Re: [Caml-list] what does let () = ... define?

["Christoph Höger" <christoph.hoeger@tu-berlin.de>]

It defines a match-expression with a single case, that is why you get
the warning on None.

In case of let () = ..., everything is fine, since unit has exactly one
constructor (that is, how unit is defined, btw.). You could also have
written let my_unit = ..., but since my_unit = () (even == (), I think),
you do not gain anything from this variable (and hopefully the compiler
gets rid of it). In order to provide more readable and safe code, you
should thus use the ()-syntax (since now everyone, including the
typechecker sees that the rhs is intended to yield unit). The same holds
for records, tuples etc. think of it as a shallow and cheap type annotation.

regards,

Christoph

Am 18.03.2016 um 19:04 schrieb Mr. Herr:
> Hi,
> 
> in a small presentation of OCaml (Linux User Group Bremen) I got some
> interesting questions, and
> trying to answer I noticed I took something for granted that was not
> fully understood.
> 
> Looking at this in the toplevel:
> 
> # let () = () ;;
> # () ;;
> - : unit = ()
> # let _ = () ;;
> - : unit = ()
> # let None = None;;
> Warning 8: this pattern-matching is not exhaustive.
> Here is an example of a value that is not matched:
> Some _
> #
> 
> ... the question is: okay, pattern matching left of the equal sign, but
> what does it define?
> It defines unit and None in the environment, and then that value is just
> sitting there?
> 
> /Str.


-- 
Christoph Höger

Technische Universität Berlin
Fakultät IV - Elektrotechnik und Informatik
Übersetzerbau und Programmiersprachen

Sekr. TEL12-2, Ernst-Reuter-Platz 7, 10587 Berlin

Tel.: +49 (30) 314-24890
E-Mail: christoph.hoeger@tu-berlin.de