It neednt necessarily be couched in terms of some form of declare to a library. My suggestion in the FR is that we uses an existing RB function as the "template" for the arguments and return value. Thus no soft declare would be needed, although I agree that something along the lines of a soft declare could also be used ( minus the library path bit since this is not needed )
I think is simply a case of taking a memory address ( pointer ) , preparing the stack and CPU registers etc with all the parameters necessary for doing a C style call, and then issuing a "jump". On 13/4/07 03:17, "Norman Palardy" <[EMAIL PROTECTED]> wrote: > > On Apr 12, 2007, at 5:25 PM, Daniel Stenning wrote: > >> Hence the FR. >> >> Even 4D has had this ability for well over a decade and they proved >> extremely useful ( even though 4D sucks these days imo ) so it cant >> be that >> difficult to implement. > > Being able to load and call a routine from a library with a "soft > declare" where the library is named in a variable should not be. > I think the hold up is what are the semantics and scope ? > > something like > > dim s as string > > s = f.absolutepath (or whatever) > > soft declare someRoutine lib s .... > > s = someOtherPath > > call someRoutine > > is the routine set as being in the library s named when the soft > declare was encountered ? or the library that is named by s when it > is finally called ? > > that's at least one issue I know of > > other than that loading and calling a method should be possible > _______________________________________________ > Unsubscribe or switch delivery mode: > <http://www.realsoftware.com/support/listmanager/> > > Search the archives: > <http://support.realsoftware.com/listarchives/lists.html> > Regards, Dan _______________________________________________ Unsubscribe or switch delivery mode: <http://www.realsoftware.com/support/listmanager/> Search the archives: <http://support.realsoftware.com/listarchives/lists.html>
