davecb@nexus.yorku.ca (David Collier-Brown) (08/27/90)
I've just been reminded of the problems of minimality by a discussion
about 4GLs: I wonder how few separate constructs are sufficient for a
Problem-Oriented Language for describing and manipulating screens.
I used to work with Uniface(tm), a front-end product from Amsterdam, which
had assignment, comparison, if-then-else and call-whole-screen-routine as
its primitives, and did reasonably well.
I'm looking at writing a program that is defined narrowly as a manipulator
of windows and subwindows, and wonder if I'm missing anything.
I propose a substrate that already looks like
{subwindow,event} => action
and am considering the constructs I need for ``action '. I think that
assignment to variables and fields
comparison of variables/fields
if-then-else
call entire named window with parameters and return value
call named C routine with parameters and return value
are what I need in the mini-language, and I need to add a group
of field/subwindow mini-editors to the environment.
I wonder if I need substring and sed-like constructs...
Who can comment on how small a 4GL I can get away with...
--dave
--
David Collier-Brown, | davecb@Nexus.YorkU.CA, ...!yunexus!davecb or
72 Abitibi Ave., | {toronto area...}lethe!dave
Willowdale, Ontario, |
CANADA. 416-223-8968 |
--
Send compilers articles to compilers@esegue.segue.boston.ma.us
{ima | spdcc | world}!esegue. Meta-mail to compilers-request@esegue.