Problem Statement

The use of   abbreviations   (also sometimes called synonyms, nicknames, AKAs, or aliases)   can be an easy way to add flexibility when specifying or using commands, sub─commands, options, etc.

It would make a list of words easier to maintain   (as words are added, changed, and/or deleted)   if the minimum abbreviation length of that list could be automatically (programmatically) determined.

For this task, use the list (below) of the days-of-the-week names that are expressed in about a hundred languages   (note that there is a blank line in the list). Caveat:   The list (above) most surely contains errors (or, at the least, differences) of what the actual (or true) names for the days-of-the-week.

To make this Rosetta Code task page as small as possible, if processing the complete list, read the days-of-the-week from a file (that is created from the above list).

Notes concerning the above list of words

Let's start with the solution:

# syntax: GAWK -f ABBREVIATIONS_AUTOMATIC.AWK ABBREVIATIONS_AUTOMATIC.TXT
{ dow_arr[NR] = $0 }
END {
    for (i=1; i<=NR; i++) {
      if (split(dow_arr[i],arr1,FS) != 7) {
        printf("NG %s\n",dow_arr[i])
        continue
      }
      col_width = 0
      for (j=1; j<=7; j++) {
        col_width = max(col_width,length(arr1[j]))
      }
      for (col=1; col<=col_width; col++) {
        delete arr2
        for (j=1; j<=7; j++) {
          arr2[toupper(substr(arr1[j],1,col))]
        }
        if (length(arr2) == 7) {
          break
        }
        if (col >= col_width) { # catches duplicate day names
          col = "NG"
          break
        }
      }
      printf("%2s %s\n",col,dow_arr[i])
    }
    exit(0)
}
function max(x,y) { return((x > y) ? x : y) }