Bug 29277, discussion today, SC and DC in dynamic function invocations

> Action A-628-01: Jonathan to enter the following modification to
> MDyck's proposal, into the Bug 29277 and ask Abel / Mike whether this
> resolves the bug:

I think that with whatever text we end up with, the following becomes clear(er):

1) Difference between / definition of external functions and other functions
2) Clear definition of when we talk about the (dyn) function declaration (static) and the function item (use, dynamic)
3) Whether the location of the declaration defines the SC and the DC, or whether the target (use) defines these
4) What it closes over: dynamic parts inside the function, will they use DC of target site, or DC of declaration site (in other words, do they carry the dyn context with them, or not, or is this defined in the other specs)

And another thought: the definition how inline functions are described is pretty clear (although complex), by creating a wrapper function around the actual internal function. Perhaps we can use a similar approach here, where we define that a dynamic function is in fact a new function with the same signature and certain properties that define its DC and SC and that in its body calls the dynamic function (has as body the function implementation).

I am at this moment unsure whether adding "The processor "makes the information available"" covers this. It feels like it only moves the issue to a different point, but doesn't answer the core questions.

Cheers,
Abel

Received on Tuesday, 12 January 2016 15:42:55 UTC