[m-dev.] For review: documentation for solver types

[Peter Moulder]

On Thu, Jul 22, 2004 at 05:11:56PM +1000, Ralph Becket wrote:
> + at node The any inst
> + at subsection The any inst

Suggest "The `any' inst", at least for the subsection.

> +More formally, @samp{X} is ground if for any values @samp{Y} and
> + at samp{Z} that unify with @samp{X}, it is the case that @samp{Y} and
> + at samp{Z} also unify with each other.

`any' is a danger word in definitions: it can mean either forall or
exists.  I initially read it as exists in the above.  Please change `any
values' to `each pair of values' or `all values'.

> + at end example
> +
> +declare types @samp{t1/0} and @samp{t2/2} to be abstract solver types.

Please add `@noindent' on a line by itself before `declare'.

> + at samp{solver_type} values may be represented by
> +ground at samp{representation_type} values (in the context of the

Presumably add a space before `ground' and `@samp{rep...}'.

Btw, I believe @samp{...} adds single quotes around the `...'.  I don't
think this necessary for most uses of solver_type; @code{...} would
suffice.  E.g. I think "@samp{free} @samp{solver_type} variable" will
look strange with all the quotation.

> +Calls to this predicate are inserted automatically by the compiler when
> +a @samp{free} @samp{solver_type} variable has to be converted to @samp{any}
> +(the initialisation predicate is responsible for registering the new,
> +unbound variable with the corresponding constraint solver state).

Use a separate sentence for this parenthesis:

 {any}.  (The ... state.)

> +Under the current design the compiler does not have enough information

Easier to parse with a comma after "design".

> +We intend to lift this
> +restriction with an extension to the current design in the future.

`in the future' is redundant.  (If you disagree, please at least move it
to just after "lift this restriction" rather than let people spend time
exploring the parse possibility "the current design in the future".)

pjrm.
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------