[kepler-dev] FYI: Dataflow Exchange Language

[Shawn Bowers]

Two comments/questions below.

On Thu, 23 Sep 2004, Bertram Ludaescher wrote:

>
> FYI:
> I think a language like the following would be suitable for some of
> our static WF analysis tools (yet to be developed ;-), including
> semantic type analysis, QPN optimization, and others.

I briefly skimmed through the paper -- but it doesn't appear there is any
support in the language for types. For example, assigning input or output
types like string, list of double, etc. Is this true?

>
> Best of all: I think the language has "0-cost" implementation by means
> of a Prolog operator precedence grammar, i.e., a bunch of ':- op(...)'
> declarations will do and we have a parser.