Re: Tail call elimination

[Tim Murphy]

A question would be do we want to use GMP or are 64 bit ints enough? I'm inclined to say ints are ok of they are wide. 

Sorry to jump to such a basic question. I sort of think it helps frame a discussion about what people really want.

I mostly needed addition, comparison and decrement of values that might be as large as a file or a rim image at worst.

So for me an add function that took signed numbers would be enough if it came with < > = since we already have "or" and "and". e.g. X:=$(add $(X),-1) gives you decrement.

If one is minimal then there is less chance of creating a carbuncle.

Regards,

Tim

Regards,

Tim

On Tue, 19 May 2020, 14:26 Paul Smith, <address@hidden> wrote:

On Mon, 2020-05-18 at 21:40 -0400, Daniel Herring wrote:
> This decision causes a difficult and error-prone ambiguity when the
> return value is really true and empty.  For example, the operation
> succeeded and the result was "", versus the operation failed and thus
> returned "".  So Scheme added separate #t and #f values, C++ added
> true and false keywords, newer languages are adding an
> "optional/maybe" construct, etc.
>
> The present Make behavior is perfect for things like dependency
> lists. However, it feels limiting for other use cases.

I would prefer to talk about concrete issues so we can consider them.
I personally haven't run into situations where the fact that "" equates
to false is a problem but obviously others may have very different
experiences.

I'm really reluctant to start adding new higher-level language features
to GNU make unless there is a clear and obvious need that can't be
reasonably addressed any other way.