Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!snorkelwacker!bloom-beacon!eru!hagbard!sunic!kth.se!cyklop.nada.kth.se!krona
From: krona@nada.kth.se (Kjell Krona)
Newsgroups: comp.sys.mac.hypercard
Subject: Re: WindowIt! and CompileIt!
Message-ID: <1990Sep14.184246.17989@nada.kth.se>
Date: 14 Sep 90 18:42:46 GMT
References: <90254.173048KPURCELL@LIVERPOOL.AC.UK> <57391@microsoft.UUCP>
Sender: krona@nada.kth.s
Organization: Royal Institute of Technology, Stockholm, Sweden
Lines: 24

I have been using CompileIt! 1.5 for a couple of months, and I have to
agree about the slow compilation times. This would be acceptable if all
HyperTalk code could be reliable translated, but as pointed out this is
not always so. A few weeks ago I came across an astonishing bug:

In an "if" statement, a ">" (I think) comparison reliable turned out an
erroneus result; however, if I subtracted the values and compared to 0,
it worked as expected! Of course, either way worked in HyperCard. This 
took me about half a day to figure out, thinking something was wrong
with the parameter passing; you do not expect such a "simple" fault..

However, if the time of the script is critical - and it often is, HC 2.0
notwithstanding - it is still much faster than writing an XCMD or XFCN
from scratch. It would certainly be nice if it was faster and more
reliable, but I still think it is well worth its price for any serious
hyperCard programmer.

(I would like to add that the bug described above has only surfaced on
one occassion, and I have not had the time to check if it is reproducible
in every script. I was quite fed up at the time -:))

Kjell Krona                      krona@nada.kth.se
Dept. of Architecture/Dept. of Numerical Analysis and Computer Science
S-100 44 Stockholm, Sweden