[M3devel] A question for our language lawyers

[Daniel Alejandro Benavides D.]

Hi all:
yes, it could be, but VAXen and Alpha's I believe it did not cause the wrong behavior to show that  incorrect initialization at  start time, that most part of it trust on it (Alphas just throw an exception to show that it was changed).
I didn't know it was wrong for sure, but I guess that confirms the initialization code is not working by vicious value initialization.
Did you see the Baby Modula-3 (in p.10 - 11, s 3.1 - Relation to Modula-3) it says you can do overriding at the type level overriding of fields to override defaults? 
Thanks in advance

--- El sáb, 7/7/12, Dirk Muysers <dmuysers at hotmail.com> escribió:

De: Dirk Muysers <dmuysers at hotmail.com>
Asunto: [M3devel] A question for our language lawyers
Para: m3devel at elegosoft.com
Fecha: sábado, 7 de julio, 2012 07:06



 

I reread ParseParams.m3 and, yes, they initialise the 
array of booleans.
One should never trust one's memory, especially past 
a certain age. Yet
I am sure having seen one of the library modules relying 
on zero initialisation.
For my excuse, I never (except an occasional INC, 
where C would use ++)
place two statements on the same line, so when I 
quickly browse through some
code, the second statement often escapes my eyes. 
Nevertheless the
initialisation question was worth to be 
mentionned.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m3lists.elegosoft.com/pipermail/m3devel/attachments/20120707/b332fef3/attachment-0002.html>