DomainKey-Signature: a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id
	:list-unsubscribe:list-subscribe:list-archive:list-post
	:list-help:sender:message-id:date:from:mime-version:to:subject
	:content-type:content-transfer-encoding; q=dns; s=default; b=tnt
	7btKTwzsZ5d05cGAM5mkCqNXujThdUi4EFQOTDbn26vrY5fiFvU3kk7gUoOmAFci
	oaK3iqNTwn12BO76+BryEhqMSPhU3o74S9xNkcn6gxwDzaTIb2g4Sn/trc1AR7xT
	6QuTnLTUgCCUqTcZ5DY5QChj9z+eJVacAYj4AqiY=
Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm
Sender: cygwin-owner@cygwin.com
Mail-Followup-To: cygwin@cygwin.com
Message-ID: <53726A51.3010105@breisch.org>
Date: Tue, 13 May 2014 14:54:09 -0400
From: "Chris J. Breisch" <chris.ml@breisch.org>
User-Agent: Postbox 3.0.9 (Windows/20140128)
MIME-Version: 1.0
To: "cygwin@cygwin.com" <cygwin@cygwin.com>
Subject: Does /etc/profile need to set MANPATH?
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

I've been working on getting man-db in a state where I can submit a 
package for it. I had been baffled by the failure of one of the tests 
that was dealing with overriding the man_db.conf file, so had been 
struggling to get a version that passes everything before I submitted it.

I finally discovered that the root cause was that MANPATH was being set 
from /etc/profile and that this value was conflicting with the test and 
causing it to fail. Unsetting MANPATH allowed the test to succeed[1].

My question is then, is it necessary to set this value? Is it something 
required by the current version of man that we're using? If so, man-db 
seems to override this necessity as it makes considerable effort to work 
out a reasonable manpath on it's own.

Perhaps this discussion belongs in cygwin-apps. I'll be happy to take it 
elsewhere.

-----------
[1] The failure of this test also seems to me to indicate an upstream 
bug. If MANPATH is set, man-db always ignores the data in the 
man_db.conf file, even if the conf file is specified on the command 
line. Worse, man-db actually reads the conf file, but then ignores the 
data. Perhaps this is working as designed, but if so, the design seems 
flawed from my perspective. From my research, this does not appear to be 
a Cygwin specific issue, although I haven't yet tried to reproduce it on 
a Linux system.


-- 
Chris J. Breisch

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple