Fri, 01 Jun 2001 18:36:01 +1000, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:
I like it. Minor issues: - import is not a specialid, it's already a reservedid. - Safeness should be optional. (And maybe s/safeness/safety/ ?) - Would anyone really use explicit "static"? -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTÊPCZA QRCZAK
Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> writes:
Would anyone really use explicit "static"?
I would. Even if rarely used, I think it is easier to teach people to use it if we say there are two forms of definition (safe and unsafe, static and dynamic, etc.) and then later teach them that one of them is the default if they say nothing. It also provides people with the right vocabulary for distinguishing two variants. With the old ffi, we had "foreign export dynamic" and "foreign export" but "foreign export" is ambiguous because you might want to use it to include "foreign export dynamic" or you might want to exclude "foreign export dynamic". We all solved that problem by talking about "foreign export static" but this was a slightly funny construction because you couldn't write "static" in actual ffi code. -- Alastair Reid reid@cs.utah.edu http://www.cs.utah.edu/~reid/
"Marcin 'Qrczak' Kowalczyk" <qrczak@knm.org.pl> wrote,
Fri, 01 Jun 2001 18:36:01 +1000, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:
I like it.
Fine :-)
Minor issues: - import is not a specialid, it's already a reservedid. - Safeness should be optional. (And maybe s/safeness/safety/ ?)
Ok, I changed all that.
- Would anyone really use explicit "static"?
Yes, because when you want to import the C identifier `dynamic' or `wrapper' and not provide a header file. Moreover, I think, it is more orthogonal. Cheers, Manuel
participants (3)
-
Alastair David Reid -
Manuel M. T. Chakravarty -
Marcin 'Qrczak' Kowalczyk