[Bug] SETUP.EXE regression -- proposing going backward again

[David A. Cobb]

Robert Collins wrote:

>----- Original Message -----
>From: "David A. Cobb" <superbiskit@cox.net>
>To: <cygwin-apps@cygwin.com>
>Sent: Tuesday, July 09, 2002 8:11 AM
>Subject: [Bug] SETUP.EXE regression -- proposing going backward again
>
>
>  
>
>>I see in SETUP.EXE vsn 2.249.2.5 that setup is back to proposing to
>>over-write "test" versions with the "older" current revision.  IIRC, we
>>all agreed once upon a time that this ain't right.
>>
>>Specifically: GCC   3.1.1-2 proposed replacement is 2.95.3-5
>>              Guile 1.5.6-2    ....                 1.4-2
>>              Lilypond 1.5.65-1 ....                1.4.14jcn4-1
>>
>>    
>>
>
>Gcc 3.1.1-2 is marked as test. We agreed that setup would never try to
>uninstall a version, but we didn't agree on management of the test versions.
>
>Rob
>
I'll sift thru the pile, but I believe we agreed that setup would never 
recommend rolling a version-number backward.  All of these do that.

-- 
David A. Cobb, Software Engineer, Public Access Advocate
"By God's Grace I am a Christian man, by my actions a great sinner." -- The Way of a Pilgrim; R. M. French, tr.
Life is too short to tolerate crappy software.
.