Tom Pollard <pollard@schrodinger.com> writes: >On Mon, 16 Jun 1997, Paul J. Schinder wrote: >> In fact, I'd love for someone to define "scripting language". The >> definition would have to be pretty convoluted to exclude MacPerl, IMHO. > Not at all. To me, a scripting language is a language in which you can > naturally specify a sequence of actions, that you would ordinarily execute > 'by hand', in a 'script'. [...] Because the Mac has > no standard command-line interface (it's graphically oriented), these > models have to be modified for the Mac. There, AppleScript and Frontier > are real scripting languages for the Macintosh because they let you > control AppleScriptable applications in a way that's very closely tied to > their standard gui interfaces. That's a plausible argument, but if I remember Frontier correctly, high quality control of applications is sometimes achieved by writing custom "glue tables", i.e. libraries. So in that respect, I don't see why MacPerl should be any less of a scripting language when working with applications such as MacPGP or Anarchie, for which Chris Nandor has written interface libraries. It is true, though, that MacPerl currently controls very few applications as conveniently as this, because it lacks terminology tools and mapping. However, I see no *syntactic* or *semantic* reasons that would prevent MacPerl from being as much of a Mac scripting language as Frontier is. AppleScript is willing to have its syntax determined by the terminology resources, which is not something that MacPerl can reproduce. >Under this definition, MacPerl is no more a scripting language than C is. >You can just as easily concoct a "DoAppleScript" function in C as in >MacPerl. Except that C is not interpreted on the Mac. Interpretation or very fast compilation to me is another essential property defining a "scripting language". >In both cases it's just as foreign to the basic language. >Contrast this with Unix shells and Tcl. Shell scripts have very frequently embedded sed, awk, perl, or m4 scripts, which are essentially the same technique of embedding different languages to achieve special effects. >PS: I'm amazed that Matthias hasn't reprimanded me for starting this >discussion. I never *reprimand* participants in such discussions, I merely *remind* them of the purpose of the list (or my interpretation thereof) :-). As for the present discussion, it is valuable to me in shaping ideas about improving AppleEvent support, which ultimately might be the core competency of MacPerl to migrate into a yellow box rhapsody perl. Thus, I think the discussion still has some usefulness left in it if participants are willing to keep some discipline. Matthias ----- Matthias Neeracher <neeri@iis.ee.ethz.ch> http://www.iis.ee.ethz.ch/~neeri "This comment reminds me of the APL93 meeting where people declared APL to be a thriving, growing language. It's like going to a wake where only the next-of-kin don't seem to be aware that the guest of honor is actually dead." -- Richard J. Gaylord <gaylord-100993020849@mm-mac18.mse.uiuc.edu> ***** Want to unsubscribe from this list? ***** Send mail with body "unsubscribe" to mac-perl-request@iis.ee.ethz.ch