about summary refs log tree commit diff
path: root/other/pamarith.c
blob: c67c73260557c843433c5b95e9f99611560cef3e (plain) (blame)
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
#include <assert.h>
#include <string.h>
#include <limits.h>

#include "pm_c_util.h"
#include "mallocvar.h"
#include "nstring.h"
#include "shhopt.h"
#include "pam.h"

enum function {FN_ADD, FN_SUBTRACT, FN_MULTIPLY, FN_DIVIDE, FN_DIFFERENCE,
               FN_MINIMUM, FN_MAXIMUM, FN_MEAN, FN_COMPARE,
               FN_AND, FN_OR, FN_NAND, FN_NOR, FN_XOR,
               FN_SHIFTLEFT, FN_SHIFTRIGHT
              };



static bool
isDyadic(enum function const function) {

    bool retval;

    switch (function) {
    case FN_ADD:
    case FN_MEAN:
    case FN_AND:
    case FN_OR:
    case FN_XOR:
        retval = FALSE;
        break;
    case FN_SUBTRACT:
    case FN_DIFFERENCE:
    case FN_MINIMUM:
    case FN_MAXIMUM:
    case FN_COMPARE:
    case FN_MULTIPLY:
    case FN_DIVIDE:
    case FN_NAND:
    case FN_NOR:
    case FN_SHIFTLEFT:
    case FN_SHIFTRIGHT:
        retval = TRUE;
        break;
    }
    return retval;
}



struct CmdlineInfo {
    /* All the information the user supplied in the command line,
       in a form easy for the program to use.
    */
    enum function function;
    unsigned int operandCt;
    const char ** operandFileNames;
};



static void
parseCommandLine(int argc, const char ** const argv,
                 struct CmdlineInfo * const cmdlineP) {
/*----------------------------------------------------------------------------
   Note that the file spec array we return is stored in the storage that
   was passed to us as the argv array.
-----------------------------------------------------------------------------*/
    optEntry * option_def;
        /* Instructions to OptParseOptions2 on how to parse our options.
         */
    optStruct3 opt;

    unsigned int option_def_index;

    unsigned int addSpec, subtractSpec, multiplySpec, divideSpec,
        differenceSpec,
        minimumSpec, maximumSpec, meanSpec, compareSpec,
        andSpec, orSpec, nandSpec, norSpec, xorSpec,
        shiftleftSpec, shiftrightSpec;

    MALLOCARRAY_NOFAIL(option_def, 100);

    option_def_index = 0;   /* incremented by OPTENTRY */
    OPTENT3(0, "add",         OPT_FLAG,   NULL, &addSpec,        0);
    OPTENT3(0, "subtract",    OPT_FLAG,   NULL, &subtractSpec,   0);
    OPTENT3(0, "multiply",    OPT_FLAG,   NULL, &multiplySpec,   0);
    OPTENT3(0, "divide",      OPT_FLAG,   NULL, &divideSpec,     0);
    OPTENT3(0, "difference",  OPT_FLAG,   NULL, &differenceSpec, 0);
    OPTENT3(0, "minimum",     OPT_FLAG,   NULL, &minimumSpec,    0);
    OPTENT3(0, "maximum",     OPT_FLAG,   NULL, &maximumSpec,    0);
    OPTENT3(0, "mean",        OPT_FLAG,   NULL, &meanSpec,       0);
    OPTENT3(0, "compare",     OPT_FLAG,   NULL, &compareSpec,    0);
    OPTENT3(0, "and",         OPT_FLAG,   NULL, &andSpec,        0);
    OPTENT3(0, "or",          OPT_FLAG,   NULL, &orSpec,         0);
    OPTENT3(0, "nand",        OPT_FLAG,   NULL, &nandSpec,       0);
    OPTENT3(0, "nor",         OPT_FLAG,   NULL, &norSpec,        0);
    OPTENT3(0, "xor",         OPT_FLAG,   NULL, &xorSpec,        0);
    OPTENT3(0, "shiftleft",   OPT_FLAG,   NULL, &shiftleftSpec,  0);
    OPTENT3(0, "shiftright",  OPT_FLAG,   NULL, &shiftrightSpec, 0);

    opt.opt_table = option_def;
    opt.short_allowed = FALSE;  /* We have no short (old-fashioned) options */
    opt.allowNegNum = FALSE;  /* We have no parms that are negative numbers */

    pm_optParseOptions3(&argc, (char **)argv, opt, sizeof(opt), 0);
        /* Uses and sets argc, argv, and some of *cmdlineP and others. */

    if (addSpec + subtractSpec + multiplySpec + divideSpec + differenceSpec +
        minimumSpec + maximumSpec + meanSpec + compareSpec +
        andSpec + orSpec + nandSpec + norSpec + xorSpec +
        shiftleftSpec + shiftrightSpec > 1)
        pm_error("You may specify only one function");

    if (addSpec)
        cmdlineP->function = FN_ADD;
    else if (subtractSpec)
        cmdlineP->function = FN_SUBTRACT;
    else if (multiplySpec)
        cmdlineP->function = FN_MULTIPLY;
    else if (divideSpec)
        cmdlineP->function = FN_DIVIDE;
    else if (differenceSpec)
        cmdlineP->function = FN_DIFFERENCE;
    else if (minimumSpec)
        cmdlineP->function = FN_MINIMUM;
    else if (maximumSpec)
        cmdlineP->function = FN_MAXIMUM;
    else if (meanSpec)
        cmdlineP->function = FN_MEAN;
    else if (compareSpec)
        cmdlineP->function = FN_COMPARE;
    else if (andSpec)
        cmdlineP->function = FN_AND;
    else if (orSpec)
        cmdlineP->function = FN_OR;
    else if (nandSpec)
        cmdlineP->function = FN_NAND;
    else if (norSpec)
        cmdlineP->function = FN_NOR;
    else if (xorSpec)
        cmdlineP->function = FN_XOR;
    else if (shiftleftSpec)
        cmdlineP->function = FN_SHIFTLEFT;
    else if (shiftrightSpec)
        cmdlineP->function = FN_SHIFTRIGHT;
    else
        pm_error("You must specify a function (e.g. '-add')");

    if (argc-1 < 2)
        pm_error("You must specify at least two arguments: the files which "
                 "are the operands of the function.  You specified %u",
                 argc-1);
    else {
        if (isDyadic(cmdlineP->function) && argc-1 > 2)
            pm_error("You specified %u arguments, but a dyadic function.  "
                     "For a dyadic function, you must specify 2 arguments:  "
                     "the operands of the function", argc-1);
        else {
            cmdlineP->operandCt = argc-1;
            cmdlineP->operandFileNames = &argv[1];
        }
    }
}



enum category {
    CATEGORY_FRACTIONAL_ARITH,
        /* Arithmetic in which each sample represents a the fraction
           sample/maxval.
        */
    CATEGORY_BITSTRING,
        /* And, Or, etc.  Maxval isn't a scale factor at all; it's a mask. */
    CATEGORY_SHIFT
        /* Left argument is a bit string, but right argument is a whole
           number (left maxval is a mask; right maxval is meaningless).
        */
};



static enum category
functionCategory(enum function const function) {

    enum category retval;

    switch (function) {
    case FN_ADD:
    case FN_SUBTRACT:
    case FN_DIFFERENCE:
    case FN_MINIMUM:
    case FN_MAXIMUM:
    case FN_MEAN:
    case FN_COMPARE:
    case FN_MULTIPLY:
    case FN_DIVIDE:
        retval = CATEGORY_FRACTIONAL_ARITH;
        break;
    case FN_AND:
    case FN_OR:
    case FN_NAND:
    case FN_NOR:
    case FN_XOR:
        retval = CATEGORY_BITSTRING;
        break;
    case FN_SHIFTLEFT:
    case FN_SHIFTRIGHT:
        retval = CATEGORY_SHIFT;
        break;
    }
    return retval;
}



static int
outFmtForCompare(int const format1,
                 int const format2) {
/*----------------------------------------------------------------------------
   The format for the output image, given that the function is Compare and
   'format1' and 'format2' are the input image formats.

   Note that Compare is special because its maxval is always 2, and that
   won't work for PBM.
-----------------------------------------------------------------------------*/
    int const tentativeFormat = MAX(format1, format2);

    int retval;

    switch (tentativeFormat) {
    case PBM_FORMAT:  retval = PGM_FORMAT;      break;
    case RPBM_FORMAT: retval = RPGM_FORMAT;     break;
    default:          retval = tentativeFormat; break;
    }
    return retval;
}



static void
computeOutputType(struct pam *  const outpamP,
                  struct pam    const inpam1,
                  struct pam    const inpam2,
                  enum function const function) {

    outpamP->size        = sizeof(struct pam);
    outpamP->len         = PAM_STRUCT_SIZE(tuple_type);
    outpamP->file        = stdout;
    outpamP->plainformat = FALSE;
    outpamP->height      = inpam1.height;
    outpamP->width       = inpam1.width;
    outpamP->depth       = MAX(inpam1.depth, inpam2.depth);


    if (function == FN_COMPARE)
        outpamP->format = outFmtForCompare(inpam1.format, inpam2.format);
    else
        outpamP->format = MAX(inpam1.format, inpam2.format);

    switch (functionCategory(function)) {
    case CATEGORY_FRACTIONAL_ARITH:
        if (function == FN_COMPARE)
            outpamP->maxval = 2;
        else
            outpamP->maxval = MAX(inpam1.maxval, inpam2.maxval);
        break;
    case CATEGORY_BITSTRING:
        if (inpam2.maxval != inpam1.maxval)
            pm_error("For a bit string operation, the maxvals of the "
                     "left and right image must be the same.  You have "
                     "left=%u and right=%u",
                     (unsigned)inpam1.maxval, (unsigned)inpam2.maxval);

        if (pm_bitstomaxval(pm_maxvaltobits(inpam1.maxval)) != inpam1.maxval)
            pm_error("For a bit string operation, the maxvals of the inputs "
                     "must be a full binary count, i.e. a power of two "
                     "minus one such as 0xff.  You have 0x%x",
                     (unsigned)inpam1.maxval);

        outpamP->maxval = inpam1.maxval;
        break;
    case CATEGORY_SHIFT:
        if (pm_bitstomaxval(pm_maxvaltobits(inpam1.maxval)) != inpam1.maxval)
            pm_error("For a bit shift operation, the maxval of the left "
                     "input image "
                     "must be a full binary count, i.e. a power of two "
                     "minus one such as 0xff.  You have 0x%x",
                     (unsigned)inpam1.maxval);
        outpamP->maxval = inpam1.maxval;
    }
    outpamP->bytes_per_sample = (pm_maxvaltobits(outpamP->maxval)+7)/8;

    if (outpamP->maxval > 1 && strneq(inpam1.tuple_type, "BLACKANDWHITE", 13))
        strcpy(outpamP->tuple_type, "");
    else
        strcpy(outpamP->tuple_type, inpam1.tuple_type);
}



static samplen
samplenSum(samplen      const operands[],
           unsigned int const operandCt) {

    unsigned int i;
    samplen total;

    for (i = 1, total = operands[0]; i < operandCt; ++i) {
        total += operands[1];
        if (total > 1.)
            total = 1.;
    }
    return total;
}



static samplen
samplenMin(samplen      const operands[],
           unsigned int const operandCt) {

    unsigned int i;
    samplen min;

    for (i = 1, min = operands[0]; i < operandCt; ++i) {
        if (operands[i] < min)
            min = operands[i];
    }
    return min;
}



static samplen
samplenMax(samplen      const operands[],
           unsigned int const operandCt) {

    unsigned int i;
    samplen max;

    for (i = 1, max = operands[0]; i < operandCt; ++i) {
        if (operands[i] > max)
            max = operands[i];
    }
    return max;
}



static samplen
samplenMean(samplen      const operands[],
            unsigned int const operandCt) {

    unsigned int i;
    double sum;

    for (i = 0, sum = 0.; i < operandCt; ++i)
        sum += operands[i];

    return sum / operandCt;
}



static samplen
samplenProduct(samplen      const operands[],
               unsigned int const operandCt) {

    unsigned int i;
    double product;

    for (i = 1, product = operands[0]; i < operandCt; ++i)
        product *= operands[i];

    return product;
}



static samplen
applyNormalizedFunction(enum function const function,
                        samplen       const operands[],
                        unsigned int  const operandCt) {

    samplen result;

    switch (function) {
    case FN_ADD:
        result = samplenSum(operands, operandCt);
        break;
    case FN_SUBTRACT:
        result = MAX(0., operands[0] - operands[1]);
        break;
    case FN_MULTIPLY:
        result = samplenProduct(operands, operandCt);
        break;
    case FN_DIVIDE:
        result = (operands[1] > operands[0]) ?
        operands[0] / operands[1] : 1.;
        break;
    case FN_DIFFERENCE:
        result = operands[0] > operands[1] ?
            operands[0] - operands[1] : operands[1] - operands[0];
        break;
    case FN_MINIMUM:
        result = samplenMin(operands, operandCt);
        break;
    case FN_MAXIMUM:
        result = samplenMax(operands, operandCt);
        break;
    case FN_MEAN:
        result = samplenMean(operands, operandCt);
        break;
    case FN_COMPARE:
        result =
            operands[0] > operands[1] ?
            1. : operands[0] < operands[1] ?
            0. : .5;
        break;
    default:
        pm_error("Internal error.  applyNormalizedFunction() called "
                 "for a function it doesn't know how to do: %u", function);
    }

    return result;
}



static void
doNormalizedArith(struct pam *  const inpam1P,
                  struct pam *  const inpam2P,
                  struct pam *  const outpamP,
                  enum function const function) {

    /* Some of the logic in this subroutine is designed for future
       expansion into non-dyadic computations.  But for now, all
       computations have exactly two operands.
    */
    unsigned int const operandCt = 2;

    tuplen ** tuplerown;
        /* tuplerown[0] is the current row in the first operand image */

    tuplen * tuplerownOut;
    unsigned int row;
    samplen * operands;
        /* operand[0] is the first operand in the current one-sample
           computation
        */
    unsigned int * plane;
        /* plane[0] is the plane number in the first operand image for
           the current one-sample computation.  plane[1] is the plane number
           in the second operand image, etc.
         */

    MALLOCARRAY_NOFAIL(operands, operandCt);
    MALLOCARRAY_NOFAIL(plane, operandCt);
    MALLOCARRAY_NOFAIL(tuplerown, operandCt);

    tuplerown[0] = pnm_allocpamrown(inpam1P);
    tuplerown[1] = pnm_allocpamrown(inpam2P);
    tuplerownOut = pnm_allocpamrown(outpamP);

    for (row = 0; row < outpamP->height; ++row) {
        unsigned int col;
        pnm_readpamrown(inpam1P, tuplerown[0]);
        pnm_readpamrown(inpam2P, tuplerown[1]);

        for (col = 0; col < outpamP->width; ++col) {
            unsigned int outplane;

            for (outplane = 0; outplane < outpamP->depth; ++outplane) {
                unsigned int op;

                plane[0] = MIN(outplane, inpam1P->depth-1);
                plane[1] = MIN(outplane, inpam2P->depth-1);

                for (op = 0; op < operandCt; ++op)
                    operands[op] = tuplerown[op][col][plane[op]];

                tuplerownOut[col][outplane] =
                    applyNormalizedFunction(function, operands, operandCt);
                assert(tuplerownOut[col][outplane] >= 0.);
                assert(tuplerownOut[col][outplane] <= 1.);
            }
        }
        pnm_writepamrown(outpamP, tuplerownOut);
    }

    pnm_freepamrown(tuplerown[0]);
    pnm_freepamrown(tuplerown[1]);
    free(tuplerown);
    pnm_freepamrown(tuplerownOut);
    free(plane);
    free(operands);
}



static sample
sampleSum(sample       const operands[],
          unsigned int const operandCt,
          sample       const maxval) {

    unsigned int i;
    sample total;

    for (i = 1, total = operands[0]; i < operandCt; ++i) {
        total += operands[i];
        if (total > maxval)
            total = maxval;
    }
    return total;
}



static sample
sampleMin(sample       const operands[],
          unsigned int const operandCt) {

    unsigned int i;
    sample min;

    for (i = 1, min = operands[0]; i < operandCt; ++i) {
        if (operands[i] < min)
            min = operands[i];
    }
    return min;
}



static sample
sampleMax(sample       const operands[],
          unsigned int const operandCt) {

    unsigned int i;
    sample max;

    for (i = 1, max = operands[0]; i < operandCt; ++i) {
        if (operands[i] > max)
            max = operands[i];
    }
    return max;
}



static sample
sampleMean(sample       const operands[],
           unsigned int const operandCt) {

    unsigned int i;
    unsigned int sum;

    for (i = 0, sum = 0; i < operandCt; ++i) {
        sum += operands[i];
        if (UINT_MAX - operands[i] < sum)
            pm_error("Arithmetic overflow adding samples for mean");
    }
    return ROUNDDIV(sum, operandCt);
}



static sample
sampleProduct(sample       const operands[],
              unsigned int const operandCt,
              sample       const maxval) {

    unsigned int i;
    double product;

    for (i = 0, product = 1.0; i < operandCt; ++i) {
        product *= ((double)operands[i]/maxval);
    }
    return (sample)(product * maxval + 0.5);
}



static sample
sampleAnd(sample       const operands[],
          unsigned int const operandCt) {

    unsigned int i;
    sample accum;

    for (i = 1, accum = operands[0]; i < operandCt; ++i) {
        accum &= operands[i];
    }
    return accum;
}



static sample
sampleOr(sample       const operands[],
         unsigned int const operandCt) {

    unsigned int i;
    sample accum;

    for (i = 1, accum = operands[0]; i < operandCt; ++i) {
        accum |= operands[i];
    }
    return accum;
}



static sample
sampleNand(sample       const operands[],
           unsigned int const operandCt,
           sample       const maxval) {

    unsigned int i;
    sample accum;

    for (i = 1, accum = operands[0]; i < operandCt; ++i) {
        accum &= operands[i];
    }
    return ~accum & maxval;
}



static sample
sampleNor(sample       const operands[],
          unsigned int const operandCt,
          sample       const maxval) {

    unsigned int i;
    sample accum;

    for (i = 1, accum = operands[0]; i < operandCt; ++i) {
        accum |= operands[i];
    }
    return ~accum & maxval;
}



static sample
sampleXor(sample       const operands[],
          unsigned int const operandCt) {

    unsigned int i;
    sample accum;

    for (i = 1, accum = operands[0]; i < operandCt; ++i) {
        accum ^= operands[i];
    }
    return accum;
}



static sample
applyUnNormalizedFunction(enum function const function,
                          sample        const operands[],
                          unsigned int  const operandCt,
                          sample        const maxval) {
/*----------------------------------------------------------------------------
   Apply function 'function' to the arguments operands[] (there are
   'operandCt' of them), assuming both are based on the same maxval
   'maxval'.  Return a value which is also a fraction of 'maxval'.

   Exception: for the shift operations, operands[1] is not based on any
   maxval.  It is an absolute bit count.

   We assume 'operandCount' is sensible for 'function'.  E.g. if
   'function' is FN_DIFFERENCE, 'operandCt' is 2.

   For a function that has a concept of left and right argument,
   operands[0] is the left and operands[1] is the right.
-----------------------------------------------------------------------------*/
    sample result;

    switch (function) {
    case FN_ADD:
        result = sampleSum(operands, operandCt, maxval);
        break;
    case FN_SUBTRACT:
        result = MAX(0, (int)operands[0] - (int)operands[1]);
        break;
    case FN_DIFFERENCE:
        result = operands[0] > operands[1] ?
            operands[0] - operands[1] : operands[1] - operands[0];
        break;
    case FN_MINIMUM:
        result = sampleMin(operands, operandCt);
        break;
    case FN_MAXIMUM:
        result = sampleMax(operands, operandCt);
        break;
    case FN_MEAN:
        result = sampleMean(operands, operandCt);
        break;
    case FN_COMPARE:
        result = operands[0] > operands[1] ?
            2 : operands[0] < operands[1] ? 0 : 1;
        break;
    case FN_MULTIPLY:
        result = sampleProduct(operands, operandCt, maxval);
        break;
    case FN_DIVIDE:
        result = (operands[1] > operands[0]) ?
            ROUNDDIV(operands[0] * maxval, operands[1]) : maxval;
        break;

    case FN_AND:
        result = sampleAnd(operands, operandCt);
        break;
    case FN_OR:
        result = sampleOr(operands, operandCt);
        break;
    case FN_NAND:
        result = sampleNand(operands, operandCt, maxval);
        break;
    case FN_NOR:
        result = sampleNor(operands, operandCt, maxval);
        break;
    case FN_XOR:
        result = sampleXor(operands, operandCt);
        break;
    case FN_SHIFTLEFT:
        result = (operands[0] << operands[1]) & maxval;
        break;
    case FN_SHIFTRIGHT:
        result = operands[0] >> operands[1];
        break;
    default:
        pm_error("Internal error.  applyUnNormalizedFunction() called "
                 "for a function it doesn't know how to do: %u", function);
    }

    return result;
}



static void
doUnNormalizedArith(struct pam *  const inpam1P,
                    struct pam *  const inpam2P,
                    struct pam *  const outpamP,
                    enum function const function) {
/*----------------------------------------------------------------------------
   Take advantage of the fact that both inputs and the output use the same
   maxval to do the computation without time-consuming normalization of
   sample values.
-----------------------------------------------------------------------------*/
    /* Some of the logic in this subroutine is designed for future
       expansion into non-dyadic computations.  But for now, all
       computations have exactly two operands.
    */
    unsigned int const operandCt = 2;

    sample const maxval = outpamP->maxval;

    tuple ** tuplerow;
        /* tuplerow[0] is the current row in the first operand image */

    tuple * tuplerowOut;
    unsigned int row;
    sample * operands;
        /* operand[0] is the first operand in the current one-sample
           computation
        */
    unsigned int * plane;
        /* plane[0] is the plane number in the first operand image for
           the current one-sample computation.  plane[1] is the plane number
           in the second operand image, etc.
         */

    /* Input conditions: */
    assert(inpam1P->maxval == maxval);
    assert(inpam2P->maxval == maxval);
    assert(outpamP->maxval == maxval);

    MALLOCARRAY_NOFAIL(operands, operandCt);
    MALLOCARRAY_NOFAIL(plane, operandCt);
    MALLOCARRAY_NOFAIL(tuplerow, operandCt);

    tuplerow[0]   = pnm_allocpamrow(inpam1P);
    tuplerow[1]   = pnm_allocpamrow(inpam2P);
    tuplerowOut = pnm_allocpamrow(outpamP);

    for (row = 0; row < outpamP->height; ++row) {
        unsigned int col;
        pnm_readpamrow(inpam1P, tuplerow[0]);
        pnm_readpamrow(inpam2P, tuplerow[1]);

        for (col = 0; col < outpamP->width; ++col) {
            unsigned int outplane;

            for (outplane = 0; outplane < outpamP->depth; ++outplane) {
                unsigned int op;

                plane[0] = MIN(outplane, inpam1P->depth-1);
                plane[1] = MIN(outplane, inpam2P->depth-1);

                for (op = 0; op < operandCt; ++op)
                    operands[op] = tuplerow[op][col][plane[op]];

                tuplerowOut[col][outplane] =
                    applyUnNormalizedFunction(function, operands, operandCt,
                                              maxval);

                assert(tuplerowOut[col][outplane] >= 0);
                assert(tuplerowOut[col][outplane] <= outpamP->maxval);
            }
        }
        pnm_writepamrow(outpamP, tuplerowOut);
    }

    pnm_freepamrow(tuplerow[0]);
    pnm_freepamrow(tuplerow[1]);
    free(tuplerow);
    pnm_freepamrow(tuplerowOut);
    free(plane);
    free(operands);
}



int
main(int argc, const char *argv[]) {

    struct CmdlineInfo cmdline;
    struct pam inpam1;
    struct pam inpam2;
    struct pam outpam;
    FILE * if1P;
    FILE * if2P;

    pm_proginit(&argc, argv);

    parseCommandLine(argc, argv, &cmdline);

    if (cmdline.operandCt != 2)
        /* Code for > 2 operands not written yet */
        pm_error("You specified %u operands.  We understand only 2.",
                 cmdline.operandCt);

    if1P = pm_openr(cmdline.operandFileNames[0]);
    if2P = pm_openr(cmdline.operandFileNames[1]);

    pnm_readpaminit(if1P, &inpam1, PAM_STRUCT_SIZE(tuple_type));
    pnm_readpaminit(if2P, &inpam2, PAM_STRUCT_SIZE(tuple_type));

    if (inpam1.width != inpam2.width || inpam1.height != inpam2.height)
        pm_error("The two images must be the same width and height.  "
                 "The first is %ux%ux%u, but the second is %ux%ux%u",
                 inpam1.width, inpam1.height, inpam1.depth,
                 inpam2.width, inpam2.height, inpam2.depth);

    if (inpam1.depth != 1 && inpam2.depth != 1 && inpam1.depth != inpam2.depth)
        pm_error("The two images must have the same depth or one of them "
                 "must have depth 1.  The first has depth %u and the second "
                 "has depth %u", inpam1.depth, inpam2.depth);

    computeOutputType(&outpam, inpam1, inpam2, cmdline.function);

    pnm_writepaminit(&outpam);

    switch (functionCategory(cmdline.function)) {
    case CATEGORY_FRACTIONAL_ARITH:
        if (inpam1.maxval == inpam2.maxval && inpam2.maxval == outpam.maxval)
            doUnNormalizedArith(&inpam1, &inpam2, &outpam, cmdline.function);
        else
            doNormalizedArith(&inpam1, &inpam2, &outpam, cmdline.function);
        break;
    case CATEGORY_BITSTRING:
    case CATEGORY_SHIFT:
        doUnNormalizedArith(&inpam1, &inpam2, &outpam, cmdline.function);
        break;
    }

    pm_close(if1P);
    pm_close(if2P);

    return 0;
}