Someone often writes like this when extracting characters one by one from the beginning to the end of a string.
For example:
function rev1(str, i, ret) {
for (i = 1; i <= length(str); i++) {
ret = substr(str, i, 1) ret;
}
return ret;
}
I think don't necessarily have to use the length().
function rev2(str, ch, i, ret) {
while (ch = substr(str, ++i, 1)) {
ret = ch ret;
}
return ret;
}
It works, but is it wrong?
You're not wrong (assuming it's input really is a string and not an array), but the resulting code is less clear since it's relying on the condition doing a string comparison since the return from substr() is a string. At a glance I thought it'd fail if the string contained a 0 but it won't thanks to that string comparison being enforced.
length(str) in every iteration of the loop like in the code you posted, you'd call it once before the loop and set a variable as in n = length(str); for (i=1; i<=n; i++). It's hard to believe calling length() would be heavy and slow given that. Since you mention you're using gawk though you could do n=split(str,arr,""); for (i=1; i<=n; i++) ret = arr[i] ret and so exchange both length() and substr() calls for one call to split(). idk which would be faster but you could try it.
strwas an array, so second function will be fail