Re: Jinja2 syntax for conditionals

[Sam Thursfield <sam thursfield codethink co uk>]

On 21/09/17 04:27, Tristan Van Berkom wrote:
> Warning: This email may be straying a bit off topic...

Agreed! I decided not to reply last week, but hopefully with the dust 
settled a bit I can be forgiven for one last attempt to work out where 
our thinking diverges...

> > I don't see this as the question at all. Either way we're adding logical
> > expressions to what was previously a serialization format, and making
> > something that is (a) a serialization format with syntax sugar for
> > variants, or (b) something else.
>
> Probably late to reply this, and probably veering off topic, but I re-
> read this while replying to Sander and feel that the above is an
> inaccurate assessment.
>
> We already stray from pure YAML in a couple of places for our
> serialization, for instance we parse out urls expressed as aliases and
> split on the ':' character. I dont think this causes the format to no
> longer be a serialization format.
>
> The existing 'arches' and 'variants' are logical expressions as much as
> they would be if expressed a bit differently for the user's
> convenience.
>
> I think you are making a deep association of S-Expressions themselves
> with the lisp programming language, while I on the other hand see them
> as an merely an easy to parse, more structured and defined approach to
> CSV (it sits somewhere in between CSV and JSON) - sending S-Expressions
> over a wire for instance is quite typical and I feel is akin to restful
> JSON apis, these are just simple data serialization mechanisms, one of
> which just happens to be used as a vehicle for some simple but turing
> complete programming languages.

I agree with this. S-Expressions just represent lists of atoms and I 
consider them to be a data format. Lisp is a family of programming 
languages where lists of atoms are interpreted as programs -- the first 
item in the list is the name of a function which the interpreter will 
execute and will pass the rest of the list as arguments. (It's not 
always that simple, but you know what I mean).

In the example that you attached earlier on in this thread, you had a 
200 line program which would parse S-Expressions into a list and then 
execute a function based on the first item in the list. To me, that's 
already starting to look like a minimal, non-Turing-complete programming 
language.

Sam


--
Sam Thursfield, Codethink Ltd.
Office telephone: +44 161 236 5575