In c, there's this very nice function strncmp(s1,s2,count) which compares two c strings, using at most count characters.  count can be less than, more than, or equal to either or both of the lengths of the two strings.  It can be used to see if two c-strings have the same prefix of some length.

Now, strncmp indeed seems to be packaged up in core.stdc.string, but i would like to use some something like it on 2 d strings (which, as i understand it, need not be zero-terminated).

I suppose it would be possible to do some conversion with toStringz() or something and then invoke the strncmp(), but that seems very wordy and also it's not clear that it would handle all pairs of d strings (e.g., what if there were some 0's in the first count characters?).

So i would like to call a d function which works on d strings, but don't want to write my own if one already exists.  (At the very least, i'd have to get a much sharper understanding of d strings, whether internal 0's can occur, etc.  And i would not want to do egregious string allocation.)

TIA for any info!

int strncmp(string a, string b, int n) {
	if(a.length > n)
		a = a[0 .. n];
	if(b.length > n)
		b = b[0 .. n];
	import std.algorithm.comparison : cmp;
	return cmp(a, b);
}

On Monday, August 22, 2016 00:14:31 Adam D. Ruppe via Digitalmars-d-learn wrote:
> int strncmp(string a, string b, int n) {
>   if(a.length > n)
>       a = a[0 .. n];
>   if(b.length > n)
>       b = b[0 .. n];
>   import std.algorithm.comparison : cmp;
>   return cmp(a, b);
> }

Aside from the imports, it can be turned into a one-liner if you use take:

return cmp(take(a, n), take(b, n));

- Jonathan M Davis

On Monday, 22 August 2016 at 01:45:02 UTC, Jonathan M Davis wrote:
> On Monday, August 22, 2016 00:14:31 Adam D. Ruppe via Digitalmars-d-learn wrote:
>> int strncmp(string a, string b, int n) {
>>   if(a.length > n)
>>       a = a[0 .. n];
>>   if(b.length > n)
>>       b = b[0 .. n];
>>   import std.algorithm.comparison : cmp;
>>   return cmp(a, b);
>> }
>
> Aside from the imports, it can be turned into a one-liner if you use take:
>
> return cmp(take(a, n), take(b, n));
>
> - Jonathan M Davis

Thanks Adam and Jonathan for your solutions.

For reference, one of the imports Jonathan is referring to is
   import std.range;

I did not know about take.  Well, i also did not know about cmp.  So my code is probably not very idiomatic.  But i do appreciate all of you d-learn people!

import std.algorithm.searching: startsWith, commonPrefix;
if(s1.startsWith(s2)) {...}
string prefix = commonPrefix(s1,s2);