digits(i) + '0') 982 %if (point = i) %then s = s."." 983 %repeat 984 %if (sf = point) %then s = s."0" 985 %if (exponent < 0) %then s=s."@-".itos(-exponent,0) 986 %if (exponent > 0) %then s=s."@+".itos(exponent,0) 987 science = s 988 989 ! So, now for the other floating point layouts 990 ! we should see which version will fit into the places allowed 991 ! 1) 0.00000nnnn (point <= 0) 992 ! OR 993 ! 2) nnnn.0 (point = sf) 994 ! OR 995 ! 3) nnnn00000.0 (point > sf) 996 ! 997 length(s) = 0 998 point = exponent + 1 999 ! First form the sig fig digit string 1000 ! possibly including a decimal point 1001 %for i = 1,1,sf %cycle 1002 s = s.tostring(digits(i) + '0') 1003 %if (point = i) %then s = s."." 1004 %repeat 1005 1006 ! Checking for format (1) 1007 ! 1) 0.00000nnnn (point <= 0) 1008 %if (point <= 0) %start 1009 %for i=1,1,-point %cycle 1010 s = "0".s 1011 %repeat 1012 s = "0.".s 1013 %finish 1014 1015 ! Checking for format (2) 1016 ! 2) nnnn.0 (point = sf) 1017 %if (point = sf) %start 1018 s = s."0" 1019 %finish 1020 1021 ! Checking for format (3) 1022 ! 3) nnnn00000.0 (point > sf) 1023 %if (point > sf) %start 1024 %for i = 1,1,(point - sf) %cycle 1025 s = s."0" 1026 %repeat 1027 s = s.".0" 1028 %finish 1029 1030 ! Check to see which fits scientific v decimal notation 1031 ! Preference is for the decimal notation 1032 %if (length(s) > places) %start 1033 s = science 1034 %finish 1035 1036 %if (sign # 0) %start 1037 s = "-".s 1038 %finish %else %start 1039 s = " ".s 1040 %finish 1041 1042 %if (length(s) < places) %start 1043 %while (length(s) < places) %cycle 1044 s = s." " 1045 %repeat 1046 %finish 1047 %end 1048 1049 %result = s 1050 %end ?SX unused 1051 1052 %routine imp77 print(%longreal f, %integer sf,places) 1053 printstring( imp77 F to S( f, sf, places ) ) 1054 %end 1055 1056 ! Print a floating point number out, along the lines of 1057 ! +/-nnn.nnn@+/-nn, to occupy Places character spaces. 1058 ! Note - there's a bug in this code such that it does not 1059 ! round the number properly. EG 3.999999999 to 4 places 1060 ! should be 4.00 but we print 3.99 1061 1062 %routine imp77 printfl(%longreal x, %integer places) 1063 %integer exponent, digit, point, printexpo 1064 1065 %if x = 0 %start 1066 printsymbol('0') 1067 printsymbol('.') 1068 printsymbol('0') 1069 %while places > 3 %cycle 1070 printsymbol('0') 1071 places = places - 1 1072 %repeat 1073 %return 1074 %finish 1075 1076 %if x < 0 %then printsymbol('-') %and x = -x %and places = places - 1 1077 1078 %if places < 3 %then places = 3 1079 1080 exponent = 0 1081 printexpo = 0 1082 1083 %while x < 1 %cycle 1084 x = x * 10 1085 exponent = exponent - 1 1086 %repeat 1087 1088 %while x >= 10 %cycle 1089 x = x / 10 1090 exponent = exponent + 1 1091 %repeat 1092 1093 ! Now X is between 1.0 and 9.99 and exponent is set accordingly 1094 ! If the exponent is "large" we will use scientific notation 1095 point = places - 2; ! for useful digits after the "0." 1096 %if exponent >= places %or exponent < -point %start 1097 printexpo = exponent 1098 exponent = 0 1099 places = places - 2 1100 %finish 1101 1102 ! Now the exponent is small-ish 1103 %if exponent < 0 %start; ! 0.nnnn 1104 printsymbol('0') 1105 printsymbol('.') 1106 places = places - 2 1107 1108 %while exponent < -1 %cycle 1109 printsymbol('0') 1110 exponent = exponent + 1 1111 places = places - 1 1112 %repeat 1113 1114 point = -1; ! because we've already passed that 1115 %else; ! nnn.nnn 1116 point = exponent 1117 %finish 1118 1119 %while places > 0 %cycle 1120 digit = int pt(x) 1121 ! Rounding as we go through this loop can "oversize" the digit. This 1122 ! of course tells us that we should have printed (eg) 40000 but we 1123 ! are now stuck with printing 39999 1124 %if digit > 9 %then digit = 9 1125 printsymbol(digit + '0') 1126 x = (x - digit)*10 1127 %if point = 0 %then printsymbol('.') %and places = places - 1 1128 point = point - 1 1129 places = places - 1 1130 %repeat 1131 1132 %if printexpo # 0 %start 1133 printsymbol('@') 1134 write68k(printexpo, 1) 1135 %finish 1136 %end 1137 1138 ! ######################################################################################### 1139 1140 %routine writetest(%string (255) procname, %routine testedwrite(%integer n, p)) 1141 %const %integer GAP = 15 ;! allow for largest int (+ or -) plus manditory spaces 1142 %const %integer COLS = 13 ;! 1 beyond what is an exact fit 1143 %integer n, m, shift, startpos, ALIGN 1144 %for ALIGN = 1, -2, -1 %cycle 1145 n = 1 1146 printstring(procname) 1147 printstring(": ") 1148 newlines(2) 1149 %if ALIGN < 0 %then %start 1150 printstring("In IMP77, the total number of print positions to be used is defined"); newline 1151 printstring("by the modulus of the second parameter (n). If this parameter is"); newline 1152 printstring("negative, no space character should output before a positive value:"); newline 1153 newlines(2) 1154 %finish 1155 printstring(" \ n | write(m,n):"); newline 1156 printstring(" \ |"); newline 1157 printstring(" \ |"); newline 1158 printstring(" m \|"); newline 1159 printstring(" ----+ ") 1160 %for m = 0, 1, cols %cycle 1161 startpos = outpos 1162 write68k(ALIGN*m,0) 1163 spaces(GAP-(outpos-startpos)) 1164 %repeat 1165 newline 1166 printstring("MIN INT ") 1167 %for m = 0, 1, cols %cycle 1168 startpos = outpos 1169 printsymbol('[') 1170 !tested write(-2147483648, m) 1171 testedwrite(16_80000000, ALIGN*m) 1172 printsymbol(']') 1173 spaces(gap - (outpos - startpos)) 1174 %repeat 1175 newline 1176 n = 1000000000 1177 shift = 9 1178 %cycle 1179 printstring(" -10^") 1180 write68k(shift, 0) 1181 spaces(2) 1182 %for m = cols, -1, 0 %cycle 1183 ! examine the number of characters between the [] brackets 1184 startpos = outpos 1185 printsymbol('[') 1186 testedwrite(-n, ALIGN*(cols - m)) 1187 printsymbol(']') 1188 spaces(gap - (outpos - startpos)) 1189 %repeat 1190 newline 1191 n = n // 10 1192 shift = shift - 1 1193 %repeat %until n = 0 1194 printstring(" 0 ") 1195 %for m = 0, 1, cols %cycle 1196 startpos = outpos 1197 printsymbol('[') 1198 testedwrite(0, ALIGN*m) 1199 printsymbol(']') 1200 spaces(gap - (outpos - startpos)) 1201 %repeat 1202 newline 1203 n = 1 1204 shift = 1 1205 %cycle 1206 printstring(" 10^") 1207 write68k(shift - 1, 0) 1208 spaces(2) 1209 %for m = 0, 1, cols %cycle 1210 ! examine the number of characters between the [] brackets 1211 startpos = outpos 1212 printsymbol('[') 1213 testedwrite(n, ALIGN*m) 1214 printsymbol(']') 1215 spaces(gap - (outpos - startpos)) 1216 %repeat 1217 newline 1218 %exit %if n = 1000000000 1219 n = n * 10 1220 shift = shift + 1 1221 %repeat 1222 printstring("MAX INT ") 1223 %for m = 0, 1, cols %cycle 1224 startpos = outpos 1225 printsymbol('[') 1226 testedwrite(2147483647, ALIGN*m) 1227 printsymbol(']') 1228 spaces(gap - (outpos - startpos)) 1229 %repeat 1230 newlines(3) 1231 %repeat 1232 %end 1233 1234 ! ######################################################################################### 1235 1236 %routine printfltest(%string (255) procname, %routine testedprintfl(%longreal n, %integer p)) 1237 %const %integer GAP = 26 ;! allow for accuracy of long reals 1238 %const %integer COLS = 17 1239 %long %real n, k 1240 %integer m, exponent, startpos, ALIGN 1241 %for ALIGN = 1, 1, 1 %cycle ; ! %for ALIGN = 1, -2, -1 %cycle 1242 n = 1 1243 %if ALIGN < 0 %then printstring("Possibly left-justified ") 1244 printstring(procname) 1245 !%if neg < 0 %then printstring(" with negative numbers") 1246 printstring(": ") 1247 newlines(2) 1248 1249 printstring(" \ n | printfl(m,n):"); newline 1250 printstring(" \ |"); newline 1251 printstring(" \ |"); newline 1252 printstring(" m \|"); newline 1253 printstring(" ----+ ") 1254 %for m = 0, 1, cols %cycle 1255 startpos = outpos 1256 write68k(m,0) 1257 spaces(GAP-(outpos-startpos)) 1258 %if REM(m, 6)=5 %then newline %and print string(" "); 1259 %repeat 1260 newline 1261 n = 1.23456789012345678901234567890 1262 1263 k = 1@100; exponent = 100 1264 %cycle 1265 printstring("-1@") 1266 write68k(exponent, 0) 1267 space %while outpos < 8 1268 %for m = cols, -1, 0 %cycle 1269 ! examine the number of characters between the [] brackets 1270 startpos = outpos 1271 printsymbol('[') 1272 testedprintfl(-k*n, ALIGN*(cols - m)) 1273 printsymbol(']') 1274 spaces(gap - (outpos - startpos)) 1275 %if REM(cols-m, 6)=5 %then newline %and print string(" "); 1276 %repeat 1277 newline 1278 %if exponent = 100 %start 1279 k = 1@10; exponent = 10 1280 %else %if exponent = 10 1281 k = 1@1; exponent = 1 1282 %else %if exponent = 1 1283 k = 1@0; exponent = 0 1284 %else %if exponent = 0 1285 k = 1@-1; exponent = -1 1286 %else %if exponent = -1 1287 k = 1@-10; exponent = -10 1288 %else %if exponent = -10 1289 k = 1@-100; exponent = -100 1290 %finish %else %exit 1291 %repeat 1292 1293 printstring(" 0@0 ") 1294 %for m = 0, 1, cols %cycle 1295 startpos = outpos 1296 printsymbol('[') 1297 testedprintfl(0, ALIGN*m) 1298 printsymbol(']') 1299 spaces(gap - (outpos - startpos)) 1300 %if REM(m, 6)=5 %then newline %and print string(" "); 1301 %repeat 1302 newline 1303 1304 k = 1@100; exponent = 100 1305 %cycle 1306 printstring(" 1@") 1307 write68k(exponent, 0) 1308 space %while outpos < 8 1309 %for m = cols, -1, 0 %cycle 1310 ! examine the number of characters between the [] brackets 1311 startpos = outpos 1312 printsymbol('[') 1313 testedprintfl(k*n, ALIGN*(cols - m)) 1314 printsymbol(']') 1315 spaces(gap - (outpos - startpos)) 1316 %if REM(cols-m, 6)=5 %then newline %and print string(" "); 1317 %repeat 1318 newline 1319 %if exponent = 100 %start 1320 k = 1@10; exponent = 10 1321 %else %if exponent = 10 1322 k = 1@1; exponent = 1 1323 %else %if exponent = 1 1324 k = 1; exponent = 0 1325 %else %if exponent = 0 1326 k = 0.1; exponent = -1 1327 %else %if exponent = -1 1328 k = 1@-10; exponent = -10 1329 %else %if exponent = -10 1330 k = 1@-100; exponent = -100 1331 %finish %else %exit 1332 %repeat 1333 1334 newlines(3) 1335 %repeat 1336 %end 1337 1338 1339 ! WRITE TESTS write(num, places) 1340 1341 writetest("Copy of current (imp2024) write", write3l) 1342 writetest("EMAS write", emaswrite) 1343 writetest("EMAS write (from imp77 compatibility library)", emasi77write) 1344 writetest("Vax write", vaxwrite) 1345 writetest("Dec10 (imp77) write", dec10 write) 1346 writetest("Mouses (imp77) write", mouseswrite) 1347 writetest("APM (imp68k) write", write68k) 1348 1349 ! PRINTFL TESTS printfl(num, places) 1350 1351 ! Current imp77 library uses some weird names. I've extracted the equivalents of 1352 ! the standard printfl and print, and renamed them to avoid confusion. 1353 ! Their output is not compatible with the ercc, vax, or dec10 versions :-( 1354 1355 ! As with write, I had to provide a local copy in order to support 'outpos' 1356 ! which is necessary to handle padding correctly when the spacing of the 1357 ! routine under test is unknown. 1358 1359 printfltest("copy of current (imp2024) printfl (called 'print' in the imp2024 library)", imp77 printfl) 1360 printfltest("EMAS (imp9) printfl", emas printfl) 1361 printfltest("ERCC (imp80) printfl", ercc printfl) 1362 printfltest("Vax (imp77) printfl", vax printfl) 1363 printfltest("Dec10 (imp77) printfl", dec10 printfl) 1364 printfltest("Mouses (imp77) printfl", mouses printfl) 1365 printfltest("APM (imp68k) printfl", dec10 printfl) 1366 1367 ! TO DO: PRINT TESTS print(num, iplaces, dplaces) 1368 1369 1370 %end %of %programm ?IMP77PRINT unused ?PRINTFL68K unused ?DEC10PRINT unused ?WRITE unused 1110 Statements compiled