Seed7 as inspiration for Ada202X

Seed7 as inspiration for Ada202X

[Dan'l Miller]

http://seed7.sourceforge.net/faq.htm
As seen in Seed7's FAQ and reference webpages, Seed7 has a variant of some of the bold fresh features that were mentioned by various posters in the recent Ada202X-wishlist thread here on comp.lang.ada.  For example, Seed7 has:
1) types as truly first-class citizens (as objects with stage-n+1 reflection of properties of types generated in stage-n) that can blossom in the next stage of:
2) multistage programming that not only rejects modern C++'s misguided poor-man's emulation of functional programming, but also fundamentally integrates/imbues the concept of a generic/template throughout the rest of the language, which blurs the line between:
3) stage-n interpreter within the stage-n+1 compiler as a compile-time source-code-generator-like presentation of source code to the n+1 stage that not only rejects C's, C++'s, and gnatprep's misguided preprocessor, but Seed7's breadth of compile-time variant selection/generation seems to go so far as to eclipse such Ada's narrower-vision patchwork of discriminants and child-package substitution for tailoring source-code to compile in multiple variant targets or for multiple variant problemspaces.
4) extensible syntax for creating domain-specific problem-space-specific "little languages" (à la OCaml & its p4), as well as a multistage progression of compiler-frontends because:
5) all statements are effectively user-defined syntax (with some basic usual branching constructs provided as a sort of standard library);
6) multiple dispatch.

The worst complaint that I have with Seed7 is that it is not a proper superset of Ada, especially regarding the lack of low-level bit & bit-string declarations and the apparent lack of ability to declare constraints on subtypes.

Re: Seed7 as inspiration for Ada202X

[Yannick Duchêne (Hibou57)]

Funny, it uses a syntax where type name comes first then entity name. Ex.  
`Natural: Width` instead of `Width: Natural`. That's counter intuitive, as  
with most natural languages (at least English and french), the `:` sign  
introduces explanations, details or a definition.

Le Thu, 15 May 2014 15:30:41 +0200, Dan'l Miller <optikos@verizon.net> a  
écrit:

> http://seed7.sourceforge.net/faq.htm
> As seen in Seed7's FAQ and reference webpages, Seed7 has a variant of  
> some of the bold fresh features that were mentioned by various posters  
> in the recent Ada202X-wishlist thread here on comp.lang.ada.  For  
> example, Seed7 has:
> 1) types as truly first-class citizens (as objects with stage-n+1  
> reflection of properties of types generated in stage-n) that can blossom  
> in the next stage of:
> 2) multistage programming that not only rejects modern C++'s misguided  
> poor-man's emulation of functional programming, but also fundamentally  
> integrates/imbues the concept of a generic/template throughout the rest  
> of the language, which blurs the line between:
> 3) stage-n interpreter within the stage-n+1 compiler as a compile-time  
> source-code-generator-like presentation of source code to the n+1 stage  
> that not only rejects C's, C++'s, and gnatprep's misguided preprocessor,  
> but Seed7's breadth of compile-time variant selection/generation seems  
> to go so far as to eclipse such Ada's narrower-vision patchwork of  
> discriminants and child-package substitution for tailoring source-code  
> to compile in multiple variant targets or for multiple variant  
> problemspaces.
> 4) extensible syntax for creating domain-specific problem-space-specific  
> "little languages" (à la OCaml & its p4), as well as a multistage  
> progression of compiler-frontends because:
> 5) all statements are effectively user-defined syntax (with some basic  
> usual branching constructs provided as a sort of standard library);
> 6) multiple dispatch.
>
> The worst complaint that I have with Seed7 is that it is not a proper  
> superset of Ada, especially regarding the lack of low-level bit &  
> bit-string declarations and the apparent lack of ability to declare  
> constraints on subtypes.


-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University