Hi nicolas I do not know what are < expr: (foo) results: bar > may be this is even me that tried something (shame on me) but I think that we should clean them. Now I use "foo >>> bar" because we can have non literal and we will start to build tools based on them. I would like to clean the proliferation of pragmas too. Stef On Mon, Nov 20, 2017 at 9:33 PM, Nicolas Cellier <nicolas.cellier.aka.nice@gmail.com> wrote:
Hi, I encountered some annotations in the form:
< expr: (foo) result: bar >
and some meta comments of the form:
" foo >>> bar "
I think both are aimed at providing examples of method usage for documentation, non regression testing, etc...
I note that the former is requiring a literal result, which is quite limiting (1/3) is not a literal for example, 0@0 neither... IMO it should be extended for handling any valid expression. But that still makes two different ways of expressing the same intent. So what's the canonical one?