Re: xxd -i vs DIY Was: C23 thoughts and opinions

On Tue, 28 May 2024 23:23:15 +0300
Michael S <already5chosen@yahoo.com> wrote:

 
Also, I think that random numbers are close to worst case for branch
predictor / loop length predictor in my inner loop.
Were I thinking about random case upfront, I'd code an inner loop
differently. I'd always copy 4 octets (comma would be stored in the
same table). After that I would update outptr by length taken from
additional table, similarly, but not identically to your method below.
 

That's what I had in mind:

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

static const char usage[] =
"bin_to_list - convert binary file to comma-delimited list of decimal
numbers\n" "Usage:\n"
" bin_to_list infile [oufile]\n"
"When output file is not specified, the result is written to standard
output.\n" ;
int main(int argz, char** argv)
{
  // process command line
  if (argz < 2) {
    fprintf(stderr, "%s", usage);
    return 1;
  }

  char* infilename = argv[1];
  static const char *help_aliases[] = { "-h", "-H", "-?", "--help",
  "--?" }; const int n_help_aliases =
  sizeof(help_aliases)/sizeof(help_aliases[0]); for (int i = 0; i <
  n_help_aliases; ++i) { if (strcmp(infilename, help_aliases[i])==0) {
      fprintf(stderr, "%s", usage);
      return 0;
    }
  }

  // open files
  FILE* fpin = fopen(infilename, "rb");
  if (!fpin) {
    perror(infilename);
    return 1;
  }

  FILE* fpout = stdout;
  char* outfilename = NULL;
  if (argz > 2) {
    outfilename = argv[2];
    fpout = fopen(outfilename, "w");
    if (!fpout) {
      perror(outfilename);
      fclose(fpin);
      return 1;
    }
  }

  enum { MAX_CHAR_PER_LINE = 80, MAX_CHAR_PER_NUM = 4,
    ALMOST_FULL_THR = MAX_CHAR_PER_LINE-MAX_CHAR_PER_NUM };
  // Initialize table
  unsigned char bin2dec[256][MAX_CHAR_PER_NUM+1]; //
  bin2dec[MAX_CHAR_PER_NUM] => length for (int i = 0; i < 256;++i) {
    char tmp[8];
    int len = sprintf(tmp, "%d,", i);
    memcpy(bin2dec[i], tmp, MAX_CHAR_PER_NUM);
    bin2dec[i][MAX_CHAR_PER_NUM] = (unsigned char)len;
  }

  // main loop
  int err = 0;
  int c;
  unsigned char outbuf[MAX_CHAR_PER_LINE+MAX_CHAR_PER_NUM]; // provide
  space for EOL unsigned char* outptr = outbuf;
  while ((c = fgetc(fpin)) >= 0) {
    unsigned char* dec = bin2dec[c & 255];
    memcpy(outptr, dec, MAX_CHAR_PER_NUM);
    outptr += dec[MAX_CHAR_PER_NUM];
    if (outptr > &outbuf[ALMOST_FULL_THR]) { // spill output buffer
      *outptr++ = '\n';
      ptrdiff_t wrlen = fwrite(outbuf, 1, outptr-outbuf, fpout);
      if (wrlen != outptr-outbuf) {
        err = 2;
        break;
      }
      outptr = outbuf;
    }
  }
  if (ferror(fpin)) {
    perror(infilename);
    err = 1;
  }
  // last line
  if (outptr != outbuf && err == 0) {
    *outptr++ = '\n';
    ptrdiff_t wrlen = fwrite(outbuf, 1, outptr-outbuf, fpout);
    if (wrlen != outptr-outbuf)
      err = 2;
  }

  // completion and cleanup
  if (err == 2 && outfilename)
    perror(outfilename);

  fclose(fpin);
  if (outfilename) {
    fclose(fpout);
    if (err)
      remove(outfilename);
  }
  return err;
}