develooper Front page | perl.perl5.porters | Postings from June 2021

Re: opsig feature...

mah.kitteh via perl5-porters
June 1, 2021 04:50
Re: opsig feature...
Message ID:

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Monday, May 31, 2021 9:28 PM, L A Walsh <> wrote:

> On 2021/05/24 10:08, Aaron Priven wrote:
> > I agree that sigil-less perl wouldn’t be perl and it wouldn’t make sense to make that a feature of the language.
>     Side note: the base note did NOT propose a sigil-less perl.
>     Since any requirement that sigils not be used would prevent
>     existing perl programs from being used.
> I would probably question my understanding before responding to such
> a proposal.
> However, regarding the Ratfor model (a full C++ 1.0 compiler that
> produced Fortran code), I don't think it would be doable in
> Perl, since perl doesn't have separate compile & execution
> phases. An Opsig perl would need to have all the information
> determined at compile also available at runtime.

Ratfor preceeded C++ by nearly a decade. It was orginally a pre-processor for FORTRAN 66. There might be a C++ implementation now, but it was definitely not the case in 1976.

If there is a modern equivalent, the Ragel State Machine comes to mind. Though this is not a really good since it really just abstracts the construction of the state machines and allows target language code to be injected verbatim.

The problem with sigiless Perl comes down to the disambiguation of:

* data structures
* contexts
* the basis coercion in prototypes

Just to name a few off the top of my head.

Since the sigil is the basis for most things with Perl, I do agree with the statement that "sigil-less Perl is not perl".

Brett Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About