Add strings.ncompare (and backing strncmp_mem) to compare up to n characters of a string.

[gillham]

Compares up to 'length' characters of the two strings. This allows strings of different lengths to be compared.

For example:
strings.ncompare('abc', 'abcdefg', 3) would only look at the first 3 letters and match (returning 0).

Currently with strings.compare strings of unequal length cannot match.

[irmen]

What is your use case if I may ask?

[gillham]

My simplest use case is when accepting commands with a "shell" or even say a text adventure game.
Below is a simple example, but basically to accept a less than the full command (just the unique length), it is currently necessary to do something like parse1 below. If de is entered for delete it is necessary to copy the first two characters of each command to a temp string, then do the comparison.
(or write your own comparison of course)

The strncmp style feature just lets me match the first N characters directly against the valid command string.

The same thing for a text adventure, 'g n' could be valid for 'go north' or similar.

Maybe there is a more obvious way to do this that I am missing, I am just used to strncmp from C.

%zeropage basicsafe
%import strings
%import textio

main {
    str[5] commands = ["catalog", "delete", "directory", "show", "type"]
    ubyte[5] unique = [1, 2, 2, 1, 1]
    str cmd = " " * 64
    str temp = " " * 64
    sub start() {
        void txt.input_chars(cmd)
        txt.nl()

        parse1()
        txt.nl()
        parse2()
        txt.nl()
    }

    sub parse1() {
        ubyte i
        for i in 0 to 4 {
            strings.left(commands[i], unique[i], temp)
            if strings.compare(cmd, temp) == 0 {
                txt.print("Valid command: ")
                txt.print(commands[i])
                }
        }
    }

    sub parse2() {
        ubyte i
        for i in 0 to 4 {
            if strings.ncompare(cmd, commands[i], unique[i]) == 0 {
                txt.print("Valid command: ")
                txt.print(commands[i])
                }
        }
    }

}

[irmen]

Looks solid. Just 1 missed spot; the new routine should get a mention in the libraries.rst documentation chapter as well. There's a section in there detailing the "string manipulation routines".

[gillham]

Ok, let me look at that.

[irmen]

thank you