NetBSD Problem Report #23017

Received: (qmail 13796 invoked by uid 605); 30 Sep 2003 19:19:10 -0000
Message-Id: <6164-1064949544@rainier.reedmedia.net>
Date: Tue, 30 Sep 2003 12:19:04 -0700
From: reed@reedmedia.net
Sender: gnats-bugs-owner@NetBSD.org
Reply-To: reed@reedmedia.net
To: gnats-bugs@gnats.netbsd.org
Subject: pkgsrc/sysutils/fam mntent.h include nested too deeply
X-Send-Pr-Version: 3.95

>Number:         23017
>Category:       pkg
>Synopsis:       pkgsrc/sysutils/fam mntent.h include nested too deeply
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    pkg-manager
>State:          closed
>Class:          change-request
>Submitter-Id:   net
>Arrival-Date:   Tue Sep 30 19:20:00 +0000 2003
>Closed-Date:    Wed Mar 31 19:16:31 +0000 2004
>Last-Modified:  Wed Mar 31 19:16:31 +0000 2004
>Originator:     
>Release:        NetBSD 1.6.1_STABLE
>Organization:
http://bsd.reedmedia.net/
>Environment:


System: NetBSD rainier.reedmedia.net 1.6.1_STABLE NetBSD 1.6.1_STABLE (GENERIC) #0: Tue Aug 12 02:52:57 PDT 2003 reed@rainier.reedmedia.net:/usr/src/sys/arch/i386/compile/GENERIC i386
Architecture: i386
Machine: i386
>Description:
sysutils/fam build failed with:
mntent.h:31:10: #include nested too deeply

In file included from mntent.h:31,
                 from mntent.h:31,
                 from mntent.h:31,
		... and around 195 times ...

This is cpp (GCC) 3.3.1 as installed from gcc3-c-3.3.1.
>How-To-Repeat:

>Fix:
patches/patch-ao provides fam/mntent.h which has:

#if defined(HAVE_MNTENT_H)
#include <mntent.h>
#else

It is reloading itself there.
>Release-Note:
>Audit-Trail:

From: "Jeremy C. Reed" <reed@reedmedia.net>
To: gnats-bugs@gnats.netbsd.org
Cc:  
Subject: Re: pkg/23017: pkgsrc/sysutils/fam mntent.h include nested too deeply
Date: Tue, 30 Sep 2003 12:23:17 -0700 (PDT)

 I forgot to mention this is under Linux and my glibc-2.3.2 also provides
 /usr/include/mntent.h

    Jeremy C. Reed


From: "Jeremy C. Reed" <reed@reedmedia.net>
To: gnats-bugs@gnats.netbsd.org
Cc:  
Subject: Re: pkg/23017: pkgsrc/sysutils/fam mntent.h include nested too 
Date: Tue, 30 Sep 2003 13:03:46 -0700 (PDT)

 More information:

 The LOCALBASE is /usr. The include path must be setup to use local
 directory first. My workaround was to force the include to be
 "/usr/include/mntent.h".

 Please note that I have built and successfully used hundreds of packages
 under Linux for the past 12 months using LOCALBASE as /usr.

 Since my standard libc headers are in /usr/include they are normally
 getting used correctly.

 The fix for this fam may be one of the following:

 1 fix the patches so that they check HAVE_MNTENT_H before using the
 patch-provided mntent.h

 2 rename the patch-provided mntent.h to fam-mntent.h and use that name

 3 or fix the buildlink2 wrappers/sed expressions to allow /usr/include
 first (probably not a good idea for me, since I want it to use buildlink
 structure properly for me)

 4 or I provide a libc buildlink2.mk file (I should do that anyways.)

 Let me know if I can do 1 or 2 above and I will provide a patch. Thank
 you.

    Jeremy C. Reed
    http://bsd.reedmedia.net/


From: "Jeremy C. Reed" <reed@reedmedia.net>
To: gnats-bugs@gnats.netbsd.org
Cc:  
Subject: pkg/23017
Date: Fri, 10 Oct 2003 11:43:57 -0700 (PDT)

 This is sysutils/fam/patches/patch-ad
 $NetBSD$

 --- fam/NFSFileSystem.c++.orig	Mon Jul 15 22:23:24 2002
 +++ fam/NFSFileSystem.c++
 @@ -24,7 +24,7 @@
  #include "NFSFileSystem.h"

  #include <assert.h>
 -#include <mntent.h>
 +#include "fam-mntent.h"
  #include <stdlib.h>
  #include <stdio.h>
  #include <string.h>

 And more patches ...

 Index: sysutils/fam/distinfo
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/distinfo,v
 retrieving revision 1.5
 diff -b -u -r1.5 distinfo
 --- sysutils/fam/distinfo	2002/12/17 19:43:39	1.5
 +++ sysutils/fam/distinfo	2003/10/10 18:37:04
 @@ -5,19 +5,20 @@
  SHA1 (patch-aa) = 6aa9a0ecda1d51ffa3e4ac2a45c02a9e629bb8bd
  SHA1 (patch-ab) = 60897dc25f7e18f584cba0c6d22538b1d0b11ecf
  SHA1 (patch-ac) = 5e76e34ebb14e9db2c76f5bdcbfd31cb70991d70
 -SHA1 (patch-ae) = e07e52e8011e2f22e0391d33764a223e6ac769a4
 -SHA1 (patch-af) = c982b3e2ada3274881990c5ea183ca692ff543fe
 +SHA1 (patch-ad) = 223f08e85d541b3ee96bf4104c267f37f5c345f3
 +SHA1 (patch-ae) = a6a686d175ef3b82531a721f0375aee39237d72e
 +SHA1 (patch-af) = 572b8aa1760b055a86e7046910ed58631d8b15d9
  SHA1 (patch-ag) = 059eb7191fc03765e0840a0faed0b27eb6919b74
  SHA1 (patch-ah) = a31f816d61127636663d16ad8e87169715661e3c
  SHA1 (patch-ai) = 62e013a8559aa5207085d80ed0bb77e005a4788e
  SHA1 (patch-aj) = c442ff2fd4afc1e011953b421d20f1adc9e92b06
  SHA1 (patch-ak) = dc18202214fc0ce153ecbab2a4f7759e32d9e11f
 -SHA1 (patch-al) = dee33dd3c235be568b35a57a34f6ca703bc8f411
 -SHA1 (patch-an) = 8416a0de32fec6cbf615083d44fd0b5add5cd8e3
 -SHA1 (patch-ao) = aa9e0cd237460004d4e1f89a3f4bced11ee620b7
 -SHA1 (patch-ap) = 75ca4b6a32f1c3ed1a085457b4de9bb228242fad
 +SHA1 (patch-al) = 147c37fa689730680a2ce1e3d90d18ae1a09317f
 +SHA1 (patch-an) = a79e11cd0aca300b328f2de356580423b0cdab02
 +SHA1 (patch-ao) = a94b163691bfba8c548ea03f2c937787febfc225
 +SHA1 (patch-ap) = 7ba84b1331291f972f2494a51f3f81ef5dc27082
  SHA1 (patch-at) = 83f5680d43ad0e06e5eca517cbba2d0866ae42bc
 -SHA1 (patch-au) = f6ba3953b892f40158e6e4bde5d8b28bf930f241
 +SHA1 (patch-au) = a8922a4e96321d20c74d61cbcee737796690518a
  SHA1 (patch-av) = 18a90c53d641305ef02f71bf443c84d50a2c3fff
  SHA1 (patch-aw) = a6b75b6c2b85d1b8247cee16fd23fbaffa346c05
  SHA1 (patch-ax) = 4f18da00e202b397615588480b78e4dc4248cc23
 Index: sysutils/fam/patches/patch-ae
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-ae,v
 retrieving revision 1.2
 diff -b -u -r1.2 patch-ae
 --- sysutils/fam/patches/patch-ae	2002/08/25 19:23:21	1.2
 +++ sysutils/fam/patches/patch-ae	2003/10/10 18:37:04
 @@ -7,7 +7,7 @@
   #include "FileSystem.h"

  -#include <mntent.h>
 -+#include "mntent.h"
 ++#include "fam-mntent.h"
   #include <string.h>

   #include "Event.h"
 Index: sysutils/fam/patches/patch-af
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-af,v
 retrieving revision 1.2
 diff -b -u -r1.2 patch-af
 --- sysutils/fam/patches/patch-af	2002/08/25 19:23:21	1.2
 +++ sysutils/fam/patches/patch-af	2003/10/10 18:37:05
 @@ -10,7 +10,7 @@
   #include "FileSystemTable.h"

  -#include <mntent.h>
 -+#include "mntent.h"
 ++#include "fam-mntent.h"
   #include <stdlib.h>
   #include <string.h>
  +#include <sys/param.h>
 Index: sysutils/fam/patches/patch-al
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-al,v
 retrieving revision 1.2
 diff -b -u -r1.2 patch-al
 --- sysutils/fam/patches/patch-al	2002/08/25 19:23:21	1.2
 +++ sysutils/fam/patches/patch-al	2003/10/10 18:37:05
 @@ -6,7 +6,7 @@
     main.c++ \
     timeval.c++ \
     timeval.h \
 -+  mntent_compat.c++ \
 ++  fam-mntent_compat.c++ \
     @IMON_FUNCS@.c++

   EXTRA_fam_SOURCES = IMonIrix.c++ IMonLinux.c++ IMonNone.c++
 Index: sysutils/fam/patches/patch-an
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-an,v
 retrieving revision 1.2
 diff -b -u -r1.2 patch-an
 --- sysutils/fam/patches/patch-an	2002/08/25 19:23:21	1.2
 +++ sysutils/fam/patches/patch-an	2003/10/10 18:37:05
 @@ -6,7 +6,7 @@

   #include <assert.h>
   #include <string.h>
 -+#include "mntent.h"
 ++#include "fam-mntent.h"

   //  A StringTable maps C strings onto values.  It is a cheap O(n)
   //  implementation, suitable only for small tables that are
 Index: sysutils/fam/patches/patch-ao
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-ao,v
 retrieving revision 1.2
 diff -b -u -r1.2 patch-ao
 --- sysutils/fam/patches/patch-ao	2002/08/25 19:23:22	1.2
 +++ sysutils/fam/patches/patch-ao	2003/10/10 18:37:05
 @@ -1,11 +1,11 @@
  $NetBSD: patch-ao,v 1.2 2002/08/25 19:23:22 jlam Exp $

 ---- fam/mntent.h.orig	Sun May 12 19:15:01 2002
 -+++ fam/mntent.h
 +--- fam/fam-mntent.h.orig	Sun May 12 19:15:01 2002
 ++++ fam/fam-mntent.h
  @@ -0,0 +1,56 @@
  +/*
  + *  mntent
 -+ *  mntent.h - compatability header for BSD
 ++ *  fam-mntent.h - compatability header for BSD
  + *
  + *  Copyright (c) 2001 David Rufino <daverufino@btinternet.com>
  + *  All rights reserved.
 Index: sysutils/fam/patches/patch-ap
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-ap,v
 retrieving revision 1.2
 diff -b -u -r1.2 patch-ap
 --- sysutils/fam/patches/patch-ap	2002/08/25 19:23:22	1.2
 +++ sysutils/fam/patches/patch-ap	2003/10/10 18:37:05
 @@ -1,7 +1,7 @@
  $NetBSD: patch-ap,v 1.2 2002/08/25 19:23:22 jlam Exp $

 ---- fam/mntent_compat.c++.orig	Sun May 12 19:37:49 2002
 -+++ fam/mntent_compat.c++
 +--- fam/fam-mntent_compat.c++.orig	Sun May 12 19:37:49 2002
 ++++ fam/fam-mntent_compat.c++
  @@ -0,0 +1,174 @@
  +/*
  + * Copyright (c) 1980, 1989, 1993, 1994
 @@ -41,7 +41,7 @@
  +/* most of this was ripped from the mount(3) source */
  +
  +#include "config.h"
 -+#include "mntent.h"
 ++#include "fam-mntent.h"
  +#if !defined(HAVE_MNTENT_H)
  +#include <stdlib.h>
  +#include <string.h>
 Index: sysutils/fam/patches/patch-au
 ===================================================================
 RCS file: /cvsroot/pkgsrc/sysutils/fam/patches/patch-au,v
 retrieving revision 1.3
 diff -b -u -r1.3 patch-au
 --- sysutils/fam/patches/patch-au	2002/12/17 19:43:43	1.3
 +++ sysutils/fam/patches/patch-au	2003/10/10 18:37:06
 @@ -15,7 +15,7 @@
     main.c++ \
     timeval.c++ \
     timeval.h \
 -+  mntent_compat.c++ \
 ++  fam-mntent_compat.c++ \
     @IMON_FUNCS@.c++


 @@ -23,7 +23,7 @@
   RPC_TCP_Connector.$(OBJEXT) Scanner.$(OBJEXT) Scheduler.$(OBJEXT) \
   ServerConnection.$(OBJEXT) ServerHost.$(OBJEXT) ServerHostRef.$(OBJEXT) \
   TCP_Client.$(OBJEXT) main.$(OBJEXT) timeval.$(OBJEXT) \
 -+mntent_compat.$(OBJEXT) \
 ++fam-mntent_compat.$(OBJEXT) \
   @IMON_FUNCS@.$(OBJEXT)
   fam_DEPENDENCIES =  $(top_srcdir)/support/libsupport.a
   fam_LDFLAGS =



From: Bas van Oostveen <v.oostveen@hccnet.nl>
To: gnats-bugs@NetBSD.org
Cc:  
Subject: Subject: Re: pkg/23017
Date: Wed, 19 Nov 2003 19:21:11 +0100

 i also have this problem with pkgsrc/irix.
 again it's nesting itself up to deeply and
 /usr/include also contains a mntent.h

 although it should be noted that for native
 Irix machines with pkgsrc on it, using the 
 sgi provided fam would be better. (?)

 my not-so-pretty fix is changing mntent.h in
 work/fam-2.6.9/fam to:

 -include <mntent.h>
 +include "/usr/include/mntent.h"



From: "Jeremy C. Reed" <reed@reedmedia.net>
To: gnats-bugs@gnats.netbsd.org
Cc:  
Subject: pkg/23017
Date: Thu, 11 Mar 2004 09:48:41 -0800 (PST)

 I can confirm this. It just had exact same issue. Under Linux, it assumes
 I have a sys/ucred.h header, but I don't. And then the MNT_RDONLY,
 MNT_SYNCHRONOUS, and other macros are not defined.

 (I needed this installed quick, so I used the suggested workaround.)

    Jeremy C. Reed
    http://www.reedmedia.net/

State-Changed-From-To: open->suspended 
State-Changed-By: rh 
State-Changed-When: Sat Mar 27 22:06:36 UTC 2004 
State-Changed-Why:  
At the moment, pkgsrc does not support installation to /usr 
This change is too substantial to be warranted during the current freeze. 
If there is desire to support pkgsrc installations to /usr, this PR can 
be re-opened at a later time. 

From: "Jeremy C. Reed" <reed@reedmedia.net>
To: gnats-bugs@gnats.netbsd.org
Cc: rh@netbsd.org, <pkg-manager@netbsd.org>, <netbsd-bugs@netbsd.org>
Subject: Re: pkg/23017
Date: Sat, 27 Mar 2004 14:59:38 -0800 (PST)

 On 27 Mar 2004 rh@netbsd.org wrote:

 > State-Changed-From-To: open->suspended
 > State-Changed-By: rh
 > State-Changed-When: Sat Mar 27 22:06:36 UTC 2004
 > State-Changed-Why:
 > At the moment, pkgsrc does not support installation to /usr

 I don't understand this statement, unless you are saying it "does not
 *officially* support installation to /usr".

 For the record (again): I have used LOCALBASE as /usr starting in
 September 1992 and have used it on over 15 systems (including production
 and everyday-use systems) for hundreds of packages. I have used
 LOCALBASE=/usr under NetBSD and Linux.

 > This change is too substantial to be warranted during the current freeze.

 The freeze is over. It was documented to end today.

 As for this issue, the package already has patches forcing this -- they
 just need to be done correctly.

 Also the PR has a note from another user indicating they have same issue
 on Irix box with native mntent.h.

 Also, PR #24578 is very similar and mentions part of same problems.

 > If there is desire to support pkgsrc installations to /usr, this PR can
 > be re-opened at a later time.

 Who makes that decision?

 (Before I even started using LOCALBASE as /usr over 18 months ago, others
 posted that they have also used pkgsrc the same.)

    Jeremy C. Reed
    http://www.reedmedia.net/



From: "Greg A. Woods" <woods@weird.com>
To: NetBSD GNATS submissions and followups <gnats-bugs@gnats.netbsd.org>,
  <rh@netbsd.org>, <pkg-manager@netbsd.org>,
  NetBSD Bugs and PR posting List <netbsd-bugs@NetBSD.ORG>
Cc:  
Subject: Re: pkg/23017
Date: Sun, 28 Mar 2004 02:32:59 -0500 (EST)

 [ On Saturday, March 27, 2004 at 14:59:38 (-0800), Jeremy C. Reed wrote: ]
 > Subject: Re: pkg/23017
 >
 > (Before I even started using LOCALBASE as /usr over 18 months ago, others
 > posted that they have also used pkgsrc the same.)

 Indeed I've been successfully using LOCALBASE=/usr for my production
 NetBSD systems for even longer (at least two years).  My users very much
 appreciate this level of software integration.

 During that time I've found and fixed quite a few very questionable, or
 at least very lazy, constructs in pgksrc.  Things like making wrong
 assumptions about host software versions based solely on the non-unique
 name of a file in a non-unique directory.  Sadly new such constructs
 seem to appear on a regular basis.  For an OS project that claims to
 pride itself on doing things "Right", this makes it look more like a
 fallacy at times.

 I think it's ludicrous that pkgsrc does not directly and fully support
 the installation of packages directly into the same locations as any
 base-OS software.  The artificial segregation of third-party software
 into a (near) mirror hierarchy of the base-OS directories is almost
 completely senseless in almost the majority of scenarios, especially
 when all package contents and dependencies are carefully registered and
 they can be installed without conflicts and can be easily de-installed.
 Except for a few instances where unexpected filename conflicts occur
 (though those are trivial to automatically identify even without having
 a fully registered package set for the base OS provided you can build
 and install the package twice, once with LOCALBASE=/usr/pkg, and then
 again with LOCALBASE=/usr provided no conflicts are found in the first
 test), the only time conflicts occur is when base-OS software is
 explicitly being upgraded or replaced.

 There are a few annoyances that I've had difficulty working around as a
 third party developer, but they would be trivial to fix if the core
 project supported LOCALBASE=/usr.  (e.g. the not-quite-matching pkg
 hierarchy where "man" and "info" directories don't live in "share" where
 they should)

 Only a very tiny number of packages still clash with at least NetBSD
 base-OS components, such as GNU readline vs. NetBSD readline.  Of course
 these things wouldn't be as much of a problem if the NetBSD variants
 actually implemented a fully compatible API and thus the third party
 versions were not ever necessary for any packages.  However in all the
 cases I've encountered like this so far it's been possible to simply
 configure and build other packages which use these conflicting
 "dependencies" in such a way that they don't require the conflicing
 package(s) in the first place.  E.g. in the case of GNU readline I
 simply don't install it at all and if some other package requires it
 (i.e. won't work with the partial NetBSD native implementation) then I
 modify the configuration to simply not use any readline library at all.
 Except for myself my other users aren't normally sophisticated enough to
 notice the difference.

 -- 
 						Greg A. Woods

 +1 416 218-0098                  VE3TCP            RoboHack <woods@robohack.ca>
 Planix, Inc. <woods@planix.com>          Secrets of the Weird <woods@weird.com>

From: Min Sik Kim <minskim@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:  
Subject: pr/23017 CVS commit: pkgsrc
Date: Sun, 28 Mar 2004 22:00:05 +0000 (UTC)

 Module Name:	pkgsrc
 Committed By:	minskim
 Date:		Sun Mar 28 22:00:05 UTC 2004

 Modified Files:
 	pkgsrc/doc: CHANGES
 	pkgsrc/sysutils/fam: MESSAGE Makefile PLIST distinfo
 	pkgsrc/sysutils/fam/patches: patch-ab patch-ae patch-af patch-ag
 	    patch-ah patch-ai patch-aj patch-ao patch-ap patch-au patch-ax
 Removed Files:
 	pkgsrc/sysutils/fam/patches: patch-ak patch-an patch-av patch-aw

 Log Message:
 Update fam to 2.7.0.  Fixes the following PR's:

   - PR pkg/23017
   - PR pkg/23495
   - PR pkg/23496
   - PR pkg/24578

 The patches for 23017 and 24578 were provided by reed@.  Although rh@
 suspended PR 23017 saying that pkgsrc does not support installation to
 /usr, this still needs to be fixed because the same problem occurs
 even when LOCALBASE != /usr.

 Changes since 2.6.10:
     Changed NOGROUP usage in src/Cred.c++ to -1 as NOGROUP is not defined
     in IRIX.
     (mraymond@sgi.com)
     removed const modifier from Scheduler.h to fix GCC 3.x build.
     (wardle@sgi.com)
     created new GNU autotools files:
     - test whether we have IMon rather than testing OS.
     - test whether we have sockaddr length members rather than testing OS.
     - install daemon in sbin rather than bin.
     - install manual pages in typical GNU/Linux sections.
     - rename daemon to famd.
     - rename fam source directory to src.
     - rename libfam source directory to lib.
     - remove editconf and util subdirectory.
     - remove rpm target and build subdirectory.
     - remove support target and subdirectory.
     [wardle@sgi.com] (wardle@sgi.com)
     fixed warnings about typenames in BTree.h.
     (wardle@sgi.com)


 To generate a diff of this commit:
 cvs rdiff -r1.5325 -r1.5326 pkgsrc/doc/CHANGES
 cvs rdiff -r1.2 -r1.3 pkgsrc/sysutils/fam/MESSAGE
 cvs rdiff -r1.11 -r1.12 pkgsrc/sysutils/fam/Makefile
 cvs rdiff -r1.3 -r1.4 pkgsrc/sysutils/fam/PLIST
 cvs rdiff -r1.7 -r1.8 pkgsrc/sysutils/fam/distinfo
 cvs rdiff -r1.2 -r1.3 pkgsrc/sysutils/fam/patches/patch-ab \
     pkgsrc/sysutils/fam/patches/patch-ae pkgsrc/sysutils/fam/patches/patch-af \
     pkgsrc/sysutils/fam/patches/patch-ai pkgsrc/sysutils/fam/patches/patch-ao \
     pkgsrc/sysutils/fam/patches/patch-ap
 cvs rdiff -r1.3 -r1.4 pkgsrc/sysutils/fam/patches/patch-ag \
     pkgsrc/sysutils/fam/patches/patch-ah
 cvs rdiff -r1.4 -r1.5 pkgsrc/sysutils/fam/patches/patch-aj \
     pkgsrc/sysutils/fam/patches/patch-au pkgsrc/sysutils/fam/patches/patch-ax
 cvs rdiff -r1.3 -r0 pkgsrc/sysutils/fam/patches/patch-ak \
     pkgsrc/sysutils/fam/patches/patch-av pkgsrc/sysutils/fam/patches/patch-aw
 cvs rdiff -r1.2 -r0 pkgsrc/sysutils/fam/patches/patch-an

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.


From: Alistair Crooks <agc@pkgsrc.org>
To: NetBSD GNATS submissions and followups <gnats-bugs@gnats.netbsd.org>,
  NetBSD Bugs and PR posting List <netbsd-bugs@NetBSD.org>
Cc: rh@NetBSD.org, pkg-manager@NetBSD.org
Subject: Re: pkg/23017
Date: Mon, 29 Mar 2004 18:50:36 +0100

 On Sun, Mar 28, 2004 at 02:32:59AM -0500, Greg A. Woods wrote:
 > [ On Saturday, March 27, 2004 at 14:59:38 (-0800), Jeremy C. Reed wrote: ]
 > > Subject: Re: pkg/23017
 > >
 > > (Before I even started using LOCALBASE as /usr over 18 months ago, others
 > > posted that they have also used pkgsrc the same.)
 > 
 > Indeed I've been successfully using LOCALBASE=/usr for my production
 > NetBSD systems for even longer (at least two years).  My users very much
 > appreciate this level of software integration.
 > 
 > During that time I've found and fixed quite a few very questionable, or
 > at least very lazy, constructs in pgksrc.  Things like making wrong
 > assumptions about host software versions based solely on the non-unique
 > name of a file in a non-unique directory.  Sadly new such constructs
 > seem to appear on a regular basis.  For an OS project that claims to
 > pride itself on doing things "Right", this makes it look more like a
 > fallacy at times.
 > 
 > I think it's ludicrous that pkgsrc does not directly and fully support
 > the installation of packages directly into the same locations as any
 > base-OS software.  The artificial segregation of third-party software
 > into a (near) mirror hierarchy of the base-OS directories is almost
 > completely senseless in almost the majority of scenarios, especially
 > when all package contents and dependencies are carefully registered and
 > they can be installed without conflicts and can be easily de-installed.
 > Except for a few instances where unexpected filename conflicts occur
 > (though those are trivial to automatically identify even without having
 > a fully registered package set for the base OS provided you can build
 > and install the package twice, once with LOCALBASE=/usr/pkg, and then
 > again with LOCALBASE=/usr provided no conflicts are found in the first
 > test), the only time conflicts occur is when base-OS software is
 > explicitly being upgraded or replaced.
 > 
 > There are a few annoyances that I've had difficulty working around as a
 > third party developer, but they would be trivial to fix if the core
 > project supported LOCALBASE=/usr.  (e.g. the not-quite-matching pkg
 > hierarchy where "man" and "info" directories don't live in "share" where
 > they should)
 > 
 > Only a very tiny number of packages still clash with at least NetBSD
 > base-OS components, such as GNU readline vs. NetBSD readline.  Of course
 > these things wouldn't be as much of a problem if the NetBSD variants
 > actually implemented a fully compatible API and thus the third party
 > versions were not ever necessary for any packages.  However in all the
 > cases I've encountered like this so far it's been possible to simply
 > configure and build other packages which use these conflicting
 > "dependencies" in such a way that they don't require the conflicing
 > package(s) in the first place.  E.g. in the case of GNU readline I
 > simply don't install it at all and if some other package requires it
 > (i.e. won't work with the partial NetBSD native implementation) then I
 > modify the configuration to simply not use any readline library at all.
 > Except for myself my other users aren't normally sophisticated enough to
 > notice the difference.

 Thank you for your thoughts, Greg (and with which I completely disagree).
 I shall treat your remarks with the respect they deserve.

 In passing, I'd like to invite you not to use such terms as "lazy", or
 trying to ride on a wave of invective by claiming that some things in
 pkgsrc are not done "Right" simply because they don't coincide with
 your desires/wants/needs.

 Alistair

From: "Greg A. Woods" <woods@weird.com>
To: Alistair Crooks <agc@pkgsrc.org>
Cc: NetBSD Bugs and PR posting List <netbsd-bugs@NetBSD.ORG>,
  NetBSD GNATS submissions and followups <gnats-bugs@gnats.netbsd.org>
Subject: Re: pkg/23017
Date: Mon, 29 Mar 2004 20:42:47 -0500 (EST)

 [ On Monday, March 29, 2004 at 18:50:36 (+0100), Alistair Crooks wrote: ]
 > Subject: Re: pkg/23017
 >
 > In passing, I'd like to invite you not to use such terms as "lazy", or
 > trying to ride on a wave of invective by claiming that some things in
 > pkgsrc are not done "Right" simply because they don't coincide with
 > your desires/wants/needs.

 It must be you who are confusing facts with your own desires/wants/needs.

 First off do you have even the slightest clue what questionable _or_
 lazy constructs I might have been referring to?  I didn't mention them
 exactly because they are not really relevant to this meta discussion.

 Are you trying to defend yourself, the project, or someone else?  As far
 as I can tell you're not personally responsible for any of the
 constructs I have in mind (at least not according to "cvs annotate"),
 unless perhaps you're taking some higher level of responsibility for
 what's been done.

 The claims I've made are in fact completely in line with the stated
 technical goals of the NetBSD project (of which I happen to agree almost
 entirely with).  They are also clearly not very well aligned with the
 requirements one has when using pkgsrc on systems other than NetBSD.  If
 you don't think what I've said is indeed "Right" (though how can you
 since you don't even know exactly what I'm referring to?), then you must
 be disagreeing with the stated project goals, not just me.

 In this case the fact that these constructs can clash with the use of
 LOCALBASE==/usr is a mere side-effect and I mentioned them by reference
 only as an example of some things that should be cleaned up anyway and
 which when cleaned up would make using LOCALBASE=/usr even easier.  In
 fact some of them have been cleaned up recently, though usually in a
 still somewhat questionable manner where only the symptoms of the real
 problems are addressed.

 -- 
 						Greg A. Woods

 +1 416 218-0098                  VE3TCP            RoboHack <woods@robohack.ca>
 Planix, Inc. <woods@planix.com>          Secrets of the Weird <woods@weird.com>
State-Changed-From-To: suspended->closed 
State-Changed-By: reed 
State-Changed-When: Wed Mar 31 19:15:01 UTC 2004 
State-Changed-Why:  
minskim@ updated fam. He used some of my ideas and it appears to build under 
Linux fine for me now. 
>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.