DOS ain't dead

Or, you could write a small program to do this for you. Compile this:

#include <stdio.h>

#define is_upper(A) ((A >= 'A') && (A <= 'Z'))
#define is_lower(A) ((A >= 'a') && (A <= 'z'))

#define to_lower(A) (A - 'A' + 'a')
#define to_upper(A) (A - 'a' + 'A')

int
main()
{
  int ch;

  while ((ch = getchar()) != EOF) {
    if (is_upper(ch)) {
      putchar( to_lower(ch) );
    }
    else {
      putchar(ch);
    }
  }

  return 0;
}

This is a very simple program (it doesn't read files, it relies on file redirection) to convert text from uppercase to lowercase. It leaves all other characters as they are (such as punctuation, numbers, etc.)

Save it as tolower.c and compile it with

> WCL tolower.c

(You'll need to install the Watcom C Compiler package and run the OWSETENV.BAT file to set your environment.)

To turn this program into a "convert to uppercase" program, copy the file to toupper.c and change the two lines inside the while loop to this:

    if (is_lower(ch)) {
      putchar( to_upper(ch) );

Then compile the new program with:

> WCL toupper.c

> But I didn't find a tolower() function. My guess: If the user supplied,
> e.g., lowercase "bäh" on the command line, DOS/command.com had to look for
> files like BÄH.EXE or BÄH.BAT, because uppercase is how filenames are
> stored on (FAT) disks. There simply was no need for a tolower() function.

Exactly. The DOS kernel really doesn't have a need for a tolower() function.

> Oh, INT 21 -
> DOS 3.3+ - GET EXTENDED COUNTRY INFORMATION reads "03h (DOS 6.2+
> COUNTRY.SYS) get pointer to lowercase table".

That is correct, but the contents of that table depend on the current Code Page. E.g., the Code Page I normally use is 437, but I have played around with some of the other Code Pages as well due to the nature of some of my programs (both released and unreleased/experimental). I think for at least most of the Code Pages, there are both lower-case and upper-case of the Latin-based letters that are in the Code Page. In such cases, you can go back and forth between lower- and upper-case if you need to.

But Code Page 437 also has some Greek letters (e.g., Alpha and Beta and Mu and Omega) but only has one of the two cases in the Code Page. So, even though those letters have case-equivalents you may never know it based on what DOS can tell you. DOS can't do a case-conversion unless there is something to display on the screen, and that depends on the Code Page.