Ahh.  Yeah, I see it now -- the output wasn't a string, but a float.  It just has the word "string"
in the type name to confuse me.

~~ Robert.

DooMeeR wrote:
> Robert Fischer a écrit :
>> How is the compiler magically getting from the float to a string
>> value?  Can someone break down
>> what's actually happening here for me?
> 
> It's not "getting to" a "string" value but to a "string t" value, which
> is an abbreviation for "float". The value has type "float", but the user
> requires type "string t", so the compiler unifies "float" and "string t"
> which is actually "float" so it's ok, and then the compiler gives the
> type requested by the user to the value, i.e., "string t".
> 
> Well, at least, this is how I understand this :)
>