Bug 51113

Summary: /usr/ports/lang/ruby build failure; sed?
Product: Ports & Packages Reporter: cswiger <cswiger>
Component: Individual Port(s)Assignee: Akinori MUSHA <knu>
Status: Closed FIXED    
Severity: Affects Only Me    
Priority: Normal    
Version: Latest   
Hardware: Any   
OS: Any   

Description cswiger 2003-04-18 01:50:08 UTC
155-sec# make                                              /usr/ports/lang/ruby
FreeBSD: not found
"/usr/ports/Mk/bsd.port.mk", line 772: warning: "FreeBSD -p" returned non-zero s
tatus
FreeBSD: not found
"/usr/ports/Mk/bsd.port.mk", line 780: warning: "FreeBSD -s" returned non-zero s
tatus
FreeBSD: not found
===>  Extracting for ruby-1.6.8.2003.01.19

Fix: 

Unknown.  Perhaps look into using autoconf/gmake/GNU sed, etc...

-Chuck
How-To-Repeat: 
cd /usr/ports/lang/ruby
make
Comment 1 Edwin Groothuis freebsd_committer freebsd_triage 2003-04-18 08:26:33 UTC
Responsible Changed
From-To: freebsd-ports-bugs->knu

Over to maintainer.
Comment 2 Akinori MUSHA freebsd_committer freebsd_triage 2003-04-23 07:25:53 UTC
State Changed
From-To: open->feedback

Can you reproduce this with the latest ports tree and a vanilla 
make.conf?  I can hardly believe that this has anything to do 
with ruby or sed. 

Those messages suggest that you have somehow UNAME defined as 
"FreeBSD" somewhere...
Comment 3 chuck 2003-04-23 15:21:00 UTC
A quick test suggests that unsetting $UNAME in my environment does allow 
ruby to build.  To the extent that this PR represents "yes, one could 
shoot oneself in the foot", it can be closed.  :-)

-Chuck

PS: To the extent that this may represent a legitimate issue with 
documentation, or a lack of error checking of user-configured env 
variables passed to the ports build system, perhaps there is something 
to consider.
Comment 4 Edwin Groothuis freebsd_committer freebsd_triage 2003-10-10 12:51:59 UTC
State Changed
From-To: feedback->closed

User error, user had a variable called UNAME in its environment 
which interfered with the definition of it in Mk/bsd.port.mk