Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!mailrus!cornell!uw-beaver!zephyr.ens.tek.com!tekcrl!tekgvs!toma
From: toma@tekgvs.LABS.TEK.COM (Tom Almy)
Newsgroups: comp.lang.modula2
Subject: Re: TopSpeed M2 WrBin bug fix
Keywords: TopSpeed Modula-2
Message-ID: <7773@tekgvs.LABS.TEK.COM>
Date: 9 Jul 90 20:14:05 GMT
References: <831@massey.ac.nz>
Reply-To: toma@tekgvs.LABS.TEK.COM (Tom Almy)
Organization: Tektronix, Inc., Beaverton,  OR.
Lines: 30

In article <831@massey.ac.nz> SLovatt@massey.ac.nz (S.J. Lovatt) writes:
>Bug Fix - JPI TopSpeed M2 Rel 1.04b
>Symptom:
>When using FIO.WrLn, I found that it was producing: CR CR LF
>... rather than CR LF, as it should.
>Solution:
>The problem was traced to FIO.WrBin, where the procedure CLB.write was
>called to do the actual writing. Elsewhere, this procedure was called as:
>CLB._write     ... but here, it was called as:    CLB.write

This message has me baffled because I recompiled (from older TS M2) an
application that wrote binary files and didn't see this problem.

But my release is "1.04" with no letter suffix. In the sources CLB.write
is called *everywhere* except FIO.Truncate().

The CLB.write() routine checks a text mode bit before converting LF to CR LF.
All of the M2 file opening routines open the file in binary (not text) mode
so this reported problem shouldn't occur. (BTW, that the text mode code
exists in the executable file points out part of the inefficiencies of the
new TS M2. I am impressed that the new M2 runs faster, because I certainly
didn't expect it.)

So this brings up a new question, are there different versions for different
countries? Did they introduce new errors in some revision from 1.04 to 1.04b?
Enquiring minds want to know.

Tom Almy
toma@tekgvs.labs.tek.com
Standard Disclaimers Apply