Copyright © 1994-2006 The NetBSD Foundation, Inc
$NetBSD: pkgsrc.xml,v 1.24 2006/11/11 05:39:09 rillig Exp $
Abstract
pkgsrc is a centralized package management system for Unix-like operating systems. This guide provides information for users and developers of pkgsrc. It covers installation of binary and source packages, creation of binary and source packages and a high-level overview about the infrastructure.
Table of Contents
CFLAGS
?Makefile
smk.conf
List of Tables
Table of Contents
There is a lot of software freely available for Unix-based systems, which is usually available in form of the source code. Before such software can be used, it needs to be configured to the local system, compiled and installed, and this is exactly what The NetBSD Packages Collection (pkgsrc) does. pkgsrc also has some basic commands to handle binary packages, so that not every user has to build the packages for himself, which is a time-costly task.
pkgsrc currently contains several thousand packages, including:
www/apache
- The Apache
web server
www/firefox
- The Firefox
web browser
meta-pkgs/gnome
- The GNOME
Desktop Environment
meta-pkgs/kde3
- The K
Desktop Environment
...just to name a few.
pkgsrc has built-in support for handling varying dependencies, such as pthreads and X11, and extended features such as IPv6 support on a range of platforms.
pkgsrc provides the following key features:
Easy building of software from source as well as the creation and installation of binary packages. The source and latest patches are retrieved from a master or mirror download site, checksum verified, then built on your system. Support for binary-only distributions is available for both native platforms and NetBSD emulated platforms.
All packages are installed in a consistent directory tree, including binaries, libraries, man pages and other documentation.
Package dependencies, including when performing package updates, are handled automatically. The configuration files of various packages are handled automatically during updates, so local changes are preserved.
Like NetBSD, pkgsrc is designed with portability in mind and consists of highly portable code. This allows the greatest speed of development when porting to new a platform. This portability also ensures that pkgsrc is consistent across all platforms.
The installation prefix, acceptable software licenses, international encryption requirements and build-time options for a large number of packages are all set in a simple, central configuration file.
The entire source (not including the distribution files) is freely available under a BSD license, so you may extend and adapt pkgsrc to your needs. Support for local packages and patches is available right out of the box, so you can configure it specifically for your environment.
The following principles are basic to pkgsrc:
“It should only work if it's right.”
— That means, if a package contains bugs, it's better to find
them and to complain about them rather than to just install the package
and hope that it works. There are numerous checks in pkgsrc that try to
find such bugs: Static analysis tools (pkgtools/pkglint
), build-time checks (portability
of shell scripts), and post-installation checks (installed files,
references to shared libraries, script interpreters).
“If it works, it should work everywhere” — Like NetBSD has been ported to many hardware architectures, pkgsrc has been ported to many operating systems. Care is taken that packages behave the same on all platforms.
pkgsrc consists of both a source distribution and a binary distribution for these operating systems. After retrieving the required source or binaries, you can be up and running with pkgsrc in just minutes!
pkgsrc was derived from FreeBSD's ports system, and initially developed for NetBSD only. Since then, pkgsrc has grown a lot, and now supports the following platforms:
This document is divided into three parts. The first, The pkgsrc user's guide, describes how one can use one of the packages in the Package Collection, either by installing a precompiled binary package, or by building one's own copy using the NetBSD package system. The second part, The pkgsrc developer's guide, explains how to prepare a package so it can be easily built by other NetBSD users without knowing about the package's building details. The third part, The pkgsrc infrastructure internals is intended for those who want to understand how pkgsrc is implemented.
This document is available in various formats: HTML, PDF, PS, TXT.
There has been a lot of talk about “ports”, “packages”, etc. so far. Here is a description of all the terminology used within this document.
A set of files and building instructions
that describe what's necessary
to build a certain piece of software using
pkgsrc. Packages are traditionally stored under
/usr/pkgsrc
.
This is the former name of “pkgsrc”. It is part of the NetBSD operating system and can be bootstrapped to run on non-NetBSD operating systems as well. It handles building (compiling), installing, and removing of packages.
This term describes the file or files that are
provided by the author of the piece of software to
distribute his work. All the changes necessary to build on
NetBSD are reflected in the corresponding package. Usually
the distfile is in the form of a compressed tar-archive,
but other types are possible, too. Distfiles are usually
stored below
/usr/pkgsrc/distfiles
.
This is the term used by FreeBSD and OpenBSD people for what we call a package. In NetBSD terminology, “port” refers to a different architecture.
A set of binaries built with pkgsrc from a distfile
and stuffed together in a single .tgz
file so it can be installed on machines of the same
machine architecture without the need to
recompile. Packages are usually generated in
/usr/pkgsrc/packages
; there is also
an archive on ftp.NetBSD.org.
Sometimes, this is referred to by the term “package” too, especially in the context of precompiled packages.
The piece of software to be installed which will be constructed from all the files in the distfile by the actions defined in the corresponding package.
The pkgsrc users are people who use the packages provided by pkgsrc. Typically they are system administrators. The people using the software that is inside the packages (maybe called “end users”) are not covered by the pkgsrc guide.
There are two kinds of pkgsrc users: Some only want to install pre-built binary packages. Others build the pkgsrc packages from source, either for installing them directly or for building binary packages themselves. For pkgsrc users Part I, “The pkgsrc user's guide” should provide all necessary documentation.
A package maintainer creates packages as described in Part II, “The pkgsrc developer's guide”.
These people are involved in all those files
that live in the mk/
directory and below.
Only these people should need to read through Part III, “The pkgsrc infrastructure internals”, though others might be curious,
too.
Table of Contents
CFLAGS
?Table of Contents
Before you download and extract the files, you need to decide
where you want to extract them. When using pkgsrc as root user, pkgsrc
is usually installed in /usr/pkgsrc
. You are though
free to install the sources and binary packages wherever you want in
your filesystem, provided that the pathname does not contain white-space
or other characters that are interpreted specially by the shell and some
other programs. A safe bet is to use only letters, digits, underscores
and dashes.
Before you download any pkgsrc files, you should decide
whether you want the current branch or the
stable branch. The latter is forked on a
quarterly basis from the current branch and only gets modified
for security updates. The names of the stable branches are built
from the year and the quarter, for example
2006Q4
.
The second step is to decide how you want to download pkgsrc. You can get it as a tar file, via SUP, or via CVS. All three ways are described here.
The primary download location for all pkgsrc files is ftp://ftp.NetBSD.org/pub/pkgsrc/. There are a number of subdirectories for different purposes, which are described in detail in Appendix C, Directory layout of the pkgsrc FTP server.
The tar file for the current branch is in the directory
current
and is called pkgsrc.tar.gz
.
It is autogenerated daily.
The tar file for the stable branch 2006Q4 is in the
directory pkgsrc-2006Q4
and is also called pkgsrc-2006Q4.tar.gz
.
After downloading the tar file, change to the directory
where you want to have pkgsrc. This is usually
/usr
. Then, run gzcat
pkgsrc.tar.gz | tar xf - to extract the files.
As an alternative to the tar file, you can get pkgsrc via the Software Update Protocol, SUP. To do so, make sure your supfile has a line
release=pkgsrc
in it, see the examples in
/usr/share/examples/supfiles
, and that the
/usr/pkgsrc
directory exists. Then, simply
run sup -v
/path/to/your/supfile
.
To get pkgsrc via CVS, make sure you have cvs(1) installed. To do an initial (full) checkout of pkgsrc, you first have to set some environment variables. For the C-Shell, type:
%
setenv CVSROOT anoncvs@anoncvs.NetBSD.org:/cvsroot
%
setenv CVS_RSH ssh
Or, the same for the bourne shell:
$
CVSROOT="anoncvs@anoncvs.NetBSD.org:/cvsroot"
$
CVS_RSH="ssh"
$
export CVSROOT CVS_RSH
Then, you change to the directory where you want to have
your copy of pkgsrc. In most cases this is
/usr
. In that directory you run the
checkout command, which is cvs -q checkout -P
pkgsrc for the current branch and cvs -q
checkout -rpkgsrc-2006Q4 -P pkgsrc for the stable
branch. This command will create a directory called
pkgsrc
with all the pkgsrc files in
it.
The preferred way to keep pkgsrc up-to-date is via CVS (which also works if you have first installed it via a tar file). It saves bandwidth and hard disk activity, compared to downloading the tar file again.
When updating from a tar file, you first need to completely remove the old pkgsrc directory. Otherwise those files that have been removed from pkgsrc in the mean time will not be removed on your local disk, resulting in inconsistencies. When removing the old files, any changes that you have done to the pkgsrc files will be lost after updating. Therefore updating via CVS is strongly recommended.
Note that by default the distfiles and the binary packages
are saved in the pkgsrc tree, so don't forget to rescue them
before updating. You can also configure pkgsrc to use other than
the default directories by setting the
DISTDIR
and PACKAGES
variables. See Chapter 5, Configuring pkgsrc for the details.
To update pkgsrc from a tar file, download the tar file as explained above. Then, make sure that you have not made any changes to the files in the pkgsrc directory. Remove the pkgsrc directory and extract the new tar file. Done.
To update pkgsrc via CVS, make sure the environment
variable CVS_RSH
is set as above. Then,
change to the pkgsrc directory and run cvs -q update
-dP.
When updating pkgsrc, the CVS program keeps track of the branch you selected. But if you, for whatever reason, want to switch from the stable branch to the current one, you can do it by adding the option “-A” after the “update” keyword. To switch from the current branch back to the stable branch, add the “-rpkgsrc-2006Q4” option.
When you update pkgsrc, the CVS program will only touch those files that are registered in the CVS repository. That means that any packages that you created on your own will stay unmodified. If you change files that are managed by CVS, later updates will try to merge your changes with those that have been done by others. See the CVS manual, chapter “update” for details.
Table of Contents
We provide a bootstrap kit in both source and binary form for other platforms than NetBSD, consisting of the pkg administration tools and other tools required to use pkgsrc and build packages.
Older binary snapshots are available in the bootstrap-pkgsrc/archive directory on ftp.NetBSD.org.
Table 3.1. Binary kits and available packages
Platform | Latest snapshot | Binary kit | Binary packages |
---|---|---|---|
Darwin 5.5/powerpc (Mac OS X 10.1.5) | 20021209 | binary kit | |
Darwin 6.6/powerpc (Mac OS X 10.2.6) | 20030623 | binary kit | |
Darwin 7.6/powerpc (Mac OS X 10.3.6) | 20041219 | binary kit | binary packages |
Darwin 7.8/powerpc (Mac OS X 10.3.8) | 20050320 | binary kit | |
Darwin 8.1/powerpc (Mac OS X 10.4.1) | 20050625 | binary kit | binary packages |
Debian GNU Linux/i386 | 20031023 | binary kit | |
Fedora Core 2 Linux/i386 | 20050618 | binary kit | |
Fedora Core 4 Linux/i386 | 20060105 | binary kit | |
FreeBSD 3.5/i386 | 20030411 | binary kit | |
FreeBSD 4.7/i386 | 20021211 | binary kit | |
FreeBSD 5.0/i386 | 20030411 | binary kit | |
FreeBSD 5.1/i386 | 20030630 | binary kit | |
FreeBSD 5.2.1/i386 | 20040227 | binary kit | |
FreeBSD 5.3/i386 | 20050119 | binary kit | |
Interix 3.5 | 20061106 | binary kit | |
IRIX 6.5 n32-bit ABI | 20040911 | binary kit | binary packages |
IRIX 6.5 64-bit ABI | 20040912 | binary kit | binary packages |
OpenBSD 3.2/i386 | 20030420 | binary kit | |
OpenBSD 3.3/i386 | 20030503 | binary kit | |
OpenBSD 3.5/i386 | 20040703 | binary kit | |
Slackware Linux 8.1/i386 | 20030417 | binary kit | |
Slackware Linux 9/i386 | 20040703 | binary kit | |
Solaris 8/sparc | 20050220 | binary kit | |
Solaris 8/i386 | 20050220 | binary kit | |
Solaris 9/sparc | 20060713 | binary kit | |
Solaris 9/i386 | 20030411 | binary kit |
Simply download the binary kit for your platform, and extract it into
/
e.g.
#
cd /
#
gzip -c -d /tmp/bootstrap-pkgsrc-SunOS-5.9-sparc-20031023.tar.gz \ | tar -xpf -
Of course, as with any binary distributions, you should verify the checksum against the SUM or CKSUM file and inspect the contents before extracting it.
Installing the bootstrap kit from source should be as simple as:
#
env CVS_RSH=ssh cvs -d anoncvs@anoncvs.NetBSD.org:/cvsroot checkout pkgsrc
#
cd pkgsrc/bootstrap
#
./bootstrap
See Chapter 2, Where to get pkgsrc and how to keep it up-to-date for other ways to get
pkgsrc before bootstrapping. The given
bootstrap command will use the defaults of
/usr/pkg
for the
prefix where programs will be installed in,
and /var/db/pkg
for the package database
directory where pkgsrc will do its internal bookkeeping.
However, these can also be set using command-line
arguments.
Binary packages for the pkgsrc tools and an initial set of packages is
available for supported platforms. An up-to-date list of these can be
found on www.pkgsrc.org.
Note that this only works for privileged builds that install
into /usr/pkg
.
The bootstrap installs a bmake tool. Use this bmake when building via pkgsrc. For examples in this guide, use bmake instead of “make”.
Here are some platform-specific notes you should be aware of.
Darwin 5.x and up are supported. There are two methods of using pkgsrc on Mac OS X, by using a disk image, or a UFS or HFSX partition.
Before you start, you will need to download and install the Mac OS X Developer Tools from Apple's Developer Connection. See http://developer.apple.com/macosx/ for details. Also, make sure you install X11 for Mac OS X and the X11 SDK from http://www.apple.com/macosx/x11/download/ if you intend to build packages that use the X11 Window System.
If you already have a UFS or HFSX partition, or have a spare partition that you can format as UFS or HFSX, it is recommended to use that instead of the disk image. It'll be somewhat faster and will mount automatically at boot time, where you must manually mount a disk image.
You cannot use an ordinary HFS+ file system for pkgsrc, because pkgsrc currently requires the file system to be case-sensitive. You can, however, use a case-sensitive HFS+ (aka HFSX) file system as found in Darwin 7.0 and newer.
Create the disk image:
#
cd pkgsrc/bootstrap
#
./darwindiskimage create ~/Documents/NetBSD 1024
# megabytes - season to taste#
./darwindiskimage mount ~/Documents/NetBSD
#
sudo chown `id -u`:`id -g` /Volumes/NetBSD
darwindiskimage will mount the filesystem nosuid, which will cause problems for packages that depend on setgid. In the case of UFS, it will also mount the filesystem asynchronous, which is somewhat dangerous according to the mount(8) man page. In the case of HFSX, it will disable journalling.
Allow suid:
#
sudo mount -u -o suid /Volumes/NetBSD
By default, /usr
will be on your root file
system, normally HFS+. It is possible to use the default
prefix of /usr/pkg
by symlinking /usr/pkg
to a directory on a UFS
or HFSX file system. Obviously, another symlink is required if you want to
place the package database directory outside the
prefix. e.g.
#
./bootstrap --pkgdbdir /usr/pkg/pkgdb
If you created your partitions at the time of installing Mac OS X
and formatted the target partition as UFS or HFSX, it should automatically
mount on /Volumes/<volume name>
when the
machine boots. If you are (re)formatting a partition as UFS or HFSX, you need
to ensure that the partition map correctly reflects
“Apple_UFS” or “Apple_HFSX” and not “Apple_HFS”.
The problem is that none of the disk tools will let you touch a disk that is booted from. You can unmount the partition, but even if you newfs it, the partition type will be incorrect and the automounter won't mount it. It can be mounted manually, but it won't appear in Finder.
You'll need to boot off of the OS X Installation (User) CD. When the Installation program starts, go up to the menu and select Disk Utility. Now, you will be able to select the partition you want to be UFS or HFSX, and Format it Apple UFS or HFSX. Quit the Disk Utility, quit the installer which will reboot your machine. The new UFS or HFSX file system will appear in Finder.
Be aware that the permissions on the new file system will be writable by root only.
This note is as of 10.2 (Jaguar) and applies to earlier versions. Hopefully Apple will fix Disk Utility in 10.3 (Panther).
FreeBSD 4.7 and 5.0 have been tested and are supported, other versions may work.
Care should be taken so that the tools that this kit installs do not conflict with the FreeBSD userland tools. There are several steps:
FreeBSD stores its ports pkg database in
/var/db/pkg
. It is therefore
recommended that you choose a different location (e.g.
/usr/pkgdb
) by
using the --pkgdbdir option to the bootstrap script.
If you do not intend to use the FreeBSD ports tools, it's probably a good idea to move them out of the way to avoid confusion, e.g.
#
cd /usr/sbin
#
mv pkg_add pkg_add.orig
#
mv pkg_create pkg_create.orig
#
mv pkg_delete pkg_delete.orig
#
mv pkg_info pkg_info.orig
An example /etc/mk.conf
file will be placed in
/etc/mk.conf.example
file
when you use the bootstrap script.
Interix is a POSIX-compatible subsystem for the Windows NT kernel, providing a Unix-like environment with a tighter kernel integration than available with Cygwin. It is part of the Windows Services for Unix package, available for free for any licensed copy of Windows 2000, XP (not including XP Home), or 2003. SFU can be downloaded from http://www.microsoft.com/windows/sfu/.
Services for Unix 3.5 has been tested. 3.0 or 3.1 may work, but are not officially supported. (The main difference in 3.0/3.1 is lack of pthreads, but other parts of libc may also be lacking.)
Services for Unix Applications (aka SUA) is an integrated component of Windows Server 2003 R2 and Windows Vista. As of this writing, SUA's Interix 5.x subsystem has not yet been tested with pkgsrc.
At an absolute minimum, the following packages must be installed from the Windows Services for Unix 3.5 distribution in order to use pkgsrc:
Utilities -> Base Utilities
Interix GNU Components -> (all)
Remote Connectivity
Interix SDK
When using pkgsrc on Interix, DO NOT install the Utilities subcomponent "UNIX Perl". That is Perl 5.6 without shared module support, installed to /usr/local, and will only cause confusion. Instead, install Perl 5.8 from pkgsrc (or from a binary package).
The Remote Connectivity subcomponent "Windows Remote Shell Service" does not need to be installed, but Remote Connectivity itself should be installed in order to have a working inetd.
During installation you may be asked whether to enable setuid behavior for Interix programs, and whether to make pathnames default to case-sensitive. Setuid should be enabled, and case-sensitivity MUST be enabled. (Without case-sensitivity, a large number of packages including perl will not build.)
NOTE: Newer Windows service packs change the way binary execution works (via the Data Execution Prevention feature). In order to use pkgsrc and other gcc-compiled binaries reliably, a hotfix containing POSIX.EXE, PSXDLL.DLL, PSXRUN.EXE, and PSXSS.EXE (899522 or newer) must be installed. Hotfixes are available from Microsoft through a support contract; however, a NetBSD developer has made most Interix hotfixes available for personal use from http://www.duh.org/interix/hotfixes.php.
In addition to the hotfix noted above, it may be necessary to disable Data Execution Prevention entirely to make Interix functional. This may happen only with certain types of CPUs; the cause is not fully understood at this time. If gcc or other applications still segfault repeatedly after installing one of the hotfixes note above, the following option can be added to the appropriate "boot.ini" line on the Windows boot drive: /NoExecute=AlwaysOff (WARNING, this will disable DEP completely, which may be a security risk if applications are often run as a user in the Administrators group!)
If SFU is already installed and you wish to alter these settings to work with pkgsrc, note the following things.
To uninstall UNIX Perl, use Add/Remove Programs, select Microsoft Windows Services for UNIX, then click Change. In the installer, choose Add or Remove, then uncheck Utilities->UNIX Perl.
To enable case-sensitivity for the file system, run REGEDIT.EXE, and change the following registry key:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\kernel
Set the DWORD value "obcaseinsensitive" to 0; then reboot.
To enable setuid binaries (optional), run REGEDIT.EXE, and change the following registry key:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Services for UNIX
Set the DWORD value "EnableSetuidBinaries" to 1; then reboot.
The package manager (either the pkgsrc "su" user, or the user running "pkg_add") must be a member of the local Administrators group. Such a user must also be used to run the bootstrap. This is slightly relaxed from the normal pkgsrc requirement of "root".
The package manager should use a umask of 002. "make install" will automatically complain if this is not the case. This ensures that directories written in /var/db/pkg are Administrators-group writeable.
The popular Interix binary packages from http://www.interopsystems.com/ use an older version of pkgsrc's pkg_* tools. Ideally, these should NOT be used in conjunction with pkgsrc. If you choose to use them at the same time as the pkgsrc packages, ensure that you use the proper pkg_* tools for each type of binary package.
The TERM setting used for DOS-type console windows (including those invoked by the csh and ksh startup shortcuts) is "interix". Most systems don't have a termcap/terminfo entry for it, but the following .termcap entry provides adequate emulation in most cases:
interix:kP=\E[S:kN=\E[T:kH=\E[U:dc@:DC@:tc=pcansi:
Though Interix suffices as a familiar and flexible substitute for a full Unix-like platform, it has some drawbacks that should be noted for those desiring to make the most of Interix.
X11:
Interix comes with the standard set of X11R6 client libraries, and can run X11 based applications, but it does not come with an X server. Some options are StarNet X-Win32, Hummingbird Exceed (available in a trimmed version for Interix from Interop Systems as the Interop X Server), and the free X11 server included with Cygwin.
X11 acceleration:
Because Interix runs in a completely different NT subsystem from Win32 applications, it does not currently support various X11 protocol extensions for acceleration (such as MIT-SHM or DGA). Most interactive applications to a local X server will run reasonably fast, but full motion video and other graphics intensive applications may require a faster-than-expected CPU.
Audio:
Interix has no native support for audio output. For audio
support, pkgsrc uses the esound client/server
audio system on Interix. Unlike on most platforms, the
audio/esound
package does
not contain the esd
server component. To output audio via an Interix host, the
emulators/cygwin_esound
package
must also be installed.
CD/DVDs, USB, and SCSI:
Direct device access is not currently supported in Interix, so it is not currently possible to access CD/DVD drives, USB devices, or SCSI devices through non-filesystem means. Among other things, this makes it impossible to use Interix directly for CD/DVD burning.
Tape drives:
Due to the same limitations as for CD-ROMs and SCSI devices, tape drives are also not directly accessible in Interix. However, support is in work to make tape drive access possible by using Cygwin as a bridge (similarly to audio bridged via Cygwin's esound server).
It is not necessary, in general, to have a "root" user on the Windows system; any member of the local Administrators group will suffice. However, some packages currently assume that the user named "root" is the privileged user. To accommodate these, you may create such a user; make sure it is in the local group Administrators (or your language equivalent).
pkg_add creates directories of mode
0755, not 0775, in $PKG_DBDIR
. For the
time being, install packages as the local Administrator (or
your language equivalent), or run the following command after
installing a package to work around the issue:
#
chmod -R g+w $PKG_DBDIR
You will need a working C compiler, either gcc or SGI's MIPS and MIPSpro
compiler (cc/c89). Please set the CC
environment variable
according to your preference. If you do not have a license for the MIPSpro
compiler suite, you can download a gcc tardist file from http://freeware.sgi.com/.
Please note that you will need IRIX 6.5.17 or higher, as this is the earliest version of IRIX providing support for if_indextoname(3), if_nametoindex(3), etc.
At this point in time, pkgsrc only supports one ABI at a time. That is, you cannot switch between the old 32-bit ABI, the new 32-bit ABI and the 64-bit ABI. If you start out using "abi=n32", that's what all your packages will be built with.
Therefore, please make sure that you have no conflicting
CFLAGS
in your environment or the
/etc/mk.conf
. Particularly, make sure that you do not
try to link n32 object files with lib64 or vice versa. Check your
/etc/compiler.defaults
!
If you have the actual pkgsrc tree mounted via NFS from a different host,
please make sure to set WRKOBJDIR
to a local directory,
as it appears that IRIX linker occasionally runs into issues when trying to
link over a network-mounted file system.
The bootstrapping process should set all the right options for programs such
as imake(1), but you may want to set some options depending on your local
setup. Please see pkgsrc/mk/defaults/mk.conf
and, of
course, your compiler's man pages for details.
If you are using SGI's MIPSPro compiler, please set
PKGSRC_COMPILER= mipspro
in /etc/mk.conf
. Otherwise, pkgsrc will assume you
are using gcc and may end up passing invalid flags to the compiler. Note that
bootstrap should create an appropriate mk.conf.example
by
default.
If you have both the MIPSPro compiler chain installed as well as gcc,
but want to make sure that MIPSPro is used, please set your PATH
to not include the location of gcc (often
/usr/freeware/bin
), and (important) pass the
'--preserve-path' flag.
Some versions of Linux (for example Debian GNU/Linux) need either libtermcap or libcurses (libncurses). Installing the distributions libncurses-dev package (or equivalent) should fix the problem.
pkgsrc supports both gcc (GNU Compiler Collection) and icc (Intel C++ Compiler). gcc is the default. icc 8.0 and 8.1 on i386 have been tested.
To bootstrap using icc, assuming the default icc installation directory:
env CC=/opt/intel_cc_80/bin/icc LDFLAGS=-static-libcxa \ ac_cv___attribute__=yes ./bootstrap
icc 8.1 needs the `-i-static' argument instead of -static-libcxa.
icc supports __attribute__, but the GNU configure test uses a nested function, which icc does not support. #undef'ing __attribute__ has the unfortunate side-effect of breaking many of the Linux header files, which cannot be compiled properly without __attribute__. The test must be overridden so that __attribute__ is assumed supported by the compiler.
After bootstrapping, you should set PKGSRC_COMPILER
in /etc/mk.conf
:
PKGSRC_COMPILER= icc
The default installation directory for icc is
/opt/intel_cc_80
, which
is also the pkgsrc default. If you have installed it into a different
directory, set ICCBASE
in
/etc/mk.conf
:
ICCBASE= /opt/icc
pkgsrc uses the static linking method of the runtime libraries provided by icc, so binaries can be run on other systems which do not have the shared libraries installed.
Libtool, however, extracts a list of libraries from the ld(1) command run when linking a C++ shared library and records it, throwing away the -Bstatic and -Bdynamic options interspersed between the libraries. This means that libtool-linked C++ shared libraries will have a runtime dependency on the icc libraries until this is fixed in libtool.
OpenBSD 3.0 and 3.2 are tested and supported.
Care should be taken so that the tools that this kit installs do not conflict with the OpenBSD userland tools. There are several steps:
OpenBSD stores its ports pkg database in
/var/db/pkg
. It is therefore
recommended that you choose a different location (e.g.
/usr/pkgdb
) by
using the --pkgdbdir option to the bootstrap script.
If you do not intend to use the OpenBSD ports tools, it's probably a good idea to move them out of the way to avoid confusion, e.g.
#
cd /usr/sbin
#
mv pkg_add pkg_add.orig
#
mv pkg_create pkg_create.orig
#
mv pkg_delete pkg_delete.orig
#
mv pkg_info pkg_info.orig
An example /etc/mk.conf
file will be placed in
/etc/mk.conf.example
file
when you use the bootstrap script. OpenBSD's make program uses
/etc/mk.conf
as well. You can work around this by enclosing all the pkgsrc-specific parts
of the file with:
.ifdef BSD_PKG_MK # pkgsrc stuff, e.g. insert defaults/mk.conf or similar here .else # OpenBSD stuff .endif
Solaris 2.6 through 9 are supported on both x86 and sparc. You will need a working C compiler. Both gcc 2.95.3 and Sun WorkShop 5 have been tested.
The following packages are required on Solaris 8 for the bootstrap process and to build packages.
SUNWsprot
SUNWarc
SUNWbtool
SUNWtoo
SUNWlibm
Please note that the use of GNU binutils on Solaris is not supported, as of June 2006.
Whichever compiler you use, please ensure the compiler tools and
your $prefix are in your PATH
. This includes
/usr/ccs/{bin,lib}
and e.g. /usr/pkg/{bin,sbin}
.
It makes life much simpler if you only use the same gcc consistently for building all packages.
It is recommended that an external gcc be used only for bootstrapping,
then either build gcc from
lang/gcc
or install a binary gcc
package, then remove gcc used during bootstrapping.
Binary packages of gcc can be found through http://www.sunfreeware.com/.
You will need at least the following packages installed (from WorkShop 5.0)
SPROcc - Sun WorkShop Compiler C 5.0
SPROcpl - Sun WorkShop Compiler C++ 5.0
SPROild - Sun WorkShop Incremental Linker
SPROlang - Sun WorkShop Compilers common components
You should set the following variables in your
mk.conf
file:
CC= cc CXX= CC CPP= cc -E CXXCPP= CC -E
The CPP
setting might break some
packages that use the C preprocessor for processing things other
than C source code.
To build 64-bit packages, you just need to have the
following lines in your mk.conf
file:
PKGSRC_COMPILER= sunpro ABI= 64
This setting has been tested for the SPARC architecture. Intel and AMD machines need some more work.
Sometimes, when using libtool,
/bin/ksh
crashes with a segmentation fault.
The workaround is to use another shell for the configure
scripts, for example by installing shells/bash
and adding the following lines
to your mk.conf
:
CONFIG_SHELL= ${LOCALBASE}/bin/bash WRAPPER_SHELL= ${LOCALBASE}/bin/bash
Then, rebuild the devel/libtool-base
package.
Table of Contents
Basically, there are two ways of using pkgsrc. The first is to only install the package tools and to use binary packages that someone else has prepared. This is the “pkg” in pkgsrc. The second way is to install the “src” of pkgsrc, too. Then you are able to build your own packages, and you can still use binary packages from someone else.
To use binary packages, you need some tools to manage
them. On NetBSD, these tools are already installed. On all other
operating systems, you need to install them first. For the
following platforms, prebuilt versions of the package tools
are available and can simply be downloaded and unpacked in the
/
directory:
Platform | URL |
---|---|
Solaris 9 | ftp://ftp0.mh.bbc.co.uk/pub/pkgsrc/packages/bootstrap-pkgsrc/ |
Solaris 10 | http://public.enst.fr/pkgsrc/packages/bootstrap-pkgsrc/ |
These pre-built package tools use
/usr/pkg
for the base directory, and
/var/db/pkg
for the database of installed
packages. If you cannot use these directories for whatever
reasons (maybe because you're not root), you have to build the
package tools yourself, which is explained in Section 3.2, “Bootstrapping pkgsrc”.
To install binary packages, you first need to know from where to get them. You can get them on CD-ROMs, DVDs, or via FTP or HTTP.
The binary packages can be found at the following locations.
Platform | URL |
---|---|
NetBSD | ftp://ftp.NetBSD.org/pub/NetBSD/packages/ |
Solaris 9 | ftp://ftp0.mh.bbc.co.uk/pub/pkgsrc/packages/ |
Solaris 10 | http://public.enst.fr/pkgsrc/packages/ |
Most of these directories contain binary packages for
multiple platforms. Select the appropriate subdirectories,
according to your machine architecture and operating system,
until you find a directory called All
. This
directory contains all the binary packages. Further, there are
subdirectories for categories that contain symbolic links that
point to the actual binary package in
../All
. This directory layout is used for
all package repositories, no matter if they are accessed via
HTTP, FTP, NFS, CD-ROM, or the local filesystem.
If you have the files on a CD-ROM or downloaded them to your hard disk, you can install them with the following command (be sure to su to root first):
#
pkg_add /path/to/package.tgz
If you have FTP access and you don't want to download the packages via FTP prior to installation, you can do this automatically by giving pkg_add an FTP URL:
#
pkg_add ftp://ftp.NetBSD.org/pub/NetBSD/packages/<OSVERSION>/<ARCH>/All/package
Note that any prerequisite packages needed to run the package in question will be installed, too, assuming they are present where you install from.
To save some typing, you can set the
PKG_PATH
environment variable to a semicolon-separated
list of paths (including remote URLs); trailing slashes are not allowed.
Additionally to the All
directory
there exists a vulnerable
directory to
which binary packages with known vulnerabilities are
moved, since removing them could cause missing dependencies. To
use these packages, add the vulnerable
directory to your PKG_PATH
. However, you should run
security/audit-packages
regularly,
especially after installing new packages, and verify that the
vulnerabilities are acceptable for your configuration. An example
PKG_PATH
would be:
ftp://ftp.NetBSD.org/pub/NetBSD/packages/<OSVERSION>/<ARCH>/All;ftp://ftp.NetBSD.org/pub/NetBSD/packages/<OSVERSION>/<ARCH>/vulnerable
Please note that semicolon (';') is a shell meta-character, so
you'll probably have to quote it.
After you've installed packages, be sure to have
/usr/pkg/bin
and /usr/pkg/sbin
in your
PATH
so you can actually start the just
installed program.
To deinstall a package, it does not matter whether it was
installed from source code or from a binary package. The
pkg_delete command does not know it anyway.
To delete a package, you can just run pkg_delete
package-name
. The package
name can be given with or without version number. Wildcards can
also be used to deinstall a set of packages, for example
*emacs*
. Be sure to include them in quotes,
so that the shell does not expand them before
pkg_delete
sees them.
The -r
option is very powerful: it
removes all the packages that require the package in question
and then removes the package itself. For example:
#
pkg_delete -r jpeg
will remove jpeg and all the packages that used it; this allows upgrading the jpeg package.
The pkg_info shows information about installed packages or binary package files.
The NetBSD Security-Officer and Packages Groups maintain a list of known security vulnerabilities to packages which are (or have been) included in pkgsrc. The list is available from the NetBSD FTP site at ftp://ftp.NetBSD.org/pub/pkgsrc/distfiles/vulnerabilities.
Through security/audit-packages
,
this list can be downloaded
automatically, and a security audit of all packages installed on a system
can take place.
There are two components to
security/audit-packages
. The first
component, “download-vulnerability-list”, is for downloading
the list of vulnerabilities from the NetBSD FTP site. The second
component, “audit-packages”, checks to see if any of your
installed packages are vulnerable. If a package is vulnerable, you
will see output similar to the following:
Package samba-2.0.9 has a local-root-shell vulnerability, see http://www.samba.org/samba/whatsnew/macroexploit.html
One can set up security/audit-packages
to download the
vulnerabilities
file daily, and include a package audit in the daily security script.
Details on this are located in the MESSAGE
file for security/audit-packages
.
Install pkgtools/pkglint
and run
lintpkgsrc with the “-i”
argument to check if your packages are up-to-date, e.g.
%
lintpkgsrc -i
... Version mismatch: 'tcsh' 6.09.00 vs 6.10.00
You can then use make update to update the package on your system and rebuild any dependencies.
The pkg_admin executes various administrative functions on the package system.
Please pay very careful attention to the warnings expressed in the pkg_add(1) manual page about the inherent dangers of installing binary packages which you did not create yourself, and the security holes that can be introduced onto your system by indiscriminate adding of such files.
The same warning of course applies to every package you install from source when you haven't completely read and understood the source code of the package, the compiler that is used to build the package and all the other tools that are involved.
After obtaining pkgsrc, the pkgsrc
directory now contains a set of packages, organized into
categories. You can browse the online index of packages, or run
make readme from the pkgsrc
directory to build local README.html
files for
all packages, viewable with any web browser such as www/lynx
or www/firefox
.
The default prefix for installed packages
is /usr/pkg
. If you wish to change this, you
should do so by setting LOCALBASE
in
mk.conf
. You should not try to use multiple
different LOCALBASE
definitions on the same
system (inside a chroot is an exception).
The rest of this chapter assumes that the package is already in pkgsrc. If it is not, see Part II, “The pkgsrc developer's guide” for instructions how to create your own packages.
To build packages from source, you need a working C compiler. On NetBSD, you need to install the “comp” and the “text” distribution sets. If you want to build X11-related packages, the “xbase” and “xcomp” distribution sets are required, too.
The first step for building a package is downloading the distfiles (i.e. the unmodified source). If they have not yet been downloaded, pkgsrc will fetch them automatically.
If you have all files that you need in the
distfiles
directory,
you don't need to connect. If the distfiles are on CD-ROM, you can
mount the CD-ROM on /cdrom
and add:
DISTDIR=/cdrom/pkgsrc/distfiles
to your mk.conf
.
By default a list of distribution sites will be randomly
intermixed to prevent huge load on servers which holding popular
packages (for example, SourceForge.net mirrors). Thus, every
time when you need to fetch yet another distfile all the mirrors
will be tried in new (random) order. You can turn this feature
off by setting MASTER_SORT_RANDOM=NO
(for
PKG_DEVELOPER
s it's already disabled).
You can overwrite some of the major distribution sites to
fit to sites that are close to your own. By setting one or two
variables you can modify the order in which the master sites are
accessed. MASTER_SORT
contains a whitespace
delimited list of domain suffixes.
MASTER_SORT_REGEX
is even more flexible, it
contains a whitespace delimited list of regular expressions. It
has higher priority than MASTER_SORT
. Have a
look at pkgsrc/mk/defaults/mk.conf
to find
some examples. This may save some of your bandwidth and
time.
You can change these settings either in your shell's environment, or,
if you want to keep the settings, by editing the
/etc/mk.conf
file,
and adding the definitions there.
If a package depends on many other packages (such as
meta-pkgs/kde3
), the build process may
alternate between periods of
downloading source, and compiling. To ensure you have all the source
downloaded initially you can run the command:
%
make fetch-list | sh
which will output and run a set of shell commands to fetch the
necessary files into the distfiles
directory. You can
also choose to download the files manually.
Once the software has downloaded, any patches will be applied, then it will be compiled for you. This may take some time depending on your computer, and how many other packages the software depends on and their compile time.
If using bootstrap or pkgsrc on a non-NetBSD system, use the pkgsrc bmake command instead of “make” in the examples in this guide.
For example, type
%
cd misc/figlet
%
make
at the shell prompt to build the various components of the package.
The next stage is to actually install the newly compiled program onto your system. Do this by entering:
%
make install
while you are still in the directory for whatever package you are installing.
Installing the package on your system may require you to be root. However, pkgsrc has a just-in-time-su feature, which allows you to only become root for the actual installation step.
That's it, the software should now be installed and setup for use. You can now enter:
%
make clean
to remove the compiled files in the work directory, as you shouldn't need them any more. If other packages were also added to your system (dependencies) to allow your program to compile, you can tidy these up also with the command:
%
make clean-depends
Taking the figlet utility as an example, we can install it on our system by building as shown in Appendix B, Build logs.
The program is installed under the default root of the
packages tree - /usr/pkg
. Should this not
conform to your tastes, set the LOCALBASE
variable in your environment, and it will use that value as the
root of your packages tree. So, to use
/usr/local
, set
LOCALBASE=/usr/local
in your environment.
Please note that you should use a directory which is dedicated to
packages and not shared with other programs (i.e., do not try and
use LOCALBASE=/usr
). Also, you should not try
to add any of your own files or directories (such as
src/
, obj/
, or
pkgsrc/
) below the
LOCALBASE
tree. This is to prevent possible
conflicts between programs and other files installed by the
package system and whatever else may have been installed
there.
Some packages look in /etc/mk.conf
to
alter some configuration options at build time. Have a look at
pkgsrc/mk/defaults/mk.conf
to get an overview
of what will be set there by default. Environment variables such
as LOCALBASE
can be set in
/etc/mk.conf
to save having to remember to
set them each time you want to use pkgsrc.
Occasionally, people want to “look under the covers” to see what is going on when a package is building or being installed. This may be for debugging purposes, or out of simple curiosity. A number of utility values have been added to help with this.
If you invoke the make(1) command with
PKG_DEBUG_LEVEL=2
, then a huge amount of
information will be displayed. For example,
make patch PKG_DEBUG_LEVEL=2
will show all the commands that are invoked, up to and including the “patch” stage.
If you want to know the value of a certain make(1)
definition, then the VARNAME
definition
should be used, in conjunction with the show-var
target. e.g. to show the expansion of the make(1)
variable LOCALBASE
:
%
make show-var VARNAME=LOCALBASE
/usr/pkg%
If you want to install a binary package that you've either
created yourself (see next section), that you put into
pkgsrc/packages manually or that is located on a remote FTP
server, you can use the "bin-install" target. This target will
install a binary package - if available - via pkg_add(1),
else do a make package. The list of remote FTP
sites searched is kept in the variable
BINPKG_SITES
, which defaults to
ftp.NetBSD.org. Any flags that should be added to pkg_add(1)
can be put into BIN_INSTALL_FLAGS
. See
pkgsrc/mk/defaults/mk.conf
for more
details.
A final word of warning: If you set up a system that has a
non-standard setting for LOCALBASE
, be sure to
set that before any packages are installed, as you cannot use
several directories for the same purpose. Doing so will result in
pkgsrc not being able to properly detect your installed packages,
and fail miserably. Note also that precompiled binary packages are
usually built with the default LOCALBASE
of
/usr/pkg
, and that you should
not install any if you use a non-standard
LOCALBASE
.
Table of Contents
The whole pkgsrc system is configured in a single file, usually
called mk.conf
. In which directory pkgsrc looks for
that file depends on the installation. On NetBSD, when you use
make(1) from the base system, it is in the directory
/etc/
. In all other cases the default location is
${PREFIX}/etc/
, depending on where you told the
bootstrap program to install the binary packages.
During the bootstrap, an example configuration file is created. To
use that, you have to create the directory
${PREFIX}/etc
and copy the example file
there.
The format of the configuration file is that of the usual
BSD-style Makefile
s. The whole pkgsrc configuration
is done by setting variables in this file. Note that you can define all
kinds of variables, and no special error checking (for example for
spelling mistakes) takes place, so you have to try it out to see if it
works.
In this section, you can find some variables that apply to all
pkgsrc packages. A complete list of the variables that can be
configured by the user is available in
mk/defaults/mk.conf
, together with some
comments that describe each variable's intent.
LOCALBASE
: Where
packages will be installed. The default is
/usr/pkg
. Do not mix binary packages
with different LOCALBASE
s!
CROSSBASE
: Where
“cross” category packages will be
installed. The default is
${LOCALBASE}/cross
.
X11BASE
: Where
X11 is installed on the system. The default is
/usr/X11R6
.
DISTDIR
: Where to store the
downloaded copies of the original source distributions used
for building pkgsrc packages. The default is
${PKGSRCDIR}/distfiles
.
PKG_DBDIR
: Where the
database about installed packages is stored. The default is
/var/db/pkg
.
MASTER_SITE_OVERRIDE
:
If set, override the packages'
MASTER_SITES
with this value.
MASTER_SITE_BACKUP
:
Backup location(s) for distribution files and patch files
if not found locally or in
${MASTER_SITES}
or
${PATCH_SITES}
respectively.
The defaults are
ftp://ftp.NetBSD.org/pub/NetBSD/packages/distfiles/${DIST_SUBDIR}/
and
ftp://ftp.freebsd.org/pub/FreeBSD/distfiles/${DIST_SUBDIR}/
.
BINPKG_SITES
:
List of sites carrying binary pkgs. rel
and
arch
are replaced with OS
release (“2.0”, etc.) and architecture
(“mipsel”, etc.).
ACCEPTABLE_LICENSES
:
List of acceptable licenses. Whenever you try to build a package
whose license is not in this list, you will get an error message
that includes instructions on how to change this variable.
XXX
PACKAGES
: The top level
directory for the binary packages. The default is
${PKGSRCDIR}/packages
.
WRKOBJDIR
:
The top level directory where, if defined, the separate
working directories will get created, and symbolically
linked to from ${WRKDIR}
(see below).
This is useful for building packages on several
architectures, then ${PKGSRCDIR}
can be NFS-mounted while ${WRKOBJDIR}
is local to every architecture. (It should be noted that
PKGSRCDIR
should not be set by the user
— it is an internal definition which refers to the
root of the pkgsrc tree. It is possible to have many
pkgsrc tree instances.)
LOCALPATCHES
:
Directory for local patches that aren't part of pkgsrc.
See Section 10.3, “patches/*” for more
information.
PKGMAKECONF
: Location of
the mk.conf
file used by a package's
BSD-style Makefile. If this is not set,
MAKECONF
is set to
/dev/null
to avoid picking up
settings used by builds in /usr/src
.
DEPENDS_TARGET
:
By default, dependencies are only installed, and no binary
package is created for them. You can set this variable to
package
to automatically create binary
packages after installing dependencies.
By default, pkgsrc will use GCC to build packages. This may be overridden by setting the following variables in /etc/mk.conf:
PKGSRC_COMPILER
:This is a list of values specifying the chain of compilers to invoke when building packages. Valid values are:
distcc
:
distributed C/C++ (chainable)
ccache
:
compiler cache (chainable)
gcc
:
GNU C/C++ Compiler
mipspro
:
Silicon Graphics, Inc. MIPSpro (n32/n64)
mipspro
:
Silicon Graphics, Inc. MIPSpro (o32)
sunpro
:
Sun Microsystems, Inc. WorkShip/Forte/Sun ONE Studio
The default is
“gcc
”. You can use
ccache
and/or
distcc
with an appropriate
PKGSRC_COMPILER
setting,
e.g. “ccache gcc
”. This
variable should always be terminated with a value for
a real compiler. Note that only one real compiler
should be listed (e.g. “sunpro gcc
”
is not allowed).
GCC_REQD
:This specifies the minimum version of GCC to use when building packages. If the system GCC doesn't satisfy this requirement, then pkgsrc will build and install one of the GCC packages to use instead.
If you wish to set the CFLAGS
variable,
please make sure to use the +=
operator
instead of the =
operator:
CFLAGS+= -your -flags
Using CFLAGS=
(i.e. without the
“+”) may lead to problems with packages that need
to add their own flags. Also, you may want to take a look at
the devel/cpuflags
package if
you're interested in optimization for the current CPU.
If you want to pass flags to the linker, both in the configure
step and the build step, you can do this in two ways. Either set
LDFLAGS
or LIBS
. The difference
between the two is that LIBS
will be appended to
the command line, while LDFLAGS
come earlier.
LDFLAGS
is pre-loaded with rpath settings for ELF
machines depending on the setting of USE_IMAKE
or
the inclusion of mk/x11.buildlink3.mk
. As with
CFLAGS
, if you do not wish to override these
settings, use the +=
operator:
LDFLAGS+= -your -linkerflags
XXX
PKG_DEVELOPER
:
Run some sanity checks that package developers want:
make sure patches apply with zero fuzz
run check-shlibs to see that all binaries will find their shared libs.
PKG_DEBUG_LEVEL
: The level
of debugging output which is displayed whilst making and
installing the package. The default value for this is 0,
which will not display the commands as they are executed
(normal, default, quiet operation); the value 1 will display
all shell commands before their invocation, and the value 2
will display both the shell commands before their invocation,
and their actual execution progress with set
-x will be displayed.
Some packages have build time options, usually to select between different dependencies, enable optional support for big dependencies or enable experimental features.
To see which options, if any, a package supports, and which options are mutually exclusive, run make show-options, for example:
The following options are supported by this package: ssl Enable SSL support. Exactly one of the following gecko options is required: firefox Use firefox as gecko rendering engine. mozilla Use mozilla as gecko rendering engine. At most one of the following database options may be selected: mysql Enable support for MySQL database. pgsql Enable support for PostgreSQL database. These options are enabled by default: firefox These options are currently enabled: mozilla ssl
The following variables can be defined in
/etc/mk.conf
to select which options to
enable for a package: PKG_DEFAULT_OPTIONS
,
which can be used to select or disable options for all packages
that support them, and
PKG_OPTIONS.
,
which can be used to select or disable options specifically for
package pkgbase
pkgbase
. Options listed in
these variables are selected, options preceded by “-”
are disabled. A few examples:
$
grep "PKG.*OPTION" /etc/mk.conf
PKG_DEFAULT_OPTIONS= -arts -dvdread -esound
PKG_OPTIONS.kdebase= debug -sasl
PKG_OPTIONS.apache= suexec
The following settings are consulted in the order given, and the last setting that selects or disables an option is used:
the default options as suggested by the package maintainer
the options implied by the settings of legacy variables (see below)
PKG_DEFAULT_OPTIONS
PKG_OPTIONS.
pkgbase
For groups of mutually exclusive options, the last option selected is used, all others are automatically disabled. If an option of the group is explicitly disabled, the previously selected option, if any, is used. It is an error if no option from a required group of options is selected, and building the package will fail.
Before the options framework was introduced, build options
were selected by setting a variable (often named
USE_
) in
FOO
/etc/mk.conf
for each option. To ease
transition to the options framework for the user, these legacy
variables are converted to the appropriate options setting
(PKG_OPTIONS.
)
automatically. A warning is issued to prompt the user to update
pkgbase
/etc/mk.conf
to use the options framework
directly. Support for the legacy variables will be removed
eventually.
Table of Contents
Once you have built and installed a package, you can create a binary package which can be installed on another system with pkg_add(1). This saves having to build the same package on a group of hosts and wasting CPU time. It also provides a simple means for others to install your package, should you distribute it.
To create a binary package, change into the appropriate directory in pkgsrc, and run make package:
#
cd misc/figlet
#
make package
This will build and install your package (if not already done),
and then build a binary package from what was installed. You can
then use the pkg_* tools to manipulate
it. Binary packages are created by default in
/usr/pkgsrc/packages
, in the form of a
gzipped tar file. See Section B.2, “Packaging figlet” for a
continuation of the above misc/figlet
example.
See Chapter 20, Submitting and Committing for information on how to submit such a binary package.
If you want to get a full set of precompiled binary packages, this section describes how to get them. Beware that the bulk build will remove all currently installed packages from your system!
Having an FTP server configured either on the machine doing the bulk builds or on a nearby NFS server can help to make the packages available to other machines that can then save time by installing only the binary packages. See ftpd(8) for more information. If you use a remote NFS server's storage, be sure to not actually compile on NFS storage, as this slows things down a lot.
The build.conf
file is the main
configuration file for bulk builds. You can configure how your
copy of pkgsrc is kept up to date, how the distfiles are
downloaded, how the packages are built and how the report is
generated. You can find an annotated example file in
pkgsrc/mk/bulk/build.conf-example
. To use
it, copy build.conf-example
to
build.conf
and edit it, following the
comments in that file.
You may want to set variables in
/etc/mk.conf
.
Look at pkgsrc/mk/defaults/mk.conf
for
details of the default settings. You will want to ensure that
ACCEPTABLE_LICENSES
meet your local policy.
As used in this example, _ACCEPTABLE=yes
accepts all licenses.
PACKAGES?= ${_PKGSRCDIR}/packages/${MACHINE_ARCH} WRKOBJDIR?= /usr/tmp/pkgsrc # build here instead of in pkgsrc BSDSRCDIR= /usr/src BSDXSRCDIR= /usr/xsrc # for x11/xservers OBJHOSTNAME?= yes # use work.`hostname` FAILOVER_FETCH= yes # insist on the correct checksum PKG_DEVELOPER?= yes _ACCEPTABLE= yes
Some options that are especially useful for bulk builds
can be found at the top lines of the file
mk/bulk/bsd.bulk-pkg.mk
. The most useful
options of these are briefly described here.
If you are on a slow machine, you may want to
set USE_BULK_BROKEN_CHECK
to
“no”.
If you are doing bulk builds from a read-only
copy of pkgsrc, you have to set BULKFILESDIR
to the directory where all log files are created. Otherwise the
log files are created in the pkgsrc directory.
Another important variable is
BULK_PREREQ
, which is a list of packages that
should be always available while building other
packages.
Some other options are scattered in the pkgsrc infrastructure:
ALLOW_VULNERABLE_PACKAGES
should be set to yes
. The purpose of the bulk
builds is creating binary packages, no matter if they are
vulnerable or not. When uploading the packages to a public
server, the vulnerable packages will be put into a directory of
their own. Leaving this variable unset would prevent the bulk
build system from even trying to build them, so possible
building errors would not show up.
CHECK_FILES
(pkgsrc/mk/check/check-files.mk
) can be set to
“yes” to check that the installed set of files
matches the PLIST
.
CHECK_INTERPRETER
(pkgsrc/mk/check/check-interpreter.mk
) can be set to
“yes” to check that the installed
“#!”-scripts will find their
interpreter.
PKGSRC_RUN_TEST
can be
set to “yes
” to run each
package's self-test before installing it. Note that some
packages make heavy use of “good” random
numbers, so you need to assure that the machine on which you
are doing the bulk builds is not completely idle. Otherwise
some test programs will seem to hang, while they are just
waiting for new random data to be
available.
It is possible to configure the bulk build to perform
certain site-specific tasks at the end of the pre-build
stage. If the file
pre-build.local
exists in
/usr/pkgsrc/mk/bulk
, it will be executed
(as a sh(1) script) at the end of the usual pre-build
stage. An example use of
pre-build.local
is to have the line:
echo "I do not have enough disk space to build this pig." \ > misc/openoffice/$BROKENF
to prevent the system from trying to build a particular package which requires nearly 3 GB of disk space.
As /usr/pkg
will be completely
deleted at the start of bulk builds, make sure your login
shell is placed somewhere else. Either drop it into
/usr/local/bin
(and adjust your login
shell in the passwd file), or (re-)install it via
pkg_add(1) from /etc/rc.local
, so
you can login after a reboot (remember that your current
process won't die if the package is removed, you just can't
start any new instances of the shell any more). Also, if you
use NetBSD earlier than 1.5, or you still want to use the pkgsrc
version of ssh for some reason, be sure to install ssh before
starting it from rc.local
:
( cd /usr/pkgsrc/security/ssh ; make bulk-install ) if [ -f /usr/pkg/etc/rc.d/sshd ]; then /usr/pkg/etc/rc.d/sshd fi
Not doing so will result in you being not able to log in via ssh after the bulk build is finished or if the machine gets rebooted or crashes. You have been warned! :)
Make sure you don't need any of the packages still installed.
During the bulk build, all packages will be removed!
Be sure to remove all other things that might
interfere with builds, like some libs installed in
/usr/local
, etc. then become root and type:
#
cd /usr/pkgsrc
#
sh mk/bulk/build
If for some reason your last build didn't complete (power failure, system panic, ...), you can continue it by running:
#
sh mk/bulk/build restart
At the end of the bulk build, you will get a summary via mail,
and find build logs in the directory specified by
FTP
in the build.conf
file.
The bulk builds consist of three steps:
The script updates your pkgsrc tree via (anon)cvs, then cleans out any broken distfiles, and removes all packages installed.
This is basically “make bulk-package” with an optimised order in which packages will be built. Packages that don't require other packages will be built first, and packages with many dependencies will be built later.
Generates a report that's placed in the directory
specified in the build.conf
file
named broken.html
, a short version
of that report will also be mailed to the build's
admin.
During the build, a list of broken packages will be compiled
in /usr/pkgsrc/.broken
(or
.../.broken.${MACHINE}
if
OBJMACHINE
is set), individual build logs
of broken builds can be found in the package's
directory. These files are used by the bulk-targets to mark
broken builds to not waste time trying to rebuild them, and
they can be used to debug these broken package builds
later.
Currently, roughly the following requirements are valid for NetBSD 2.0/i386:
10 GB - distfiles (NFS ok)
8 GB - full set of all binaries (NFS ok)
5 GB - temp space for compiling (local disk recommended)
Note that all pkgs will be de-installed as soon as they are turned into a binary package, and that sources are removed, so there is no excessively huge demand to disk space. Afterwards, if the package is needed again, it will be installed via pkg_add(1) instead of building again, so there are no cycles wasted by recompiling.
If you don't want all the packages nuked from a machine (rendering it useless for anything but pkg compiling), there is the possibility of doing the package bulk build inside a chroot environment.
The first step is to set up a chroot sandbox,
e.g. /usr/sandbox
. This can be done by
using null mounts, or manually.
There is a shell script called
pkgsrc/mk/bulk/mksandbox
which will set
up the sandbox environment using null mounts. It will also
create a script called sandbox
in the
root of the sandbox environment, which will allow the null
mounts to be activated using the sandbox
mount command and deactivated using the
sandbox umount command.
To set up a sandbox environment by hand, after extracting all
the sets from a NetBSD installation or doing a make
distribution DESTDIR=/usr/sandbox in
/usr/src/etc
, be sure the following items
are present and properly configured:
Kernel
#
cp /netbsd /usr/sandbox
/dev/*
#
cd /usr/sandbox/dev ; sh MAKEDEV all
/etc/resolv.conf
(for security/smtpd
and mail):
#
cp /etc/resolv.conf /usr/sandbox/etc
Working(!) mail config (hostname, sendmail.cf):
#
cp /etc/mail/sendmail.cf /usr/sandbox/etc/mail
/etc/localtime
(for security/smtpd
):
#
ln -sf /usr/share/zoneinfo/UTC /usr/sandbox/etc/localtime
/usr/src
(system sources,
e. g. for sysutils/aperture
):
#
ln -s ../disk1/cvs .
#
ln -s cvs/src-2.0 src
Create /var/db/pkg
(not part of default install):
#
mkdir /usr/sandbox/var/db/pkg
Create /usr/pkg
(not part of default install):
#
mkdir /usr/sandbox/usr/pkg
Checkout pkgsrc via cvs into
/usr/sandbox/usr/pkgsrc
:
#
cd /usr/sandbox/usr
#
cvs -d anoncvs@anoncvs.NetBSD.org:/cvsroot checkout -d -P pkgsrc
Do not mount/link this to the copy of your pkgsrc tree you do development in, as this will likely cause problems!
Make
/usr/sandbox/usr/pkgsrc/packages
and
.../distfiles
point somewhere
appropriate. NFS- and/or nullfs-mounts may come in handy!
Edit /etc/mk.conf
, see Section 6.3.1.2, “/etc/mk.conf”.
Adjust mk/bulk/build.conf
to suit your needs.
When the chroot sandbox is set up, you can start the build with the following steps:
#
cd /usr/sandbox/usr/pkgsrc
#
sh mk/bulk/do-sandbox-build
This will just jump inside the sandbox and start building. At
the end of the build, mail will be sent with the results of
the build. Created binary pkgs will be in
/usr/sandbox/usr/pkgsrc/packages
(wherever that points/mounts to/from).
In addition to building a complete set of all packages in
pkgsrc, the pkgsrc/mk/bulk/build
script
may be used to build a subset of the packages contained in
pkgsrc. By setting SPECIFIC_PKGS
in /etc/mk.conf
, the variables
SITE_SPECIFIC_PKGS
HOST_SPECIFIC_PKGS
GROUP_SPECIFIC_PKGS
USER_SPECIFIC_PKGS
will define the set of packages which should be built. The bulk build code will also include any packages which are needed as dependencies for the explicitly listed packages.
One use of this is to do a bulk build with
SPECIFIC_PKGS
in a chroot sandbox
periodically to have a complete set of the binary packages
needed for your site available without the overhead of
building extra packages that are not needed.
This section describes how pkgsrc developers can upload binary pkgs built by bulk builds to ftp.NetBSD.org.
If you would like to automatically create checksum files for the
binary packages you intend to upload, remember to set
MKSUMS=yes
in your
mk/bulk/build.conf
.
If you would like to PGP sign the checksum files (highly
recommended!), remember to set
SIGN_AS=username@NetBSD.org
in your
mk/bulk/build.conf
. This will prompt you for
your GPG password to sign the files before uploading everything.
Then, make sure that you have RSYNC_DST
set properly in your mk/bulk/build.conf
file, i.e. adjust it to something like one of the following:
RSYNC_DST=ftp.NetBSD.org:/pub/NetBSD/packages/packages-200xQy/NetBSD-a.b.c/arch/upload
Please use appropriate values for "packages-200xQy", "NetBSD-a.b.c" and "arch" here. If your login on ftp.NetBSD.org is different from your local login, write your login directly into the variable, e.g. my local account is "feyrer", but for my login "hubertf", I use:
RSYNC_DST=hubertf@ftp.NetBSD.org:/pub/NetBSD/packages/packages-200xQy/NetBSD-a.b.c/arch/upload
A separate upload
directory is used
here to allow "closing" the directory during upload. To do
so, run the following command on ftp.NetBSD.org next:
nbftp% mkdir -p -m 750 /pub/NetBSD/packages/packages-200xQy/NetBSD-a.b.c/arch/upload
Please note that /pub/NetBSD/packages
is
only appropriate for packages for the NetBSD operating
system. Binary packages for other operating systems should go
into /pub/pkgsrc
.
Before uploading the binary pkgs, ssh authentication needs to be set up. This example shows how to set up temporary keys for the root account inside the sandbox (assuming that no keys should be present there usually):
#
chroot /usr/sandbox
chroot-#
rm $HOME/.ssh/id-dsa*
chroot-#
ssh-keygen -t dsa
chroot-#
cat $HOME/.ssh/id-dsa.pub
Now take the output of id-dsa.pub
and
append it to your ~/.ssh/authorized_keys
file on ftp.NetBSD.org. You can remove the key after the
upload is done!
Next, test if your ssh connection really works:
chroot-#
ssh ftp.NetBSD.org date
Use "-l yourNetBSDlogin" here as appropriate!
Now after all this works, you can exit the sandbox and start the upload:
chroot-#
exit
#
cd /usr/sandbox/usr/pkgsrc
#
sh mk/bulk/do-sandbox-upload
The upload process may take quite some time. Use ls(1) or
du(1) on the FTP server to monitor progress of the
upload. The upload script will take care of not uploading
restricted packages and putting vulnerable packages into the
vulnerable
subdirectory.
After the upload has ended, first thing is to revoke ssh access:
nbftp% vi ~/.ssh/authorized_keys
Gdd:x!
Use whatever is needed to remove the key you've entered
before! Last, move the uploaded packages out of the
upload
directory to have them accessible
to everyone:
nbftp%cd /pub/NetBSD/packages/packages-200xQy/NetBSD-a.b.c/arch
nbftp%mv upload/* .
nbftp%rmdir upload
nbftp%chmod 755 .
After your pkgsrc bulk-build has completed, you may wish to
create a CD-ROM set of the resulting binary packages to assist
in installing packages on other machines. The
pkgtools/cdpack
package provides
a simple tool for creating the ISO 9660 images.
cdpack arranges the packages on the CD-ROMs in a
way that keeps all the dependencies for a given package on the same
CD as that package.
Complete documentation for cdpack is found in the cdpack(1)
man page. The following short example assumes that the binary
packages are left in
/usr/pkgsrc/packages/All
and that
sufficient disk space exists in /u2
to
hold the ISO 9660 images.
#
mkdir /u2/images
#
pkg_add /usr/pkgsrc/packages/All/cdpack
#
cdpack /usr/pkgsrc/packages/All /u2/images
If you wish to include a common set of files
(COPYRIGHT
, README
,
etc.) on each CD in the collection, then you need to create a
directory which contains these files. e.g.
#
mkdir /tmp/common
#
echo "This is a README" > /tmp/common/README
#
echo "Another file" > /tmp/common/COPYING
#
mkdir /tmp/common/bin
#
echo "#!/bin/sh" > /tmp/common/bin/myscript
#
echo "echo Hello world" >> /tmp/common/bin/myscript
#
chmod 755 /tmp/common/bin/myscript
Now create the images:
#
cdpack -x /tmp/common /usr/pkgsrc/packages/All /u2/images
Each image will contain README
,
COPYING
, and bin/myscript
in their root directories.
Table of Contents
The files that are installed by pkgsrc are organized in a way that
is similar to what you find in the /usr
directory
of the base system. But some details are different. This is because
pkgsrc initially came from FreeBSD and had adopted its file system
hierarchy. Later it was largely influenced by NetBSD. But no matter
which operating system you are using pkgsrc with, you can expect the
same layout for pkgsrc.
There are mainly four root directories for pkgsrc, which are all
configurable in the bootstrap/bootstrap
script.
When pkgsrc has been installed as root, the default locations
are:
LOCALBASE= /usr/pkg PKG_SYSCONFBASE= /usr/pkg/etc VARBASE= /var PKG_DBDIR= /var/db/pkg
In unprivileged mode (when pkgsrc has been installed as any other user), the default locations are:
LOCALBASE= ${HOME}/pkg PKG_SYSCONFBASE= ${HOME}/pkg/etc VARBASE= ${HOME}/pkg/var PKG_DBDIR= ${HOME}/pkg/var/db/pkg
What these four directories are for, and what they look like is explained below.
LOCALBASE
corresponds to the
/usr
directory in the base system. It is the
“main” directory where the files are installed and contains
the well-known subdirectories like bin
,
include
, lib
,
share
and
sbin
.
VARBASE
corresponds to
/var
in the base system. Some programs (especially
games, network daemons) need write access to it during normal
operation.
PKG_SYSCONFDIR
corresponds to
/etc
in the base system. It contains configuration
files of the packages, as well as pkgsrc's mk.conf
itself.
The following directories exist in a typical pkgsrc installation
in ${LOCALBASE}
.
bin
Contains executable programs that are intended to be directly used by the end user.
emul
Contains files for the emulation layers of various other operating systems, especially for NetBSD.
etc
(the usual location of
${PKG_SYSCONFDIR}
)Contains the configuration files.
include
Contains headers for the C and C++ programming languages.
info
Contains GNU info files of various packages.
lib
Contains shared and static libraries.
libdata
Contains data files that don't change after
installation. Other data files belong into
${VARBASE}
.
libexec
Contains programs that are not intended to be used by end users, such as helper programs or network daemons.
libexec/cgi-bin
Contains programs that are intended to be executed as CGI scripts by a web server.
man
(the usual value of
${PKGMANDIR}
)Contains brief documentation in form of manual pages.
sbin
Contains programs that are intended to be used only by the super-user.
share
Contains platform-independent data files that don't change after installation.
share/doc
Contains documentation files provided by the packages.
share/examples
Contains example files provided by the packages. Among
others, the original configuration files are saved here and copied to
${PKG_SYSCONFDIR}
during
installation.
share/examples/rc.d
Contains the original files for rc.d scripts.
var
(the usual location of
${VARBASE}
)Contains files that may be modified after installation.
Table of Contents
CFLAGS
?This section contains hints, tips & tricks on special things in pkgsrc that we didn't find a better place for in the previous chapters, and it contains items for both pkgsrc users and developers.
The following mailing lists may be of interest to pkgsrc users:
pkgsrc-users: This is a general purpose list for most issues regarding pkgsrc, regardless of platform, e.g. soliciting user help for pkgsrc configuration, unexpected build failures, using particular packages, upgrading pkgsrc installations, questions regarding the pkgsrc release branches, etc. General announcements or proposals for changes that impact the pkgsrc user community, e.g. major infrastructure changes, new features, package removals, etc., may also be posted.
pkgsrc-bulk: A list where the results of pkgsrc bulk builds are sent and discussed.
pkgsrc-changes: This list is for those who are interested in getting a commit message for every change committed to pkgsrc. It is also available in digest form, meaning one daily message containing all commit messages for changes to the package source tree in that 24 hour period.
To subscribe, do:
%
echo subscribelistname
| mail majordomo@NetBSD.org
Archives for all these mailing lists are available from http://mail-index.NetBSD.org/.
Pkgviews is tightly integrated with buildlink. You can find a
pkgviews User's guide in
pkgsrc/mk/buildlink3/PKGVIEWS_UG
.
The directory pkgsrc/pkgtools
contains
a number of useful utilities for both users and developers of pkgsrc. This
section attempts only to make the reader aware of the utilities and when
they might be useful, and not to duplicate the documentation that comes
with each package.
Utilities used by pkgsrc (automatically installed when needed):
pkgtools/x11-links
:
Symlinks for use by buildlink.
OS tool augmentation (automatically installed when needed):
pkgtools/digest
:
Calculates various kinds of checksums (including SHA1).
pkgtools/libnbcompat
:
Compatibility library for pkgsrc tools.
pkgtools/mtree
: Installed on
non-BSD systems due to lack of native mtree.
pkgtools/pkg_install
:
Up-to-date replacement for
/usr/sbin/pkg_install
, or for use on operating
systems where pkg_install is not present.
Utilities used by pkgsrc (not automatically installed):
pkgtools/pkg_tarup
:
Create a binary package from an
already-installed package. Used by make replace to
save the old package.
pkgtools/dfdisk
:
Adds extra functionality to pkgsrc, allowing it to fetch distfiles
from multiple locations. It currently supports the following
methods: multiple CD-ROMs and network FTP/HTTP connections.
pkgtools/xpkgwedge
: Put X11
packages someplace else (enabled by default).
devel/cpuflags
: Determine the
best compiler flags to optimise code for your current CPU and
compiler.
Utilities for keeping track of installed packages, being up to date, etc:
pkgtools/pkg_chk
: Reports on
packages whose installed versions do not match the latest pkgsrc
entries.
pkgtools/pkgdep
: Makes
dependency graphs of packages, to aid in choosing a strategy for
updating.
pkgtools/pkgdepgraph
: Makes
graphs from the output of pkgtools/pkgdep
(uses graphviz).
pkgtools/pkglint
: The
pkglint(1) program checks a pkgsrc entry for errors, lintpkgsrc(1)
does various checks on the complete pkgsrc system.
pkgtools/pkgsurvey
: Report what
packages you have installed.
Utilities for people maintaining or creating individual packages:
pkgtools/pkgdiff
: Automate
making and maintaining patches for a package (includes pkgdiff,
pkgvi, mkpatches, etc.).
pkgtools/rpm2pkg
,
pkgtools/url2pkg
: Aids in
converting to pkgsrc.
pkgtools/gensolpkg
: Convert
pkgsrc to a Solaris package.
Utilities for people maintaining pkgsrc (or: more obscure pkg utilities)
pkgtools/pkg_comp
: Build
packages in a chrooted area.
pkgtools/libkver
: Spoof
kernel version for chrooted cross builds.
If you want to use pkgsrc as non-root user, you can set some
variables to make pkgsrc work under these conditions. At the very least,
you need to set UNPRIVILEGED
to “yes”; this
will turn on unprivileged mode and set multiple related variables to allow
installation of packages as non-root.
In case the defaults are not enough, you may want to tune some other
variables used. For example, if the automatic user/group detection leads
to incorrect values (or not the ones you would like to use), you can change
them by setting UNPRIVILEGED_USER
and
UNPRIVILEGED_GROUP
respectively.
As regards bootstrapping, please note that the
bootstrap script will ease non-root configuration when
given the “--ignore-user-check” flag, as it will choose and
use multiple default directories under ~/pkg
as the
installation targets. These directories can be overridden by the
“--prefix” flag provided by the script, as well as some others
that allow finer tuning of the tree layout.
By default, resuming transfers in pkgsrc is disabled, but you can
enable this feature by adding the option
PKG_RESUME_TRANSFERS=YES
into
/etc/mk.conf
. If, during a fetch step, an incomplete
distfile is found, pkgsrc will try to resume it.
You can also
use a different program than the default ftp(1) by changing the
FETCH_CMD
variable. Don't forget to set
FETCH_RESUME_ARGS
and
FETCH_OUTPUT_ARGS
if you are not using default
values.
For example, if you want to use
wget
to resume downloads, you'll have to use something
like:
FETCH_CMD= wget FETCH_BEFORE_ARGS= --passive-ftp FETCH_RESUME_ARGS= -c FETCH_OUTPUT_ARGS= -O
If you want to use XFree86 from pkgsrc instead of your system's own
X11 (/usr/X11R6
, /usr/openwin
,
...), you will have to add the following line into
/etc/mk.conf
:
X11_TYPE=XFree86
If you want to use X.org from pkgsrc instead of your system's own X11
(/usr/X11R6
, /usr/openwin
, ...)
you will have to add the following line into
/etc/mk.conf
:
X11_TYPE=xorg
The DragonFly operating system defaults to using this X.org X11 implementation from pkgsrc.
If you are sitting behind a firewall which does not allow direct connections to Internet hosts (i.e. non-NAT), you may specify the relevant proxy hosts. This is done using an environment variable in the form of a URL, e.g. in Amdahl, the machine “orpheus.amdahl.com” is one of the firewalls, and it uses port 80 as the proxy port number. So the proxy environment variables are:
ftp_proxy=ftp://orpheus.amdahl.com:80/ http_proxy=http://orpheus.amdahl.com:80/
This depends on which utility is used to retrieve distfiles. From
bsd.pkg.mk
, FETCH_CMD
is assigned
the first available command from the following list:
${LOCALBASE}/bin/ftp
/usr/bin/ftp
On a default NetBSD installation, this will be
/usr/bin/ftp
, which automatically tries passive
connections first, and falls back to active connections if the server
refuses to do passive. For the other tools, add the following to your
/etc/mk.conf
file:
PASSIVE_FETCH=1
.
Having that option present will prevent
/usr/bin/ftp
from falling back to active
transfers.
You would like to download all the distfiles in a single batch from work or university, where you can't run a make fetch. There is an archive of distfiles on ftp.NetBSD.org, but downloading the entire directory may not be appropriate.
The answer here is to do a make fetch-list in
/usr/pkgsrc
or one of its subdirectories, carry the
resulting list to your machine at work/school and use it there. If you
don't have a NetBSD-compatible ftp(1) (like tnftp) at work, don't
forget to set FETCH_CMD
to something that fetches a
URL:
At home:
%
cd /usr/pkgsrc
%
make fetch-list FETCH_CMD=wget DISTDIR=/tmp/distfiles >/tmp/fetch.sh
%
scp /tmp/fetch.sh work:/tmp
At work:
%
sh /tmp/fetch.sh
then tar up /tmp/distfiles
and take it
home.
If you have a machine running NetBSD, and you want to get all distfiles (even ones that aren't for your machine architecture), you can do so by using the above-mentioned make fetch-list approach, or fetch the distfiles directly by running:
%
make mirror-distfiles
If you even decide to ignore
NO_{SRC,BIN}_ON_{FTP,CDROM}
, then you can get everything
by running:
%
make fetch NO_SKIP=yes
When compiling the pkgtools/pkg_install
package, you get the error from make that it doesn't know how to make
/usr/share/tmac/tmac.andoc
? This indicates that
you don't have installed the “text” set (nroff, ...) from
the NetBSD base distribution on your machine. It is recommended to do
that to format man pages.
In the case of the pkgtools/pkg_install
package, you
can get away with setting NOMAN=YES
either in the
environment or in /etc/mk.conf
.
You didn't install the compiler set, comp.tgz
,
when you installed your NetBSD machine. Please get and install it, by
extracting it in /
:
#
cd /
#
tar --unlink -zxvpf .../comp.tgz
comp.tgz
is part of every NetBSD release. Get
the one that corresponds to your release (determine via uname
-r).
When installing packages as non-root user and using the just-in-time
su(1) feature of pkgsrc, it can become annoying to type in the root
password for each required package installed. To avoid this, the sudo
package can be used, which does password caching over a limited time. To
use it, install sudo (either as binary package or from
security/sudo
) and then put the following
into your /etc/mk.conf
:
.if exists(${LOCALBASE}/bin/sudo) SU_CMD= ${LOCALBASE}/bin/sudo /bin/sh -c .endif
As the system administrator, you can choose where configuration files
are installed. The default settings make all these files go into
${PREFIX}/etc
or some of its subdirectories; this may
be suboptimal depending on your expectations (e.g., a read-only,
NFS-exported PREFIX
with a need of per-machine
configuration of the provided packages).
In order to change the defaults, you can modify the
PKG_SYSCONFBASE
variable (in
/etc/mk.conf
) to point to your preferred configuration
directory; some common examples include /etc
or
/etc/pkg
.
Furthermore, you can change this value on a per-package basis by
setting the PKG_SYSCONFDIR.${PKG_SYSCONFVAR}
variable.
PKG_SYSCONFVAR
's value usually matches the name of the
package you would like to modify, that is, the contents of
PKGBASE
.
Note that after changing these settings, you must rebuild and reinstall any affected packages.
Please be aware that there can often be bugs in third-party software,
and some of these bugs can leave a machine vulnerable to exploitation by
attackers. In an effort to lessen the exposure, the NetBSD packages team
maintains a database of known-exploits to packages which have at one time
been included in pkgsrc. The database can be downloaded automatically, and
a security audit of all packages installed on a system can take place. To
do this, install the security/audit-packages
package. It has two
components:
download-vulnerability-list, an easy way to download a list of the security vulnerabilities information. This list is kept up to date by the NetBSD security officer and the NetBSD packages team, and is distributed from the NetBSD ftp server:
ftp://ftp.NetBSD.org/pub/NetBSD/packages/distfiles/pkg-vulnerabilities
audit-packages, an easy way to audit the current machine, checking each vulnerability which is known. If a vulnerable package is installed, it will be shown by output to stdout, including a description of the type of vulnerability, and a URL containing more information.
Use of the security/audit-packages
package is strongly recommended! After
“audit-packages” is installed, please read
the package's message, which you can get by running pkg_info -D
audit-packages
.
If this package is installed, pkgsrc builds will use it to perform a security check before building any package. See Section 5.2, “Variables affecting the build process” for ways to control this check.
When you add your own preferences to the
CFLAGS
variable in your
mk.conf
, these flags are passed in
environment variables to the ./configure
scripts and to make(1). Some package authors ignore the
CFLAGS
from the environment variable by
overriding them in the Makefile
s of their
package.
Currently there is no solution to this problem. If you
really need the package to use your CFLAGS
you should run make patch in the package
directory and then inspect any Makefile
and
Makefile.in
for whether they define
CFLAGS
explicitly. Usually you can remove
these lines. But be aware that some “smart”
programmers write so bad code that it only works for the
specific combination of CFLAGS
they have
chosen.
Make sure that your copy of pkgsrc is consistent. A case that occurs often is that people only update pkgsrc in parts, because of performance reasons. Since pkgsrc is one large system, not a collection of many small systems, there are sometimes changes that only work when the whole pkgsrc tree is updated.
Make sure that you don't have any CVS conflicts. Search for “<<<<<<” or “>>>>>>” in all your pkgsrc files.
Make sure that you don't have old copies of the packages extracted. Run make clean clean-depends to verify this.
If the problem still exists, write a mail to the
pkgsrc-users
mailing list.
You have modified a file from pkgsrc, and someone else has
modified that same file afterwards in the CVS repository. Both changes
are in the same region of the file, so when you updated pkgsrc, the
cvs
command marked the conflicting changes in the
file. Because of these markers, the file is no longer a valid
Makefile
.
Have a look at that file, and if you don't need your local changes anymore, you can remove that file and run cvs -q update -dP in that directory to download the current version.
This part of the book deals with creating and modifying packages. It starts with a “HOWTO”-like guide on creating a new package. The remaining chapters are more like a reference manual for pkgsrc.
Table of Contents
Makefile
smk.conf
Table of Contents
When you find a package that is not yet in pkgsrc, you most likely have a URL from where you can download the source code. Starting with this URL, creating a package involves only a few steps.
First, install the packages pkgtools/url2pkg
and pkgtools/pkglint
.
Then, choose one of the top-level directories as the
category in which you want to place your package. You can also create a
directory of your own (maybe called local
). In that
category directory, create another directory for your package and change
into it.
Run the program url2pkg, which will ask
you for a URL. Enter the URL of the distribution file (in most cases a
.tar.gz
file) and watch how the basic ingredients
of your package are created automatically. The distribution file is
extracted automatically to fill in some details in the
Makefile
that would otherwise have to be done
manually.
Examine the extracted files to determine the dependencies of
your package. Ideally, this is mentioned in some
README
file, but things may differ. For each of
these dependencies, look where it exists in pkgsrc, and if there is a
file called buildlink3.mk
in that directory, add a
line to your package Makefile
which includes that
file just before the last line. If the
buildlink3.mk
file does not exist, add a
DEPENDS
line to the Makefile, which specifies the
version of the dependency and where it can be found in pkgsrc. This line
should be placed in the third paragraph. If the dependency is only
needed for building the package, but not when using it, use
BUILD_DEPENDS
instead of DEPENDS
.
Your package may then look like this:
[...] BUILD_DEPENDS+= lua>=5.0:../../lang/lua DEPENDS+= screen-[0-9]*:../../misc/screen DEPENDS+= screen>=4.0:../../misc/screen [...] .include "../../category
/package
/buildlink3.mk" .include "../../devel/glib2/buildlink3.mk" .include "../../mk/bsd.pkg.mk"
Run pkglint to see what things still need to be done to make your package a “good” one. If you don't know what pkglint's warnings want to tell you, try pkglint --explain or pkglint -e, which outputs additional explanations.
In many cases the package is not yet ready to build. You can find instructions for the most common cases in the next section, Section 9.1, “Common types of packages”. After you have followed the instructions over there, you can hopefully continue here.
Run bmake clean to clean the working
directory from the extracted files. Besides these files, a lot of cache
files and other system information has been saved in the working
directory, which may become wrong after you edited the
Makefile
.
Now, run bmake to build the package. For the various things that can go wrong in this phase, consult Chapter 18, Making your package work.
When the package builds fine, the next step is to install the package. Run bmake install and hope that everything works.
Up to now, the file PLIST
, which
contains a list of the files that are installed by the package, is
nearly empty. Run bmake print-PLIST
>PLIST to generate a probably correct list. Check
the file using your preferred text editor to see if the list of
files looks plausible.
Run pkglint again to see if the generated
PLIST
contains garbage or not.
When you ran bmake install, the package
has been registered in the database of installed files, but with an
empty list of files. To fix this, run bmake deinstall
and bmake install again. Now the package is
registered with the list of files from
PLIST
.
Run bmake package to create a binary package from the set of installed files.
Simple Perl modules are handled automatically by url2pkg, including dependencies.
Looking at the file pkgsrc/doc/TODO
, I saw
that the “nvu” package has not yet been imported into
pkgsrc. As the description says it has to do with the web, the obvious
choice for the category is “www”.
$
mkdir www/nvu$
cd www/nvu
The web site says that the sources are available as a tar file, so I fed that URL to the url2pkg program:
$
url2pkg http://cvs.nvu.com/download/nvu-1.0-sources.tar.bz2
My editor popped up, and I added a PKGNAME
line
below the DISTNAME
line, as the package name should
not have the word “sources” in it. I also filled in the
MAINTAINER
, HOMEPAGE
and
COMMENT
fields. Then the package
Makefile
looked like that:
# $NetBSD$ # DISTNAME= nvu-1.0-sources PKGNAME= nvu-1.0 CATEGORIES= www MASTER_SITES= http://cvs.nvu.com/download/ EXTRACT_SUFX= .tar.bz2 MAINTAINER= rillig@NetBSD.org HOMEPAGE= http://cvs.nvu.com/ COMMENT= Web Authoring System # url2pkg-marker (please do not remove this line.) .include "../../mk/bsd.pkg.mk"
Then, I quit the editor and watched pkgsrc downloading a large source archive:
url2pkg> Running "make makesum" ... => Required installed package digest>=20010302: digest-20060826 found => Fetching nvu-1.0-sources.tar.bz2 Requesting http://cvs.nvu.com/download/nvu-1.0-sources.tar.bz2 100% |*************************************| 28992 KB 150.77 KB/s 00:00 ETA 29687976 bytes retrieved in 03:12 (150.77 KB/s) url2pkg> Running "make extract" ... => Required installed package digest>=20010302: digest-20060826 found => Checksum SHA1 OK for nvu-1.0-sources.tar.bz2 => Checksum RMD160 OK for nvu-1.0-sources.tar.bz2 work.bacc -> /tmp/roland/pkgsrc/www/nvu/work.bacc ===> Installing dependencies for nvu-1.0 ===> Overriding tools for nvu-1.0 ===> Extracting for nvu-1.0 url2pkg> Adjusting the Makefile. Remember to correct CATEGORIES, HOMEPAGE, COMMENT, and DESCR when you're done! Good luck! (See pkgsrc/doc/pkgsrc.txt for some more help :-)
Now that the package has been extracted, let's see what's inside
it. The package has a README.txt
, but that only
says something about mozilla, so it's probably useless for seeing what
dependencies this package has. But since there is a GNU configure script
in the package, let's hope that it will complain about everything it
needs.
$
bmake
=> Required installed package digest>=20010302: digest-20060826 found
=> Checksum SHA1 OK for nvu-1.0-sources.tar.bz2
=> Checksum RMD160 OK for nvu-1.0-sources.tar.bz2
===> Patching for nvu-1.0
===> Creating toolchain wrappers for nvu-1.0
===> Configuring for nvu-1.0
[...]
configure: error: Perl 5.004 or higher is required.
[...]
WARNING: Please add USE_TOOLS+=perl to the package Makefile.
[...]
That worked quite well. So I opened the package Makefile in my
editor, and since it already has a USE_TOOLS
line, I
just appended “perl” to it. Since the dependencies of the
package have changed now, and since a perl wrapper is automatically
installed in the “tools” phase, I need to build the package
from scratch.
$
bmake clean ===> Cleaning for nvu-1.0$
bmake [...] *** /tmp/roland/pkgsrc/www/nvu/work.bacc/.tools/bin/make is not \ GNU Make. You will not be able to build Mozilla without GNU Make. [...]
So I added “gmake” to the
USE_TOOLS
line and tried again (from scratch).
[...] checking for GTK - version >= 1.2.0... no *** Could not run GTK test program, checking why... [...]
Now to the other dependencies. The first question is: Where is the GTK package hidden in pkgsrc?
$
echo ../../*/gtk* [many packages ...]$
echo ../../*/gtk ../../x11/gtk$
echo ../../*/gtk2 ../../x11/gtk2$
echo ../../*/gtk2/bui* ../../x11/gtk2/buildlink3.mk
The first try was definitely too broad. The second one had exactly one result, which is very good. But there is one pitfall with GNOME packages. Before GNOME 2 had been released, there were already many GNOME 1 packages in pkgsrc. To be able to continue to use these packages, the GNOME 2 packages were imported as separate packages, and their names usually have a “2” appended. So I checked whether this was the case here, and indeed it was.
Since the GTK2 package has a buildlink3.mk
file, adding the dependency is very easy. I just inserted an
.include
line before the last line of the package
Makefile
, so that it now looks like this:
[...] .include "../../x11/gtk2/buildlink3.mk" .include "../../mk/bsd.pkg.mk
After another bmake clean && bmake, the answer was:
[...] checking for gtk-config... /home/roland/pkg/bin/gtk-config checking for GTK - version >= 1.2.0... no *** Could not run GTK test program, checking why... *** The test program failed to compile or link. See the file config.log for the *** exact error that occured. This usually means GTK was incorrectly installed *** or that you have moved GTK since it was installed. In the latter case, you *** may want to edit the gtk-config script: /home/roland/pkg/bin/gtk-config configure: error: Test for GTK failed. [...]
In this particular case, the assumption that “every package
prefers GNOME 2” had been wrong. The first of the lines above
told me that this package really wanted to have the GNOME 1 version of
GTK. If the package had looked for GTK2, it would have looked for
pkg-config instead of gtk-config.
So I changed the x11/gtk2
to
x11/gtk
in the package Makefile
,
and tried again.
[...] cc -o xpidl.o -c -DOSTYPE=\"NetBSD3\" -DOSARCH=\"NetBSD\" -I../../../dist/include/xpcom -I../../../dist/include -I/tmp/roland/pkgsrc/www/nvu/work.bacc/mozilla/dist/include/nspr -I/usr/X11R6/include -fPIC -DPIC -I/home/roland/pkg/include -I/usr/include -I/usr/X11R6/include -Wall -W -Wno-unused -Wpointer-arith -Wcast-align -Wno-long-long -pedantic -O2 -I/home/roland/pkg/include -I/usr/include -Dunix -pthread -pipe -DDEBUG -D_DEBUG -DDEBUG_roland -DTRACING -g -I/home/roland/pkg/include/glib/glib-1.2 -I/home/roland/pkg/lib/glib/include -I/usr/pkg/include/orbit-1.0 -I/home/roland/pkg/include -I/usr/include -I/usr/X11R6/include -include ../../../mozilla-config.h -DMOZILLA_CLIENT -Wp,-MD,.deps/xpidl.pp xpidl.c In file included from xpidl.c:42: xpidl.h:53:24: libIDL/IDL.h: No such file or directory In file included from xpidl.c:42: xpidl.h:132: error: parse error before "IDL_ns" [...]
The package still does not find all of its dependencies. Now the
question is: Which package provides the
libIDL/IDL.h
header file?
$
echo ../../*/*idl* ../../devel/py-idle ../../wip/idled ../../x11/acidlaunch$
echo ../../*/*IDL* ../../net/libIDL
Let's take the one from the second try. So I included the
../../net/libIDL/buildlink3.mk
file and tried
again. But the error didn't change. After digging through some of the
code, I concluded that the build process of the package was broken and
couldn't have ever worked, but since the Mozilla source tree is quite
large, I didn't want to fix it. So I added the following to the package
Makefile
and tried again:
CPPFLAGS+= -I${BUILDLINK_PREFIX.libIDL}/include/libIDL-2.0 BUILDLINK_TRANSFORM+= -l:IDL:IDL-2
The latter line is needed because the package expects the library
libIDL.so
, but only
libIDL-2.so
is available. So I told the compiler
wrapper to rewrite that on the fly.
The next problem was related to a recent change of the FreeType
interface. I looked up in www/seamonkey
which patch files were relevant for this issue and copied them to the
patches
directory. Then I retried, fixed the
patches so that they applied cleanly and retried again. This time,
everything worked.
Table of Contents
Whenever you're preparing a package, there are a number of files involved which are described in the following sections.
Building, installation and creation of a binary package are all
controlled by the package's Makefile
.
The Makefile
describes various things about
a package, for example from where to get it, how to configure,
build, and install it.
A package Makefile
contains several
sections that describe the package.
In the first section there are the following variables, which should appear exactly in the order given here. The order and grouping of the variables is mostly historical and has no further meaning.
DISTNAME
is the basename of the
distribution file to be downloaded from the package's
website.
PKGNAME
is the name of the
package, as used by pkgsrc. You only need to provide it if
DISTNAME
(which is the default) is not a good
name for the package in pkgsrc. Usually it is the pkgsrc
directory name together with the version number. It must match the
regular expression
^[A-Za-z0-9][A-Za-z0-9-_.+]*$
, that is, it
starts with a letter or digit, and contains only letters, digits,
dashes, underscores, dots and plus signs.
SVR4_PKGNAME
is the name of
the package file to create if the PKGNAME
isn't unique on a SVR4 system. The default is
PKGNAME
, which may be shortened when you use
pkgtools/gensolpkg
. Only add
SVR4_PKGNAME
if PKGNAME
does not produce an unique package name on a SVR4 system.
The length of SVR4_PKGNAME
is limited to 5
characters.
CATEGORIES
is a list of categories
which the package fits in. You can choose any of the top-level
directories of pkgsrc for it.
Currently the following values are available for
CATEGORIES
. If more than
one is used, they need to be separated by spaces:
archivers cross geography meta-pkgs security audio databases graphics misc shells benchmarks devel ham multimedia sysutils biology editors inputmethod net textproc cad emulators lang news time chat finance mail parallel wm comms fonts math pkgtools www converters games mbone print x11
MASTER_SITES
,
DYNAMIC_MASTER_SITES
,
DIST_SUBDIR
, EXTRACT_SUFX
and DISTFILES
are discussed in detail in
Section 16.5, “The fetch phase”.
The second section contains information about separately downloaded patches, if any.
PATCHFILES:
Name(s) of additional files that contain distribution patches.
There is no default. pkgsrc will look for them at
PATCH_SITES
.
They will automatically be uncompressed before patching if
the names end with .gz
or
.Z
.
PATCH_SITES
:
Primary location(s) for distribution patch files (see
PATCHFILES
below) if not found locally.
The third section contains the following variables.
MAINTAINER
is the email address
of the person who feels responsible for this package, and who is
most likely to look at problems or questions regarding this
package which have been reported with send-pr(1). Other
developers should contact the MAINTAINER
before
making major changes to the package. When packaging a new program,
set MAINTAINER
to yourself. If you really can't
maintain the package for future updates, set it to
<pkgsrc-users@NetBSD.org>
.
HOMEPAGE
is a URL where users can
find more information about the package.
COMMENT
is a one-line
description of the package (should not include the package
name).
Other variables that affect the build:
WRKSRC
: The directory where the
interesting distribution files of the package are found. The
default is ${WRKDIR}/${DISTNAME}
, which
works for most packages.
If a package doesn't create a subdirectory for itself
(most GNU software does, for instance), but extracts itself in
the current directory, you should set
WRKSRC=${WRKDIR}
.
If a package doesn't create a subdirectory with the
name of DISTNAME
but some different name,
set WRKSRC
to point to the proper name in
${WRKDIR}
, for example
WRKSRC=${WRKDIR}/${DISTNAME}/unix
. See
lang/tcl
and x11/tk
for other examples.
The name of the working directory created by pkgsrc is
taken from the WRKDIR_BASENAME
variable. By default, its value is
work
. If you want to use the same
pkgsrc tree for building different kinds of binary packages,
you can change the variable according to your needs. Two
other variables handle common cases of setting
WRKDIR_BASENAME
individually. If
OBJHOSTNAME
is defined in
/etc/mk.conf
, the first component of
the host's name is attached to the directory name. If
OBJMACHINE
is defined, the platform name
is attached, which might look like
work.i386
or
work.sparc
.
Please pay attention to the following gotchas:
Add MANCOMPRESSED
if man pages are
installed in compressed form by the package. For packages using
BSD-style makefiles which honor MANZ, there is
MANCOMPRESSED_IF_MANZ
.
Replace /usr/local
with
“${PREFIX}” in all files (see patches,
below).
If the package installs any info files, see Section 18.6.6, “Packages installing info files”.
The distinfo
file contains the message
digest, or checksum, of each distfile needed for the package. This
ensures that the distfiles retrieved from the Internet have not been
corrupted during transfer or altered by a malign force to introduce
a security hole. Due to recent rumor about weaknesses of digest
algorithms, all distfiles are protected using both SHA1 and RMD160
message digests, as well as the file size.
The distinfo
file also contains the
checksums for all the patches found in the
patches
directory (see Section 10.3, “patches/*”).
To regenerate the distinfo
file, use the
make makedistinfo or make mdi
command.
Some packages have different sets of distfiles depending on
the platform, for example www/navigator
). These are kept in the same
distinfo
file and care should be taken when
upgrading such a package to ensure distfile information is not
lost.
Many packages still don't work out-of-the box on the various
platforms that are supported by pkgsrc. Therefore, a number of custom
patch files are needed to make the package work. These patch files are
found in the patches/
directory.
In the patch phase, these patches are
applied to the files in WRKSRC
directory after
extracting them, in alphabetic
order, so patch-aa
is applied before
patch-ab
, etc.
The patch-*
files should be in
diff -bu format, and apply without a fuzz to avoid
problems. (To force patches to apply with fuzz you can set
PATCH_FUZZ_FACTOR=-F2
). Furthermore, each patch
should contain only changes for a single file, and no file should be
patched by more than one patch file. This helps to keep future
modifications simple.
Each patch file is structured as follows: In the first line, there is the RCS Id of the patch itself. The second line should be empty for aesthetic reasons. After that, there should be a comment for each change that the patch does. There are a number of standard cases:
Patches that replace the ==
operator for test(1) with =
in shell scripts
should contain a reference to , to avoid
redundant explanations.
Patches for commonly known vulnerabilities should mention the vulnerability ID (CAN, CVE).
Patches that change source code should mention the platform and other environment (for example, the compiler) that the patch is needed for.
In all other cases, the patch should be commented so that any developer who knows the code of the application can make some use of the patch. Special care should be taken for the upstream developers, since we generally want that they accept our patches, so we have less work in the future.
One important thing to mention is to pay attention that no RCS
IDs get stored in the patch files, as these will cause problems when
later checked into the NetBSD CVS tree. Use the
pkgdiff command from the pkgtools/pkgdiff
package to avoid these
problems.
For even more automation, we recommend using
mkpatches from the same package to make a
whole set of patches. You just have to backup files before you
edit them to filename.orig
, e.g. with
cp -p filename filename.orig or, easier, by
using pkgvi again from the same package. If
you upgrade a package this way, you can easily compare the new
set of patches with the previously existing one with
patchdiff. Copy the patches you want to use
or update from the work/.newpatches
directory to patches/
.
When you have finished a package, remember to generate
the checksums for the patch files by using the make
makepatchsum command, see Section 10.2, “distinfo
”.
When adding a patch that corrects a problem in the distfile (rather than e.g. enforcing pkgsrc's view of where man pages should go), send the patch as a bug report to the maintainer. This benefits non-pkgsrc users of the package, and usually makes it possible to remove the patch in future version.
If you want to share patches between multiple packages
in pkgsrc, e.g. because they use the same distfiles, set
PATCHDIR
to the path where the patch files
can be found, e.g.:
PATCHDIR= ${.CURDIR}/../xemacs/patches
Patch files that are distributed by the author or other
maintainers can be listed in
PATCHFILES
.
If it is desired to store any patches that should not be
committed into pkgsrc, they can be kept outside the pkgsrc
tree in the $LOCALPATCHES
directory. The
directory tree there is expected to have the same
“category/package” structure as pkgsrc, and
patches are expected to be stored inside these dirs (also
known as $LOCALPATCHES/$PKGPATH
). For
example, if you want to keep a private patch for
pkgsrc/graphics/png
, keep it in
$LOCALPATCHES/graphics/png/mypatch
. All
files in the named directory are expected to be patch files,
and they are applied after pkgsrc patches are
applied.
When fixing a portability issue in the code do not use preprocessor magic to check for the current operating system nor platform. Doing so hurts portability to other platforms because the OS-specific details are not abstracted appropriately.
The general rule to follow is: instead of checking for the
operating system the application is being built on, check for the
specific features you need. For example,
instead of assuming that kqueue is available under NetBSD and
using the __NetBSD__
macro to conditionalize
kqueue support, add a check that detects kqueue itself —
yes, this generally involves patching the
configure script. There is absolutely nothing
that prevents some OSes from adopting interfaces from other OSes
(e.g. Linux implementing kqueue), something that the above checks
cannot take into account.
Of course, checking for features generally involves more work on the developer's side, but the resulting changes are cleaner and there are chances they will work on many other platforms. Not to mention that there are higher chances of being later integrated into the mainstream sources. Remember: It doesn't work unless it is right!
Some typical examples:
Table 10.1. Patching examples
Where | Incorrect | Correct |
---|---|---|
configure script |
case ${target_os} in netbsd*) have_kvm=yes ;; *) have_kvm=no ;; esac |
AC_CHECK_LIB(kvm, kvm_open, have_kvm=yes, have_kvm=no) |
C source file |
#if defined(__NetBSD__) # include <sys/event.h> #endif |
#if defined(HAVE_SYS_EVENT_H) # include <sys/event.h> #endif |
C source file |
int monitor_file(...) { #if defined(__NetBSD__) int fd = kqueue(); ... #else ... #endif } |
int monitor_file(...) { #if defined(HAVE_KQUEUE) int fd = kqueue(); ... #else ... #endif } |
For more information, please read the Making packager-friendly software article (part 1, part 2). It summarizes multiple details on how to make software easier to package; all the suggestions in it were collected from our experience in pkgsrc work, so they are possibly helpful when creating patches too.
Always, always, always feed back any portability fixes or improvements you do to a package to the mainstream developers. This is the only way to get their attention on portability issues and to ensure that future versions can be built out-of-the box on NetBSD. Furthermore, any user that gets newer distfiles will get the fixes straight from the packaged code.
This generally involves cleaning up the patches (because sometimes the patches that are added to pkgsrc are quick hacks), filling bug reports in the appropriate trackers for the projects and working with the mainstream authors to accept your changes. It is extremely important that you do it so that the packages in pkgsrc are kept simple and thus further changes can be done without much hassle.
Support the idea of free software!
DESCR
A multi-line description of the piece of software. This should include any credits where they are due. Please bear in mind that others do not share your sense of humour (or spelling idiosyncrasies), and that others will read everything that you write here.
PLIST
This file governs the files that are installed on your system: all the binaries, manual pages, etc. There are other directives which may be entered in this file, to control the creation and deletion of directories, and the location of inserted files. See Chapter 12, PLIST issues for more information.
INSTALL
This shell script is invoked twice by pkg_add(1).
First time after package extraction and before files are
moved in place, the second time after the files to install
are moved in place. This can be used to do any custom
procedures not possible with @exec commands in
PLIST
. See pkg_add(1) and
pkg_create(1) for more information. See also Section 14.1, “Files and directories outside the installation prefix”.
DEINSTALL
This script is executed before and after any files are removed. It is this script's responsibility to clean up any additional messy details around the package's installation, since all pkg_delete knows is how to delete the files created in the original distribution. See pkg_delete(1) and pkg_create(1) for more information.
MESSAGE
This file is displayed after installation of the package.
Useful for things like legal notices on almost-free
software and hints for updating config files after
installing modules for apache, PHP etc.
Please note that you can modify variables in it easily by using
MESSAGE_SUBST
in the package's
Makefile
:
MESSAGE_SUBST+= SOMEVAR="somevalue"
replaces "${SOMEVAR}" with “somevalue” in
MESSAGE
. By default, substitution is
performed for PKGNAME
,
PKGBASE
, PREFIX
,
LOCALBASE
, X11PREFIX
,
X11BASE
,
PKG_SYSCONFDIR
,
ROOT_GROUP
, and
ROOT_GROUP
.
You can display a different or additional files by
setting the MESSAGE_SRC
variable. Its
default is MESSAGE
, if the file
exists.
ALTERNATIVES
FIXME: There is no documentation on the alternatives framework.
Makefile.common
This file contains arbitrary things that could
also go into a Makefile
, but its purpose is
to be used by more than one package. This file should only be
used when the packages that will use the file are known in
advance. For other purposes it is often better to write a
*.mk
file and give it a good name that
describes what it does.
buildlink3.mk
This file contains the dependency information for the buildlink3 framework (see Chapter 13, Buildlink methodology).
hacks.mk
This file contains workarounds for compiler bugs
and similar things. It is included automatically by the pkgsrc
infrastructure, so you don't need an extra
.include
line for
it.
options.mk
This file contains the code for the
package-specific options (see Chapter 15, Options handling) that can be
selected by the user. If a package has only one or two options,
it is equally acceptable to put the code directly into the
Makefile
.
When you type make, the distribution files are
unpacked into the directory denoted by
WRKDIR
. It can be removed by running
make clean. Besides the sources, this
directory is also used to keep various timestamp files.
The directory gets removed completely on clean.
The default is ${.CURDIR}/work
or ${.CURDIR}/work.${MACHINE_ARCH}
if OBJMACHINE
is set.
If you have any files that you wish to be placed in the package prior
to configuration or building, you could place these files here and use
a ${CP} command in the
“pre-configure” target to achieve
this. Alternatively, you could simply diff the file against
/dev/null
and use the patch mechanism to manage
the creation of this file.
If you want to share files in this way with other
packages, set the FILESDIR
variable to point
to the other package's files
directory,
e.g.:
FILESDIR=${.CURDIR}/../xemacs/files
Table of Contents
Pkgsrc consists of many Makefile
fragments,
each of which forms a well-defined part of the pkgsrc system. Using
the make(1) system as a programming language for a big system
like pkgsrc requires some discipline to keep the code correct and
understandable.
The basic ingredients for Makefile
programming are variables (which are actually macros) and shell
commands. Among these shell commands may even be more complex ones
like awk(1) programs. To make sure that every shell command runs
as intended it is necessary to quote all variables correctly when they
are used.
This chapter describes some patterns, that appear quite often in
Makefile
s, including the pitfalls that come along
with them.
When you are creating a file as a target of a rule, always write the data to a temporary file first and finally rename that file. Otherwise there might occur an error in the middle of generating the file, and when the user runs make(1) for the second time, the file exists and will not be regenerated properly. Example:
wrong: @echo "line 1" > ${.TARGET} @echo "line 2" >> ${.TARGET} @false correct: @echo "line 1" > ${.TARGET}.tmp @echo "line 2" >> ${.TARGET}.tmp @false @mv ${.TARGET}.tmp ${.TARGET}
When you run make wrong twice, the file
wrong
will exist, although there was an error
message in the first run. On the other hand, running make
correct gives an error message twice, as expected.
You might remember that make(1) sometimes removes
${.TARGET}
in case of error, but this only
happens when it is interrupted, for example by pressing
^C
. This does not happen
when one of the commands fails (like false(1) above).
Makefile
variables contain strings that
can be processed using the five operators ``='', ``+='', ``?='',
``:='', and ``!='', which are described in the make(1) man
page.
When a variable's value is parsed from a
Makefile
, the hash character ``#'' and the
backslash character ``\'' are handled specially. If a backslash is
followed by a newline, any whitespace immediately in front of the
backslash, the backslash, the newline, and any whitespace
immediately behind the newline are replaced with a single space. A
backslash character and an immediately following hash character are
replaced with a single hash character. Otherwise, the backslash is
passed as is. In a variable assignment, any hash character that is
not preceded by a backslash starts a comment that continues upto the
end of the logical line.
Note: Because of this parsing algorithm
the only way to create a variable consisting of a single backslash
is using the ``!='' operator, for example: BACKSLASH!=echo "\\"
.
So far for defining variables. The other thing you can do with variables is evaluating them. A variable is evaluated when it is part of the right side of the ``:='' or the ``!='' operator, or directly before executing a shell command which the variable is part of. In all other cases, make(1) performs lazy evaluation, that is, variables are not evaluated until there's no other way. The ``modifiers'' mentioned in the man page also evaluate the variable.
Some of the modifiers split the string into words and then operate on the words, others operate on the string as a whole. When a string is split into words, it is split as you would expect it from sh(1).
No rule without exception—the .for loop does not follow the shell quoting rules but splits at sequences of whitespace.
There are several types of variables that should be handled differently. Strings and two types of lists.
Strings can contain arbitrary
characters. Nevertheless, you should restrict yourself to only
using printable characters. Examples are
PREFIX
and
COMMENT
.
Internal lists are lists that
are never exported to any shell command. Their elements are
separated by whitespace. Therefore, the elements themselves cannot
have embedded whitespace. Any other characters are allowed.
Internal lists can be used in .for loops.
Examples are DEPENDS
and
BUILD_DEPENDS
.
External lists are lists that
may be exported to a shell command. Their elements can contain any
characters, including whitespace. That's why they cannot be used
in .for loops. Examples are
DISTFILES
and
MASTER_SITES
.
All variable names starting with an underscore
are reserved for use by the pkgsrc infrastructure. They shall
not be used by package
Makefile
s.
In .for loops you should use lowercase variable names for the iteration variables.
All list variables should have a ``plural''
name, e.g. PKG_OPTIONS
or
DISTFILES
.
This section presents you with some code snippets you should use in your own code. If you don't find anything appropriate here, you should test your code and add it here.
STRING= foo * bar `date` INT_LIST= # empty ANOTHER_INT_LIST= apache-[0-9]*:../../www/apache EXT_LIST= # empty ANOTHER_EXT_LIST= a=b c=d INT_LIST+= ${STRING} # 1 INT_LIST+= ${ANOTHER_INT_LIST} # 2 EXT_LIST+= ${STRING:Q} # 3 EXT_LIST+= ${ANOTHER_EXT_LIST} # 4
When you add a string to an external list (example 3), it must be quoted. In all other cases, you must not add a quoting level. You must not merge internal and external lists, unless you are sure that all entries are correctly interpreted in both lists.
EXT_LIST= # empty .for i in ${INT_LIST} EXT_LIST+= ${i:Q}"" .endfor
This code converts the internal list
INT_LIST
into the external list
EXT_LIST
. As the elements of an internal list
are unquoted they must be quoted here. The reason for appending
""
is explained below.
Sometimes you may want to print an arbitrary string. There are many ways to get it wrong and only few that can handle every nastiness.
STRING= foo bar < > * `date` $$HOME ' " EXT_LIST= string=${STRING:Q} x=second\ item all: echo ${STRING} # 1 echo "${STRING}" # 2 echo "${STRING:Q}" # 3 echo ${STRING:Q} # 4 echo x${STRING:Q} | sed 1s,.,, # 5 printf "%s\\n" ${STRING:Q}"" # 6 env ${EXT_LIST} /bin/sh -c 'echo "$$string"; echo "$$x"'
Example 1 leads to a syntax error in the shell, as the characters are just copied.
Example 2 leads to a syntax error too, and if you leave out
the last " character from ${STRING}
,
date(1) will be executed. The $HOME
shell
variable would be evaluated, too.
Example 3 outputs each space character preceded by a backslash (or not), depending on the implementation of the echo(1) command.
Example 4 handles correctly every string that does not start with a dash. In that case, the result depends on the implementation of the echo(1) command. As long as you can guarantee that your input does not start with a dash, this form is appropriate.
Example 5 handles even the case of a leading dash correctly.
Example 6 also works with every string and is the light-weight solution, since it does not involve a pipe, which has its own problems.
The EXT_LIST
does not need to be quoted
because the quoting has already been done when adding elements to
the list.
As internal lists shall not be passed to the shell, there is no example for it.
There are many possible sources of wrongly quoted variables. This section lists some of the commonly known ones.
Whenever you use the value of a list, think
about what happens to leading or trailing whitespace. If the
list is a well-formed shell expression, you can apply the
:M*
modifier to strip leading and trailing
whitespace from each word. The :M
operator
first splits its argument according to the rules of the shell,
and then creates a new list consisting of all words that match
the shell glob expression *
, that is: all.
One class of situations where this is needed is when adding a
variable like CPPFLAGS
to
CONFIGURE_ARGS
. If the configure script
invokes other configure scripts, it strips the leading and
trailing whitespace from the variable and then passes it to the
other configure scripts. But these configure scripts expect the
(child) CPPFLAGS
variable to be the same as
the parent CPPFLAGS
. That's why we better
pass the CPPFLAGS
value properly trimmed. And
here is how we do it:
CPPFLAGS= # empty CPPFLAGS+= -Wundef -DPREFIX=\"${PREFIX:Q}\" CPPFLAGS+= ${MY_CPPFLAGS} CONFIGURE_ARGS+= CPPFLAGS=${CPPFLAGS:M*:Q} all: echo x${CPPFLAGS:Q}x # leading and trailing whitespace echo x${CONFIGURE_ARGS}x # properly trimmed
The example above contains one bug: The
${PREFIX}
is a properly quoted shell
expression, but there is the C compiler after it, which also
expects a properly quoted string (this time in C syntax). The
version above is therefore only correct if
${PREFIX}
does not have embedded backslashes
or double quotes. If you want to allow these, you have to add
another layer of quoting to each variable that is used as a C
string literal. You cannot use the :Q
operator for it, as this operator only works for the
shell.
Whenever a variable can be empty, the
:Q
operator can have surprising results. Here
are two completely different cases which can be solved with the
same trick.
EMPTY= # empty empty_test: for i in a ${EMPTY:Q} c; do \ echo "$$i"; \ done for_test: .for i in a:\ a:\test.txt echo ${i:Q} echo "foo" .endfor
The first example will only print two of the three lines
we might have expected. This is because
${EMPTY:Q}
expands to the empty string, which
the shell cannot see. The workaround is to write
${EMPTY:Q}""
. This pattern can be often found
as ${TEST} -z ${VAR:Q}
or as ${TEST}
-f ${FNAME:Q}
(both of these are wrong).
The second example will only print three lines instead of
four. The first line looks like a:\ echo foo
.
This is because the backslash of the value
a:\
is interpreted as a line-continuation by
make(1), which makes the second line the arguments of the
echo(1) command from the first line. To avoid this, write
${i:Q}""
.
The pkgsrc bmake program does not handle the following
assignment correctly. In case _othervar_
contains a ``-'' character, one of the closing braces is included
in ${VAR}
after this code executes.
VAR:= ${VAR:N${_othervar_:C/-//}}
For a more complex code snippet and a workaround, see the
package regress/make-quoting
, testcase
bug1
.
Table of Contents
The PLIST
file contains a package's
“packing list”, i.e. a list of files that belong to
the package (relative to the ${PREFIX}
directory it's been installed in) plus some additional statements
- see the pkg_create(1) man page for a full list.
This chapter addresses some issues that need attention when
dealing with the PLIST
file (or files, see
below!).
Be sure to add a RCS ID line as the first thing in any
PLIST
file you write:
@comment $NetBSD$
You can use the make print-PLIST command to output a PLIST that matches any new files since the package was extracted. See Section 16.17, “Other helpful targets” for more information on this target.
If you have used any of the *-dirs packages, as explained in
Section 12.8, “Sharing directories between packages”, you may have noticed that
make print-PLIST outputs a set of
@comment
s instead of real
@dirrm
lines. You can also do this for
specific directories and files, so that the results of that
command are very close to reality. This helps a
lot during the update of packages.
The PRINT_PLIST_AWK
variable takes a set
of AWK patterns and actions that are used to filter the output of
print-PLIST. You can append any chunk of AWK
scripting you like to it, but be careful with quoting.
For example, to get all files inside the
libdata/foo
directory removed from the
resulting PLIST:
PRINT_PLIST_AWK+= /^libdata\/foo/ { next; }
And to get all the @dirrm
lines referring
to a specific (shared) directory converted to
@comment
s:
PRINT_PLIST_AWK+= /^@dirrm share\/specific/ { print "@comment " $$0; next; }
A number of variables are substituted automatically in PLISTs when a package is installed on a system. This includes the following variables:
${MACHINE_ARCH}
, ${MACHINE_GNU_ARCH}
Some packages like emacs and perl embed information
about which architecture they were built on into the
pathnames where they install their files. To handle this
case, PLIST will be preprocessed before actually used, and
the symbol
“${MACHINE_ARCH}
” will be
replaced by what uname -p gives. The
same is done if the string
${MACHINE_GNU_ARCH}
is embedded in
PLIST somewhere - use this on packages that have GNU
autoconf-created configure scripts.
There used to be a symbol
“$ARCH
” that
was replaced by the output of uname
-m, but that's no longer supported and has
been removed.
${OPSYS}
, ${LOWER_OPSYS}
, ${OS_VERSION}
Some packages want to embed the OS name and version
into some paths. To do this, use these variables in the
PLIST
:
${OPSYS}
- output of “uname -s”
${LOWER_OPSYS}
- lowercase common name (eg. “solaris”)
${OS_VERSION}
- “uname -r”
For a complete list of values which are replaced by
default, please look in bsd.pkg.mk
(and
search for PLIST_SUBST
).
If you want to change other variables not listed above, you
can add variables and their expansions to this variable in the
following way, similar to MESSAGE_SUBST
(see Section 10.5, “Optional files”):
PLIST_SUBST+= SOMEVAR="somevalue"
This replaces all occurrences of “${SOMEVAR}”
in the PLIST
with
“somevalue”.
Man pages should be installed in compressed form if
MANZ
is set (in bsd.own.mk
),
and uncompressed otherwise. To handle this in the
PLIST
file, the suffix “.gz” is
appended/removed automatically for man pages according to
MANZ
and MANCOMPRESSED
being set
or not, see above for details. This modification of the
PLIST
file is done on a copy of it, not
PLIST
itself.
To use one or more files as source for the PLIST
used
in generating the binary package, set the variable
PLIST_SRC
to the names of that file(s).
The files are later concatenated using cat(1), and the order of things is
important. The default for PLIST_SRC
is
${PKGDIR}/PLIST
.
Some packages decide to install a different set of files based on the operating system being used. These differences can be automatically handled by using the following files:
PLIST.common
PLIST.${OPSYS}
PLIST.${MACHINE_ARCH}
PLIST.${OPSYS}-${MACHINE_ARCH}
PLIST.common_end
A “shared directory” is a directory where multiple (and unrelated) packages install files. These directories are problematic because you have to add special tricks in the PLIST to conditionally remove them, or have some centralized package handle them.
Within pkgsrc, you'll find both approaches. If a directory is shared by a few unrelated packages, it's often not worth to add an extra package to remove it. Therefore, one simply does:
@unexec ${RMDIR} %D/path/to/shared/directory 2>/dev/null || ${TRUE}
in the PLISTs of all affected packages, instead of the regular "@dirrm" line.
However, if the directory is shared across many packages, two different solutions are available:
If the packages have a common dependency, the directory
can be removed in that. For example, see
textproc/scrollkeeper
, which
removes the shared directory
share/omf
.
If the packages using the directory are not related at all (they have no common dependencies), a *-dirs package is used.
From now on, we'll discuss the second solution. To get an idea of the *-dirs packages available, issue:
%
cd .../pkgsrc%
ls -d */*-dirs
Their use from other packages is very simple. The
USE_DIRS
variable takes a list of package names
(without the “-dirs” part) together with the required
version number (always pick the latest one when writing new
packages).
For example, if a package installs files under
share/applications
, it should have the
following line in it:
USE_DIRS+= xdg-1.1
After regenerating the PLIST using make print-PLIST, you should get the right (commented out) lines.
Note that even if your package is using
$X11BASE
, it must not depend on the
*-x11-dirs packages. Just specify the name without that part and
pkgsrc (in particular, mk/dirs.mk
) will take
care of it.
Table of Contents
Buildlink is a framework in pkgsrc that controls what headers and libraries are seen by a package's configure and build processes. This is implemented in a two step process:
Symlink headers and libraries for dependencies into
BUILDLINK_DIR
, which by default is a subdirectory
of WRKDIR
.
Create wrapper scripts that are used in place of the normal compiler
tools that translate -I${LOCALBASE}/include
and
-L${LOCALBASE}/lib
into references to
BUILDLINK_DIR
. The wrapper scripts also make
native compiler on some operating systems look like GCC, so that
packages that expect GCC won't require modifications to build with
those native compilers.
This normalizes the environment in which a package is built so that the
package may be built consistently despite what other software may be
installed. Please note that the normal system header and library paths,
e.g. /usr/include
,
/usr/lib
, etc., are always searched -- buildlink3 is
designed to insulate the package build from non-system-supplied
software.
The process of converting packages to use the buildlink3 framework (“bl3ifying”) is fairly straightforward. The things to keep in mind are:
Ensure that the build always calls the wrapper scripts
instead of the actual toolchain. Some packages are tricky,
and the only way to know for sure is the check
${WRKDIR}/.work.log
to see if the
wrappers are being invoked.
Don't override PREFIX
from within
the package Makefile, e.g. Java VMs, standalone shells,
etc., because the code to symlink files into
${BUILDLINK_DIR}
looks for files
relative to “pkg_info -qp pkgname
”.
Remember that only the
buildlink3.mk
files that you list in a
package's Makefile are added as dependencies for that package.
If a dependency on a particular package is required for its libraries and headers, then we replace:
DEPENDS+= foo>=1.1.0:../../category/foo
with
.include "../../category/foo/buildlink3.mk"
The buildlink3.mk files usually define the required dependencies. If you need a newer version of the dependency when using buildlink3.mk files, then you can define it in your Makefile; for example:
BUILDLINK_API_DEPENDS.foo+= foo>=1.1.0 .include "../../category/foo/buildlink3.mk"
There are several buildlink3.mk
files in pkgsrc/mk
that handle special package issues:
bdb.buildlink3.mk
chooses either
the native or a pkgsrc Berkeley DB implementation based on
the values of BDB_ACCEPTED
and
BDB_DEFAULT
.
curses.buildlink3.mk
: If the system
comes with neither Curses nor NCurses, this will take care
to install the devel/ncurses
package.
krb5.buildlink3.mk
uses the value
of KRB5_ACCEPTED
to choose between
adding a dependency on Heimdal or MIT-krb5 for packages that
require a Kerberos 5 implementation.
motif.buildlink3.mk
checks for a
system-provided Motif installation or adds a dependency on
x11/lesstif
or x11/openmotif
. The user can set
MOTIF_TYPE
to “dt”,
“lesstif”, or “openmotif” to choose
which Motif version will be used.
oss.buildlink3.mk
defines several
variables that may be used by packages that use the
Open Sound System (OSS) API.
pgsql.buildlink3.mk
will accept
either Postgres 8.0, 8.1, or 8.2, whichever is found installed. See
the file for more information.
pthread.buildlink3.mk
uses the value of
PTHREAD_OPTS
and checks for native pthreads or adds
a dependency on devel/pth
as needed.
xaw.buildlink3.mk
uses the value of
XAW_TYPE
to choose a particular Athena widgets
library.
The comments in those buildlink3.mk
files provide a more complete
description of how to use them properly.
A package's buildlink3.mk
file is
included by Makefiles to indicate the need to compile and link
against header files and libraries provided by the package. A
buildlink3.mk
file should always provide
enough information to add the correct type of dependency
relationship and include any other
buildlink3.mk
files that it needs to find
headers and libraries that it needs in turn.
To generate an initial buildlink3.mk
file for further editing, Rene Hexel's pkgtools/createbuildlink
package is highly recommended. For most packages, the following
command will generate a good starting point for
buildlink3.mk
files:
%
cd pkgsrc/
category
/pkgdir
%
createbuildlink >buildlink3.mk
The following real-life example
buildlink3.mk
is taken
from pkgsrc/graphics/tiff
:
# $NetBSD: buildlink3.mk,v 1.7 2004/03/18 09:12:12 jlam Exp $ BUILDLINK_DEPTH:= ${BUILDLINK_DEPTH}+ TIFF_BUILDLINK3_MK:= ${TIFF_BUILDLINK3_MK}+ .if !empty(BUILDLINK_DEPTH:M+) BUILDLINK_DEPENDS+= tiff .endif BUILDLINK_PACKAGES:= ${BUILDLINK_PACKAGES:Ntiff} BUILDLINK_PACKAGES+= tiff .if !empty(TIFF_BUILDLINK3_MK:M+) BUILDLINK_API_DEPENDS.tiff+= tiff>=3.6.1 BUILDLINK_PKGSRCDIR.tiff?= ../../graphics/tiff .endif # TIFF_BUILDLINK3_MK .include "../../devel/zlib/buildlink3.mk" .include "../../graphics/jpeg/buildlink3.mk" BUILDLINK_DEPTH:= ${BUILDLINK_DEPTH:S/+$//}
The header and footer manipulate
BUILDLINK_DEPTH
, which is common across all
buildlink3.mk
files and is used to track
at what depth we are including
buildlink3.mk
files.
The first section controls if the dependency on
pkg
is added.
BUILDLINK_DEPENDS
is the global list of
packages for which dependencies are added by
buildlink3.
The second section advises pkgsrc that the
buildlink3.mk
file for
pkg
has been included at some point.
BUILDLINK_PACKAGES
is the global list of
packages for which buildlink3.mk
files
have been included. It must always be
appended to within a buildlink3.mk
file.
The third section is protected from multiple inclusion
and controls how the dependency on pkg
is
added. Several important variables are set in the section:
BUILDLINK_API_DEPENDS.
is the actual dependency recorded in the installed
package; this should always be set using
+= to ensure that
we're appending to any pre-existing list of values. This
variable should be set to the first version of the
package that had an API change.
pkg
BUILDLINK_PKGSRCDIR.
is the location of the pkg
pkg
pkgsrc directory.
BUILDLINK_DEPMETHOD.
(not shown above) controls whether we use
pkg
BUILD_DEPENDS
or
DEPENDS
to add the dependency on
pkg
. The build dependency is
selected by setting
BUILDLINK_DEPMETHOD.
to “build”. By default, the full dependency is
used.pkg
BUILDLINK_INCDIRS.
and
pkg
BUILDLINK_LIBDIRS.
(not shown above) are lists of subdirectories of
pkg
${BUILDLINK_PREFIX.
to add to the header and library search paths. These
default to “include” and “lib”
respectively.pkg
}
BUILDLINK_CPPFLAGS.
(not shown above) is the list of preprocessor flags to add
to pkg
CPPFLAGS
, which are passed on to the
configure and build phases. The “-I” option
should be avoided and instead be handled using
BUILDLINK_INCDIRS.
as
above.pkg
The following variables are all optionally defined within
this second section (protected against multiple inclusion) and
control which package files are symlinked into
${BUILDLINK_DIR}
and how their names are
transformed during the symlinking:
BUILDLINK_FILES.
(not shown above) is a shell glob pattern relative to
pkg
${BUILDLINK_PREFIX.
to be symlinked into
pkg
}${BUILDLINK_DIR}
,
e.g. include/*.h
.
BUILDLINK_FILES_CMD.
(not shown above) is a shell pipeline that
outputs to stdout a list of files relative to
pkg
${BUILDLINK_PREFIX.
.
The resulting files are to be symlinked
into pkg
}${BUILDLINK_DIR}
. By default,
this takes the +CONTENTS
of a
pkg
and filters it through
${BUILDLINK_CONTENTS_FILTER.
.pkg
}
BUILDLINK_CONTENTS_FILTER.
(not shown above) is a filter command that filters
pkg
+CONTENTS
input into a list of files
relative to
${BUILDLINK_PREFIX.
on stdout. By default for overwrite packages,
pkg
}BUILDLINK_CONTENTS_FILTER.
outputs the contents of the pkg
include
and lib
directories in the package
+CONTENTS
, and for pkgviews packages,
it outputs any libtool archives in
lib
directories.
BUILDLINK_FNAME_TRANSFORM.
(not shown above) is a list of sed arguments used to
transform the name of the source filename into a
destination filename, e.g. -e
"s|/curses.h|/ncurses.h|g".pkg
The last section includes any
buildlink3.mk
needed for
pkg
's library dependencies.
Including these buildlink3.mk
files
means that the headers and libraries for these
dependencies are also symlinked into
${BUILDLINK_DIR}
whenever the pkg
buildlink3.mk
file is included.
The situation that requires increasing the dependency listed in
BUILDLINK_API_DEPENDS.
after a package update is when the API or interface to the
header files change.pkg
In this case,
BUILDLINK_API_DEPENDS.
should be adjusted to require at least the new package
version. In some cases, the packages that depend on this new
version may need their pkg
PKGREVISION
s
increased and, if they have buildlink3.mk
files, their
BUILDLINK_API_DEPENDS.
adjusted, too. This is needed so pkgsrc will require the
correct package dependency and not settle for an older one
when building the source.pkg
BUILDLINK_ABI_DEPENDS.
should be increased when the binary interface or sonames
(major number of the library version) of any installed
shared libraries change. This is needed so that binary
packages made using it will require the correct package
dependency and not settle for an older one which will not
contain the necessary shared libraries.pkg
See Section 18.1.6, “Handling dependencies” for
more information about dependencies on other packages,
including the BUILDLINK_ABI_DEPENDS
and
ABI_DEPENDS
definitions.
Please take careful consideration before adjusting
BUILDLINK_API_DEPENDS.
or
pkg
BUILDLINK_ABI_DEPENDS.
as we don't want to cause unneeded package deletions and
rebuilds. In many cases, new versions of packages work just
fine with older dependencies.pkg
Also it is not needed to set
BUILDLINK_ABI_DEPENDS.
when it is identical to
pkg
BUILDLINK_API_DEPENDS.
. pkg
Some packages in pkgsrc install headers and libraries that
coincide with headers and libraries present in the base system.
Aside from a buildlink3.mk
file, these
packages should also include a builtin.mk
file that includes the necessary checks to decide whether using
the built-in software or the pkgsrc software is
appropriate.
The only requirements of a builtin.mk file for
pkg
are:
It should set
USE_BUILTIN.
to either “yes” or “no”
after it is included.pkg
It should not override any
USE_BUILTIN.
which is already set before the
pkg
builtin.mk
file is included.
It should be written to allow multiple inclusion. This
is very important and takes careful
attention to Makefile
coding.
The following is the recommended template for builtin.mk files:
.if !defined(IS_BUILTIN.foo) # # IS_BUILTIN.foo is set to "yes" or "no" depending on whether "foo" # genuinely exists in the system or not. # IS_BUILTIN.foo?= no # BUILTIN_PKG.foo should be set here if "foo" is built-in and its package # version can be determined. # . if !empty(IS_BUILTIN.foo:M[yY][eE][sS]) BUILTIN_PKG.foo?= foo-1.0 . endif .endif # IS_BUILTIN.foo .if !defined(USE_BUILTIN.foo) USE_BUILTIN.foo?= ${IS_BUILTIN.foo} . if defined(BUILTIN_PKG.foo) . for _depend_ in ${BUILDLINK_API_DEPENDS.foo} . if !empty(USE_BUILTIN.foo:M[yY][eE][sS]) USE_BUILTIN.foo!= \ if ${PKG_ADMIN} pmatch '${_depend_}' ${BUILTIN_PKG.foo}; then \ ${ECHO} "yes"; \ else \ ${ECHO} "no"; \ fi . endif . endfor . endif .endif # USE_BUILTIN.foo CHECK_BUILTIN.foo?= no .if !empty(CHECK_BUILTIN.foo:M[nN][oO]) # # Here we place code that depends on whether USE_BUILTIN.foo is set to # "yes" or "no". # .endif # CHECK_BUILTIN.foo
The first section sets
IS_BUILTIN.
depending on if pkg
pkg
really exists
in the base system. This should not be a base system software
with similar functionality to pkg
;
it should only be “yes” if the actual package is
included as part of the base system. This variable is only
used internally within the builtin.mk
file.
The second section sets
BUILTIN_PKG.
to the version of pkg
pkg
in the base
system if it exists (if
IS_BUILTIN.
is “yes”). This variable is only used internally
within the pkg
builtin.mk
file.
The third section sets
USE_BUILTIN.
and is required in all
pkg
builtin.mk
files. The code in this
section must make the determination whether the built-in
software is adequate to satisfy the dependencies listed in
BUILDLINK_API_DEPENDS.
.
This is typically done by comparing
pkg
BUILTIN_PKG.
against each of the dependencies in
pkg
BUILDLINK_API_DEPENDS.
.
pkg
USE_BUILTIN.
must be set to the correct value by the
end of the pkg
builtin.mk
file. Note that
USE_BUILTIN.
may be “yes” even if
pkg
IS_BUILTIN.
is “no” because we may make the determination
that the built-in version of the software is similar enough to
be used as a replacement.pkg
The last section is guarded by
CHECK_BUILTIN.
,
and includes code that uses the value of
pkg
USE_BUILTIN.
set in the previous section. This typically includes, e.g.,
adding additional dependency restrictions and listing additional
files to symlink into pkg
${BUILDLINK_DIR}
(via
BUILDLINK_FILES.
).pkg
When building packages, it's possible to choose whether to set
a global preference for using either the built-in (native)
version or the pkgsrc version of software to satisfy a
dependency. This is controlled by setting
PREFER_PKGSRC
and
PREFER_NATIVE
. These variables take values
of either “yes”, “no”, or a list of
packages. PREFER_PKGSRC
tells pkgsrc to
use the pkgsrc versions of software, while
PREFER_NATIVE
tells pkgsrc to use the
built-in versions. Preferences are determined by the most
specific instance of the package in either
PREFER_PKGSRC
or
PREFER_NATIVE
. If a package is specified
in neither or in both variables, then
PREFER_PKGSRC
has precedence over
PREFER_NATIVE
. For example, to require
using pkgsrc versions of software for all but the most basic
bits on a NetBSD system, you can set:
PREFER_PKGSRC= yes PREFER_NATIVE= getopt skey tcp_wrappers
A package must have a
builtin.mk
file to be listed in PREFER_NATIVE
,
otherwise it is simply ignored in that list.
Table of Contents
This chapter describes the framework known as
pkginstall
, whose key features are:
Generic installation and manipulation of directories and files
outside the pkgsrc-handled tree, LOCALBASE
.
Automatic handling of configuration files during installation, provided that packages are correctly designed.
Generation and installation of system startup scripts.
Registration of system users and groups.
Registration of system shells.
Automatic updating of fonts databases.
The following sections inspect each of the above points in detail.
You may be thinking that many of the things described here could be
easily done with simple code in the package's post-installation target
(post-install
). This is incorrect,
as the code in them is only executed when building from source. Machines
using binary packages could not benefit from it at all (as the code itself
could be unavailable). Therefore, the only way to achieve any of the items
described above is by means of the installation scripts, which are
automatically generated by pkginstall.
As you already know, the PLIST
file holds a list
of files and directories that belong to a package. The names used in it
are relative to the installation prefix (${PREFIX}
),
which means that it cannot register files outside this directory (absolute
path names are not allowed). Despite this restriction, some packages need
to install files outside this location; e.g., under
${VARBASE}
or
${PKG_SYSCONFDIR}
. The only way to achieve this
is to create such files during installation time by using
installation scripts.
The generic installation scripts are shell scripts that can
contain arbitrary code. The list of scripts to execute is taken from
the INSTALL_FILE
variable, which defaults to
INSTALL
. A similar variable exists for package
removal (DEINSTALL_FILE
, whose default is
DEINSTALL
). These scripts can run arbitrary
commands, so they have the potential to create and manage files
anywhere in the file system.
Using these general installation files is not recommended, but may be needed in some special cases. One reason for avoiding them is that the user has to trust the packager that there is no unwanted or simply erroneous code included in the installation script. Also, previously there were many similar scripts for the same functionality, and fixing a common error involved finding and changing all of them.
The pkginstall framework offers another, standardized way. It
provides generic scripts to abstract the manipulation of such files
and directories based on variables set in the package's
Makefile
. The rest of this section describes
these variables.
The following variables can be set to request the creation of directories anywhere in the file system:
MAKE_DIRS
and OWN_DIRS
contain a list of directories that should be created and should attempt
to be destroyed by the installation scripts. The difference between
the two is that the latter prompts the administrator to remove any
directories that may be left after deinstallation (because they were
not empty), while the former does not.
MAKE_DIRS_PERMS
and
OWN_DIRS_PERMS
contain a list of tuples describing
which directories should be created and should attempt to be destroyed
by the installation scripts. Each tuple holds the following values,
separated by spaces: the directory name, its owner, its group and its
numerical mode. For example:
MAKE_DIRS_PERMS+= ${VARBASE}/foo/private ${ROOT_USER} ${ROOT_GROUP} 0700
The difference between the two is exactly the same as their
non-PERMS
counterparts.
Creating non-empty files outside the installation prefix is tricky
because the PLIST
forces all files to be inside it.
To overcome this problem, the only solution is to extract the file in the
known place (i.e., inside the installation prefix) and copy it to the
appropriate location during installation (done by the installation scripts
generated by pkginstall). We will call the former the master
file in the following paragraphs, which describe the variables
that can be used to automatically and consistently handle files outside the
installation prefix:
CONF_FILES
and
SUPPORT_FILES
are pairs of master and target files.
During installation time, the master file is copied to the target one
if and only if the latter does not exist. Upon deinstallation, the
target file is removed provided that it was not modified by the
installation.
The difference between the two is that the latter prompts the administrator to remove any files that may be left after deinstallation (because they were not empty), while the former does not.
CONF_FILES_PERMS
and
SUPPORT_FILES_PERMS
contain tuples describing master
files as well as their target locations. For each of them, it also
specifies their owner, their group and their numeric permissions, in
this order. For example:
SUPPORT_FILES_PERMS+= ${PREFIX}/share/somefile ${VARBASE}/somefile ${ROOT_USER} ${ROOT_GROUP} 0700
The difference between the two is exactly the same as their
non-PERMS
counterparts.
Configuration files are special in the sense that they are installed
in their own specific directory, PKG_SYSCONFDIR
, and
need special treatment during installation (most of which is automated by
pkginstall). The main concept you must bear in mind is that files marked
as configuration files are automatically copied to the right place (somewhere
inside PKG_SYSCONFDIR
) during installation if
and only if they didn't exist before. Similarly, they will not
be removed if they have local modifications. This ensures that
administrators never lose any custom changes they may have made.
As said before, the PKG_SYSCONFDIR
variable
specifies where configuration files shall be installed. Its contents are
set based upon the following variables:
PKG_SYSCONFBASE
: The configuration's root
directory. Defaults to ${PREFIX}/etc
although it may
be overridden by the user to point to his preferred location (e.g.,
/etc
, /etc/pkg
, etc.).
Packages must not use it directly.
PKG_SYSCONFSUBDIR
: A subdirectory of
PKG_SYSCONFBASE
under which the configuration files
for the package being built shall be installed. The definition of this
variable only makes sense in the package's
Makefile
(i.e., it is not user-customizable).
As an example, consider the Apache package,
www/apache2
, which places its
configuration files under the
httpd/
subdirectory of
PKG_SYSCONFBASE
. This should be set in the package
Makefile.
PKG_SYSCONFVAR
: Specifies the name of the
variable that holds this package's configuration directory (if
different from PKG_SYSCONFBASE
). It defaults to
PKGBASE
's value, and is always prefixed with
PKG_SYSCONFDIR
.
PKG_SYSCONFDIR.${PKG_SYSCONFVAR}
: Holds the
directory where the configuration files for the package identified by
PKG_SYSCONFVAR
's shall be placed.
Based on the above variables, pkginstall determines the value of
PKG_SYSCONFDIR
, which is the only
variable that can be used within a package to refer to its configuration
directory. The algorithm used to set its value is basically the
following:
If PKG_SYSCONFDIR.${PKG_SYSCONFVAR}
is set,
its value is used.
If the previous variable is not defined but
PKG_SYSCONFSUBDIR
is set in the package's
Makefile
, the resulting value is
${PKG_SYSCONFBASE}/${PKG_SYSCONFSUBDIR}
.
Otherwise, it is set to
${PKG_SYSCONFBASE}
.
It is worth mentioning that ${PKG_SYSCONFDIR}
is
automatically added to OWN_DIRS
. See Section 14.1.1, “Directory manipulation” what this means.
Given that pkgsrc (and users!) expect configuration files to be in a known place, you need to teach each package where it shall install its files. In some cases you will have to patch the package Makefiles to achieve it. If you are lucky, though, it may be as easy as passing an extra flag to the configuration script; this is the case of GNU Autoconf- generated files:
CONFIGURE_ARGS+= --sysconfdir=${PKG_SYSCONFDIR}
Note that this specifies where the package has to look for its configuration files, not where they will be originally installed (although the difference is never explicit, unfortunately).
As said before, pkginstall automatically handles configuration files.
This means that the packages themselves must not
touch the contents of ${PKG_SYSCONFDIR}
directly. Bad news is that many software installation scripts
will, out of the box, mess with the contents of that directory. So what is
the correct procedure to fix this issue?
You must teach the package (usually by manually patching it) to
install any configuration files under the examples hierarchy,
share/examples/${PKGBASE}/
. This way, the
PLIST
registers them and the administrator always
has the original copies available.
Once the required configuration files are in place (i.e., under the
examples hierarchy), the pkginstall framework can use them as master copies
during the package installation to update what is in
${PKG_SYSCONFDIR}
. To achieve this, the variables
CONF_FILES
and CONF_FILES_PERMS
are
used. Check out Section 14.1.2, “File manipulation” for information
about their syntax and their purpose. Here is an example, taken from the
mail/mutt
package:
EGDIR= ${PREFIX}/share/doc/mutt/samples CONF_FILES= ${EGDIR}/Muttrc ${PKG_SYSCONFDIR}/Muttrc
Note that the EGDIR
variable is specific to that
package and has no meaning outside it.
System startup scripts are special files because they must be installed in a place known by the underlying OS, usually outside the installation prefix. Therefore, the same rules described in Section 14.1, “Files and directories outside the installation prefix” apply, and the same solutions can be used. However, pkginstall provides a special mechanism to handle these files.
In order to provide system startup scripts, the package has to:
Store the script inside ${FILESDIR}
, with
the .sh
suffix appended. Considering the
print/cups
package as an example, it has a
cupsd.sh
in its files directory.
Tell pkginstall to handle it, appending the name of the script,
without its extension, to the RCD_SCRIPTS
variable.
Continuing the previous example:
RCD_SCRIPTS+= cupsd
Once this is done, pkginstall will do the following steps for each script in an automated fashion:
Process the file found in the files directory applying all the
substitutions described in the FILES_SUBST
variable.
Copy the script from the files directory to the examples
hierarchy, ${PREFIX}/share/examples/rc.d/
. Note
that this master file must be explicitly registered in the
PLIST
.
Add code to the installation scripts to copy the startup script from the examples hierarchy into the system-wide startup scripts directory.
The automatic copying of config files can be toggled by setting the
environment variable PKG_RCD_SCRIPTS
prior to package
installation. Note that the scripts will be always copied inside the
examples hierarchy, ${PREFIX}/share/examples/rc.d/
, no
matter what the value of this variable is.
If a package needs to create special users and/or groups during installation, it can do so by using the pkginstall framework.
Users can be created by adding entries to the
PKG_USERS
variable. Each entry has the following
syntax:
user:group
Further specification of user details may be done by setting
per-user variables.
PKG_UID.
is the
numeric UID for the user.
user
PKG_GECOS.
is the
user's description or comment.
user
PKG_HOME.
is the
user's home directory, and defaults to
user
/nonexistent
if not specified.
PKG_SHELL.
is the
user's shell, and defaults to user
/sbin/nologin
if
not specified.
Similarly, groups can be created by adding entries to the
PKG_GROUPS
variable, whose syntax is:
group
The numeric GID of the group may be set by defining
PKG_GID.
.group
If a package needs to create the users and groups at an earlier
stage, then it can set USERGROUP_PHASE
to
either configure
or build
to
indicate the phase before which the users and groups are created. In
this case, the numeric UIDs and GIDs of the created users and groups
are automatically hardcoded into the final installation scripts.
Packages that install system shells should register them in the shell
database, /etc/shells
, to make things easier to the
administrator. This must be done from the installation scripts to keep
binary packages working on any system. pkginstall provides an easy way to
accomplish this task.
When a package provides a shell interpreter, it has to set the
PKG_SHELL
variable to its absolute file name. This will
add some hooks to the installation scripts to handle it. Consider the
following example, taken from shells/zsh
:
PKG_SHELL= ${PREFIX}/bin/zsh
Packages that install X11 fonts should update the database files that index the fonts within each fonts directory. This can easily be accomplished within the pkginstall framework.
When a package installs X11 fonts, it must list the directories in
which fonts are installed in the
FONTS_DIRS.
variables,
where type
type
can be one of “ttf”,
“type1” or “x11”. This will add hooks to the
installation scripts to run the appropriate commands to update the fonts
database files within each of those directories. For convenience, if the
directory path is relative, it is taken to be relative to the package's
installation prefix. Consider the following example, taken from fonts/dbz-ttf
:
FONTS_DIRS.ttf= ${PREFIX}/lib/X11/fonts/TTF
Table of Contents
Many packages have the ability to be built to support different
sets of features. bsd.options.mk
is a framework
in pkgsrc that provides generic handling of those options that
determine different ways in which the packages can be built. It's
possible for the user to specify exactly which sets of options will be
built into a package or to allow a set of global default options
apply.
Global default options are listed in
PKG_DEFAULT_OPTIONS
, which is a list of the options
that should be built into every package if that option is supported.
This variable should be set in /etc/mk.conf
.
The following example shows how
bsd.options.mk
should be used
by the hypothetical ``wibble'' package, either in the package
Makefile
, or in a file,
e.g. options.mk
, that is included by the
main package Makefile
.
PKG_OPTIONS_VAR= PKG_OPTIONS.wibble PKG_SUPPORTED_OPTIONS= wibble-foo ldap PKG_OPTIONS_OPTIONAL_GROUPS= database PKG_OPTIONS_GROUP.database= mysql pgsql PKG_SUGGESTED_OPTIONS= wibble-foo PKG_OPTIONS_LEGACY_VARS+= WIBBLE_USE_OPENLDAP:ldap PKG_OPTIONS_LEGACY_OPTS+= foo:wibble-foo .include "../../mk/bsd.prefs.mk" # this package was previously named wibble2 .if defined(PKG_OPTIONS.wibble2) PKG_LEGACY_OPTIONS+= ${PKG_OPTIONS.wibble2} PKG_OPTIONS_DEPRECATED_WARNINGS+= \ "Deprecated variable PKG_OPTIONS.wibble2 used, use ${PKG_OPTIONS_VAR} instead." .endif .include "../../mk/bsd.options.mk" # Package-specific option-handling ### ### FOO support ### .if !empty(PKG_OPTIONS:Mwibble-foo) CONFIGURE_ARGS+= --enable-foo .endif ### ### LDAP support ### .if !empty(PKG_OPTIONS:Mldap) . include "../../databases/openldap-client/buildlink3.mk" CONFIGURE_ARGS+= --enable-ldap=${BUILDLINK_PREFIX.openldap-client} .endif ### ### database support ### .if !empty(PKG_OPTIONS:Mmysql) . include "../../mk/mysql.buildlink3.mk" .endif .if !empty(PKG_OPTIONS:Mpgsql) . include "../../mk/pgsql.buildlink3.mk" .endif
The first section contains the information about which build options are supported by the package, and any default options settings if needed.
PKG_OPTIONS_VAR
is the name of the
make(1) variable that the user can set to override the default
options. It should be set to
PKG_OPTIONS.pkgbase
. Do not set it to
PKG_OPTIONS.${PKGBASE}, since PKGBASE
is not defined
at the point where the options are processed.
PKG_SUPPORTED_OPTIONS
is a list of
build options supported by the package.
PKG_OPTIONS_OPTIONAL_GROUPS
is a
list of names of groups of mutually exclusive options. The options in
each group are listed in
PKG_OPTIONS_GROUP.
.
The most specific setting of any option from the group takes
precedence over all other options in the group. Options from the
groups will be automatically added to
groupname
PKG_SUPPORTED_OPTIONS
.
PKG_OPTIONS_REQUIRED_GROUPS
is like
PKG_OPTIONS_OPTIONAL_GROUPS
, but building the
packages will fail if no option from the group is
selected.
PKG_OPTIONS_NONEMPTY_SETS
is a list
of names of sets of options. At least one option from each set must
be selected. The options in each set are listed in
PKG_OPTIONS_SET.
.
Options from the sets will be automatically added to
setname
PKG_SUPPORTED_OPTIONS
. Building the package will
fail if no option from the set is selected.
PKG_SUGGESTED_OPTIONS
is a list of
build options which are enabled by default.
PKG_OPTIONS_LEGACY_VARS
is a list
of
“USE_VARIABLE
:option
”
pairs that map legacy /etc/mk.conf
variables to
their option counterparts. Pairs should be added with
“+=” to keep the listing of global legacy variables. A
warning will be issued if the user uses a legacy
variable.
PKG_OPTIONS_LEGACY_OPTS
is a list
of
“old-option
:new-option
”
pairs that map options that have been renamed to their new
counterparts. Pairs should be added with “+=” to keep
the listing of global legacy options. A warning will be issued if
the user uses a legacy option.
PKG_LEGACY_OPTIONS
is a list of
options implied by deprecated variables used. This can be used for
cases that neither PKG_OPTIONS_LEGACY_VARS
nor
PKG_OPTIONS_LEGACY_OPTS
can handle, e. g. when
PKG_OPTIONS_VAR
is renamed.
PKG_OPTIONS_DEPRECATED_WARNINGS
is
a list of warnings about deprecated variables or options used, and
what to use instead.
A package should never modify
PKG_DEFAULT_OPTIONS
or the variable named in
PKG_OPTIONS_VAR
. These are strictly user-settable.
To suggest a default set of options, use
PKG_SUGGESTED_OPTIONS
.
PKG_OPTIONS_VAR
must be defined before
including bsd.options.mk
. If none of
PKG_SUPPORTED_OPTIONS
,
PKG_OPTIONS_OPTIONAL_GROUPS
, and
PKG_OPTIONS_REQUIRED_GROUPS
are defined (as can
happen with platform-specific options if none of them is supported on
the current platform), PKG_OPTIONS
is set to the
empty list and the package is otherwise treated as not using the
options framework.
After the inclusion of bsd.options.mk
, the
variable PKG_OPTIONS
contains the list of selected
build options, properly filtered to remove unsupported and duplicate
options.
The remaining sections contain the logic that is specific to
each option. The correct way to check for an option is to check
whether it is listed in PKG_OPTIONS
:
.if !empty(PKG_OPTIONS:Moption
)
Options that enable similar features in different packages (like optional support for a library) should use a common name in all packages that support it (like the name of the library). If another package already has an option with the same meaning, use the same name.
Options that enable features specific to one package, where it's
unlikely that another (unrelated) package has the same (or a similar)
optional feature, should use a name prefixed with
.pkgname
-
If a group of related packages share an optional feature
specific to that group, prefix it with the name of the
“main” package
(e. g. djbware-errno-hack
).
For new options, add a line to
mk/defaults/options.description
. Lines have two
fields, separated by tab. The first field is the option name, the
second its description. The description should be a whole sentence
(starting with an uppercase letter and ending with a period) that
describes what enabling the option does. E. g. “Enable ispell
support.” The file is sorted by option names.
Table of Contents
This chapter gives a detailed description on how a package is
built. Building a package is separated into different
phases (for example fetch
,
build
, install
), all of which are
described in the following sections. Each phase is splitted into
so-called stages, which take the name of the
containing phase, prefixed by one of pre-
,
do-
or post-
. (Examples are
pre-configure
, post-build
.) Most
of the actual work is done in the do-*
stages.
Never override the regular targets (like
fetch
), if you have to, override the
do-*
ones instead.
The basic steps for building a program are always the same. First the program's source (distfile) must be brought to the local system and then extracted. After any pkgsrc-specific patches to compile properly are applied, the software can be configured, then built (usually by compiling), and finally the generated binaries, etc. can be put into place on the system.
To get more details about what is happening at each step,
you can set the PKG_VERBOSE
variable, or the
PATCH_DEBUG
variable if you are just interested
in more details about the patch step.
Before outlining the process performed by the NetBSD package system in the next section, here's a brief discussion on where programs are installed, and which variables influence this.
The automatic variable PREFIX
indicates
where all files of the final program shall be installed. It is
usually set to LOCALBASE
(/usr/pkg
), or CROSSBASE
for pkgs in the cross
category. The value of
PREFIX
needs to be put
into the various places in the program's source where paths to
these files are encoded. See Section 10.3, “patches/*” and Section 18.3.1, “Shared libraries - libtool” for more details.
When choosing which of these variables to use, follow the following rules:
PREFIX
always points to the location
where the current pkg will be installed. When referring to a
pkg's own installation path, use
“${PREFIX}”.
LOCALBASE
is where all non-X11 pkgs
are installed. If you need to construct a -I or -L argument
to the compiler to find includes and libraries installed by
another non-X11 pkg, use “${LOCALBASE}”. The name
LOCALBASE
stems from FreeBSD, which
installed all packages in /usr/local
. As
pkgsrc leaves /usr/local
for the system
administrator, this variable is a misnomer.
X11BASE
is where the actual X11
distribution (from xsrc, etc.) is installed. When looking for
standard X11 includes (not those
installed by a package), use “${X11BASE}”.
X11-based packages are special in that they may be
installed in either X11BASE
or
LOCALBASE
.
Usually, X11 packages should be installed under
LOCALBASE
whenever possible. Note that you
will need to include
../../mk/x11.buildlink3.mk
in them to
request the presence of X11 and to get the right compilation
flags.
Even though, there are some packages that cannot be installed
under LOCALBASE
: those that come with app-defaults
files. These packages are special and they must be placed under
X11BASE
. To accomplish this, set either
USE_X11BASE
or USE_IMAKE
in
your package.
Some notes: If you need
to find includes or libraries installed by a pkg that has
USE_IMAKE
or USE_X11BASE
in
its pkg Makefile
, you need to look in
both ${X11BASE}
and
${LOCALBASE}
. To force installation of
all X11 packages in LOCALBASE
, the
pkgtools/xpkgwedge
package
is enabled by default.
X11PREFIX
should be used to refer to
the installed location of an X11
package. X11PREFIX
will be set to
X11BASE
if xpkgwedge is not installed, and
to LOCALBASE
if xpkgwedge is
installed.
If xpkgwedge is installed, it is possible to have some
packages installed in X11BASE
and some in
LOCALBASE
. To determine the prefix of an
installed package, the EVAL_PREFIX
definition can be used. It takes pairs in the format
“DIRNAME=<package>”, and the make(1)
variable DIRNAME
will be set to the prefix
of the installed package <package>, or
“${X11PREFIX}” if the package is not
installed.
This is best illustrated by example.
The following lines are taken from
pkgsrc/wm/scwm/Makefile
:
EVAL_PREFIX+= GTKDIR=gtk+ CONFIGURE_ARGS+= --with-guile-prefix=${LOCALBASE:Q} CONFIGURE_ARGS+= --with-gtk-prefix=${GTKDIR:Q} CONFIGURE_ARGS+= --enable-multibyte
Specific defaults can be defined for the packages
evaluated using EVAL_PREFIX
, by using a
definition of the form:
GTKDIR_DEFAULT= ${LOCALBASE}
where GTKDIR
corresponds
to the first definition in
the EVAL_PREFIX
pair.
Within ${PREFIX}
, packages should
install files according to hier(7), with the exception that
manual pages go into ${PREFIX}/man
, not
${PREFIX}/share/man
.
When building a package, a number of directories is used to store source files, temporary files, pkgsrc-internal files, and so on. These directories are explained here.
Some of the directory variables contain relative pathnames. There
are two common base directories for these relative directories:
PKGSRCDIR/PKGPATH
is used for directories that are
pkgsrc-specific. WRKSRC
is used for directories
inside the package itself.
PKGSRCDIR
This is an absolute pathname that points to the pkgsrc root directory. Generally, you don't need it.
PKGDIR
This is an absolute pathname that points to the current package.
PKGPATH
This is a pathname relative to
PKGSRCDIR
that points to the current
package.
WRKDIR
This is an absolute pathname pointing to the directory where all work takes place. The distfiles are extracted to this directory. It also contains temporary directories and log files used by the various pkgsrc frameworks, like buildlink or the wrappers.
WRKSRC
This is an absolute pathname pointing to the directory
where the distfiles are extracted. It is usually a direct subdirectory
of WRKDIR
, and often it's the only directory entry
that isn't hidden. This variable may be changed by a package
Makefile
.
You can run a particular phase by typing make
phase, where phase is the name of the
phase. This will automatically run all phases that are required for this
phase. The default phase is build
, that is, when you
run make without parameters in a package directory,
the package will be built, but not installed.
The first step in building a package is to fetch the distribution files (distfiles) from the sites that are providing them. This is the task of the fetch phase.
In simple cases, MASTER_SITES
defines all URLs from where the distfile, whose name is
derived from the DISTNAME
variable, is
fetched. The more complicated cases are described
below.
The variable DISTFILES
specifies
the list of distfiles that have to be fetched. Its value
defaults to ${DISTNAME}${EXTRACT_SUFX}
,
so that most packages don't need to define it at all.
EXTRACT_SUFX
is
.tar.gz
by default, but can be changed
freely. Note that if your package requires additional
distfiles to the default one, you cannot just append the
additional filenames using the +=
operator, but you have write for example:
DISTFILES= ${DISTNAME}${EXTRACT_SUFX} additional-files.tar.gz
Each distfile is fetched from a list of sites, usually
MASTER_SITES
. If the package has multiple
DISTFILES
or multiple
PATCHFILES
from different sites, you can
set
SITES.
to the list of URLs where the file
distfile
(including the suffix) can be found.distfile
DISTFILES= ${DISTNAME}${EXTRACT_SUFX} DISTFILES+= foo-file.tar.gz SITES.foo-file.tar.gz= \ http://www.somewhere.com/somehow/ \ http://www.somewhereelse.com/mirror/somehow/
When actually fetching the distfiles, each item from
MASTER_SITES
or
SITES.*
gets the name of each distfile
appended to it, without an intermediate slash. Therefore,
all site values have to end with a slash or other separator
character. This allows for example to set
MASTER_SITES
to a URL of a CGI script
that gets the name of the distfile as a parameter. In this
case, the definition would look like:
MASTER_SITES= http://www.example.com/download.cgi?file=
There are some predefined values for
MASTER_SITES
, which can be used in
packages. The names of the variables should speak for
themselves.
${MASTER_SITE_APACHE} ${MASTER_SITE_BACKUP} ${MASTER_SITE_CYGWIN} ${MASTER_SITE_DEBIAN} ${MASTER_SITE_FREEBSD} ${MASTER_SITE_FREEBSD_LOCAL} ${MASTER_SITE_GENTOO} ${MASTER_SITE_GNOME} ${MASTER_SITE_GNU} ${MASTER_SITE_GNUSTEP} ${MASTER_SITE_IFARCHIVE} ${MASTER_SITE_KDE} ${MASTER_SITE_MOZILLA} ${MASTER_SITE_MYSQL} ${MASTER_SITE_OPENOFFICE} ${MASTER_SITE_PERL_CPAN} ${MASTER_SITE_PGSQL} ${MASTER_SITE_R_CRAN} ${MASTER_SITE_SOURCEFORGE} ${MASTER_SITE_SOURCEFORGE_JP} ${MASTER_SITE_SUNSITE} ${MASTER_SITE_SUSE} ${MASTER_SITE_TEX_CTAN} ${MASTER_SITE_XCONTRIB} ${MASTER_SITE_XEMACS}
Some explanations for the less self-explaining ones:
MASTER_SITE_BACKUP
contains backup sites
for packages that are maintained in ftp://ftp.NetBSD.org/pub/NetBSD/packages/distfiles/${DIST_SUBDIR}. MASTER_SITE_LOCAL
contains local
package source distributions that are maintained in ftp://ftp.NetBSD.org/pub/NetBSD/packages/distfiles/LOCAL_PORTS/.
If you choose one of these predefined sites, you may want to specify a subdirectory of that site. Since these macros may expand to more than one actual site, you must use the following construct to specify a subdirectory:
MASTER_SITES= ${MASTER_SITE_GNU:=subdirectory/name/} MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=project_name/}
Note the trailing slash after the subdirectory name.
The fetch phase makes sure that
all the distfiles exist in a local directory
(DISTDIR
, which can be set by the pkgsrc
user). If the files do not exist, they are fetched using
commands of the form
${FETCH_CMD} ${FETCH_BEFORE_ARGS} ${site}${file} ${FETCH_AFTER_ARGS}
where ${site}
varies through
several possibilities in turn: first,
MASTER_SITE_OVERRIDE
is tried, then the
sites specified in either SITES.file
if
defined, else MASTER_SITES
or
PATCH_SITES
, as applies, then finally the
value of MASTER_SITE_BACKUP
. The order of
all except the first and the last can be optionally sorted
by the user, via setting either
MASTER_SORT_RANDOM
, and
MASTER_SORT_AWK
or
MASTER_SORT_REGEX
.
The distfiles mirror run by the NetBSD Foundation uses the
mirror-distfiles target to mirror the
distfiles, if they are freely distributable. Packages setting
NO_SRC_ON_FTP
(usually to
“${RESTRICTED}”) will not have their distfiles
mirrored.
After the distfile(s) are fetched, their checksum is generated and compared with the checksums stored in the distinfo file. If the checksums don't match, the build is aborted. This is to ensure the same distfile is used for building, and that the distfile wasn't changed, e.g. by some malign force, deliberately changed distfiles on the master distribution site or network lossage.
When the distfiles are present on the local system, they need to be extracted, as they usually come in the form of some compressed archive format.
By default, all DISTFILES
are
extracted. If you only need some of them, you can set the
EXTRACT_ONLY
variable to the list of those
files.
Extracting the files is usually done by a little
program, mk/extract/extract
, which
already knows how to extract various archive formats, so most
likely you will not need to change anything here. But if you
need, the following variables may help you:
EXTRACT_OPTS_{BIN,LHA,PAX,RAR,TAR,ZIP,ZOO}
Use these variables to override the default
options for an extract command, which are defined in
mk/extract/extract
.
EXTRACT_USING
This variable can be set to
gtar
, nbtar
(which is the
default value), pax
, or an
absolute pathname pointing to the command with which tar
archives should be
extracted.
If the extract
program doesn't
serve your needs, you can also override the
EXTRACT_CMD
variable, which holds the
command used for extracting the files. This command is
executed in the ${WRKSRC}
directory. During execution of this command, the shell
variable extract_file
holds the absolute
pathname of the file that is going to be extracted.
And if that still does not suffice, you can override the
do-extract
target in the package
Makefile.
After extraction, all the patches named by the
PATCHFILES
, those present in the patches
subdirectory of the package as well as in
$LOCALPATCHES/$PKGPATH (e.g.
/usr/local/patches/graphics/png
) are
applied. Patchfiles ending in .Z
or
.gz
are uncompressed before they are
applied, files ending in .orig
or
.rej
are ignored. Any special options to
patch(1) can be handed in
PATCH_DIST_ARGS
. See Section 10.3, “patches/*” for more details.
By default patch(1) is given special args to make it fail if the patches apply with some lines of fuzz. Please fix (regen) the patches so that they apply cleanly. The rationale behind this is that patches that don't apply cleanly may end up being applied in the wrong place, and cause severe harm there.
This is covered in Chapter 17, Tools needed for building or running.
This phase creates wrapper programs for the compilers and linkers. The following variables can be used to tweak the wrappers.
ECHO_WRAPPER_MSG
The command used to print progress
messages. Does nothing by default. Set to
${ECHO}
to see the progress
messages.
WRAPPER_DEBUG
This variable can be set to
yes
(default) or no
,
depending on whether you want additional information in the
wrapper log file.
WRAPPER_UPDATE_CACHE
This variable can be set to
yes
or no
, depending
on whether the wrapper should use its cache, which will
improve the speed. The default value is
yes
, but is forced to
no
if the platform does not support
it.
WRAPPER_REORDER_CMDS
A list of reordering commands. A reordering
command has the form
reorder:l:
.
It ensures that that
lib1
:lib2
-l
occurs
before lib1
-l
.
lib2
WRAPPER_TRANSFORM_CMDS
A list of transformation commands. [TODO: investigate further]
Most pieces of software need information on the header files, system calls, and library routines which are available on the platform they run on. The process of determining this information is known as configuration, and is usually automated. In most cases, a script is supplied with the distfiles, and its invocation results in generation of header files, Makefiles, etc.
If the package contains a configure script, this can be
invoked by setting HAS_CONFIGURE
to
“yes”. If the configure script is a GNU autoconf
script, you should set GNU_CONFIGURE
to
“yes” instead. What happens in the
configure phase is roughly:
.for d in ${CONFIGURE_DIRS} cd ${WRKSRC} && cd ${d} && env ${CONFIGURE_ENV} \ ${CONFIGURE_SCRIPT} ${CONFIGURE_ARGS} .endfor
CONFIGURE_DIRS
(default:
“.”) is a list of pathnames relative to
WRKSRC
. In each of these directories, the
configure script is run with the environment
CONFIGURE_ENV
and arguments
CONFIGURE_ARGS
. The variables
CONFIGURE_ENV
,
CONFIGURE_SCRIPT
(default:
“./configure”) and
CONFIGURE_ARGS
may all be changed by the
package.
If the program uses an Imakefile
for configuration, the appropriate steps can be invoked by
setting USE_IMAKE
to
“yes”. (If you only want the package installed in
${X11PREFIX}
but xmkmf not being run, set
USE_X11BASE
instead.) You can add variables to
xmkmf's environment by adding them to the
SCRIPTS_ENV
variable.
If there is no configure step at all, set
NO_CONFIGURE
to “yes”.
For building a package, a rough equivalent of the following code is executed.
.for d in ${BUILD_DIRS} cd ${WRKSRC} && cd ${d} && env ${MAKE_ENV} \ ${MAKE_PROGRAM} ${BUILD_MAKE_FLAGS} \ -f ${MAKE_FILE} ${BUILD_TARGET} .endfor
BUILD_DIRS
(default:
“.”) is a list of pathnames relative to
WRKSRC
. In each of these directories,
MAKE_PROGRAM
is run with the environment
MAKE_ENV
and arguments
BUILD_MAKE_FLAGS
. The variables
MAKE_ENV
,
BUILD_MAKE_FLAGS
,
MAKE_FILE
and
BUILD_TARGET
may all be changed by the
package.
The default value of MAKE_PROGRAM
is
“gmake” if USE_TOOLS
contains
“gmake”, “make” otherwise. The
default value of MAKE_FILE
is
“Makefile”, and BUILD_TARGET
defaults to “all”.
If there is no build step at all, set
NO_BUILD
to “yes”.
Once the build stage has completed, the final step is to install the software in public directories, so users can access the programs and files.
In the install phase, a rough equivalent of the following code is executed. Additionally, before and after this code, much magic is performed to do consistency checks, registering the package, and so on.
.for d in ${INSTALL_DIRS} cd ${WRKSRC} && cd ${d} && env ${MAKE_ENV} \ ${MAKE_PROGRAM} ${INSTALL_MAKE_FLAGS} \ -f ${MAKE_FILE} ${BUILD_TARGET} .endfor
The variable's meanings are analogous to the ones in the
build phase.
INSTALL_DIRS
defaults to
BUILD_DIRS
. INSTALL_TARGET
is “install” by default, plus
“install.man” if USE_IMAKE
is
defined and NO_INSTALL_MANPAGES
is not
defined.
In the install phase, the following
variables are useful. They are all variations of the
install(1) command that have the owner, group and
permissions preset. INSTALL
is the plain
install command. The specialized variants, together with their
intended use, are:
INSTALL_PROGRAM_DIR
directories that contain binaries
INSTALL_SCRIPT_DIR
directories that contain scripts
INSTALL_LIB_DIR
directories that contain shared and static libraries
INSTALL_DATA_DIR
directories that contain data files
INSTALL_MAN_DIR
directories that contain man pages
INSTALL_PROGRAM
binaries that can be stripped from debugging symbols
INSTALL_SCRIPT
binaries that cannot be stripped
INSTALL_GAME
game binaries
INSTALL_LIB
shared and static libraries
INSTALL_DATA
data files
INSTALL_GAME_DATA
data files for games
INSTALL_MAN
man pages
Some other variables are:
INSTALLATION_DIRS
A list of directories relative to
PREFIX
that are created by pkgsrc at the
beginning of the install phase. If
this variable is set,
NO_MTREE
=“yes” is assumed,
which means that the package claims to create all needed
directories itself before installing files to it. Therefore
this variable should only be set in
Makefile
s that are under control of the
package's author. The directories are created with the
correct ownership, depending on their
name.
In the rare cases that a package shouldn't install anything,
set NO_INSTALL
to “yes”. This is
mostly relevant for packages in the regress
category.
Once the install stage has completed, a binary package of the installed files can be built. These binary packages can be used for quick installation without previous compilation, e.g. by the make bin-install or by using pkg_add.
By default, the binary packages are created in
${PACKAGES}/All
and symlinks are created in
${PACKAGES}/
,
one for each category in the category
CATEGORIES
variable. PACKAGES
defaults to
pkgsrc/packages
.
If there should be no binary package, set
NO_PACKAGE
to “yes”.
This should only be used in rare cases, like when a package
definitely is only usable on the machine where it is built and even
then, a binary package can be useful.
Once you're finished with a package, you can clean the work directory by running make clean. If you want to clean the work directories of all dependencies too, use make clean-depends.
For any of the main targets described in the previous section, two auxiliary targets exist with “pre-” and “post-” used as a prefix for the main target's name. These targets are invoked before and after the main target is called, allowing extra configuration or installation steps be performed from a package's Makefile, for example, which a program's configure script or install target omitted.
Should one of the main targets do the wrong thing, and should there be no variable to fix this, you can redefine it with the do-* target. (Note that redefining the target itself instead of the do-* target is a bad idea, as the pre-* and post-* targets won't be called anymore, etc.) You will not usually need to do this.
If you did a make install and you noticed some file was not installed properly, you can repeat the installation with this target, which will ignore the “already installed” flag.
This is the default value of
DEPENDS_TARGET
except in the case of
make update and make
package, where the defaults are
“package” and “update”,
respectively.
This target does a pkg_delete(1) in the current directory, effectively de-installing the package. The following variables can be used to tune the behaviour:
PKG_VERBOSE
Add a "-v" to the pkg_delete(1) command.
DEINSTALLDEPENDS
Remove all packages that require (depend on)
the given package. This can be used to remove any
packages that may have been pulled in by a given
package, e.g. if make deinstall
DEINSTALLDEPENDS=1 is done in
pkgsrc/x11/kde
, this is
likely to remove whole KDE. Works by adding
“-R” to the pkg_delete(1)
command line.
Install a binary package from local disk and via FTP
from a list of sites (see the
BINPKG_SITES
variable), and do a
make package if no binary package is
available anywhere. The arguments given to
pkg_add can be set via
BIN_INSTALL_FLAGS
e.g., to do verbose
operation, etc.
This target causes the current package to be
updated to the latest version. The package and all
depending packages first get de-installed, then current
versions of the corresponding packages get compiled and
installed. This is similar to manually noting which
packages are currently installed, then performing a
series of make deinstall and
make install (or whatever
UPDATE_TARGET
is set to) for these
packages.
You can use the “update” target to
resume package updating in case a previous make
update was interrupted for some reason.
However, in this case, make sure you don't call
make clean or otherwise remove the
list of dependent packages in WRKDIR
.
Otherwise, you lose the ability to automatically update
the current package along with the dependent packages
you have installed.
Resuming an interrupted make update will only work as long as the package tree remains unchanged. If the source code for one of the packages to be updated has been changed, resuming make update will most certainly fail!
The following variables can be used either on the
command line or in /etc/mk.conf
to
alter the behaviour of make
update:
UPDATE_TARGET
Install target to recursively use for the
updated package and the dependent packages.
Defaults to DEPENDS_TARGET
if
set, “install” otherwise for
make update. Other good
targets are “package” or
“bin-install”. Do not set this to
“update” or you will get stuck in an
endless loop!
NOCLEAN
Don't clean up after updating. Useful if you want to leave the work sources of the updated packages around for inspection or other purposes. Be sure you eventually clean up the source tree (see the “clean-update” target below) or you may run into troubles with old source code still lying around on your next make or make update.
REINSTALL
Deinstall each package before installing
(making DEPENDS_TARGET
). This
may be necessary if the
“clean-update” target (see below) was
called after interrupting a running make
update.
DEPENDS_TARGET
Allows you to disable recursion and hardcode
the target for packages. The default is
“update” for the update target,
facilitating a recursive update of prerequisite
packages. Only set
DEPENDS_TARGET
if you want to
disable recursive updates. Use
UPDATE_TARGET
instead to just
set a specific target for each package to be
installed during make update
(see above).
Clean the source tree for all packages that would
get updated if make update was called
from the current directory. This target should not be
used if the current package (or any of its depending
packages) have already been de-installed (e.g., after
calling make update) or you may lose
some packages you intended to update. As a rule of
thumb: only use this target before
the first time you run make update
and only if you have a dirty package tree (e.g., if you
used NOCLEAN
).
If you are unsure about whether your tree is clean, you can either perform a make clean at the top of the tree, or use the following sequence of commands from the directory of the package you want to update (before running make update for the first time, otherwise you lose all the packages you wanted to update!):
#
make clean-update
#
make clean CLEANDEPENDS=YES
#
make update
The following variables can be used either on the
command line or in /etc/mk.conf
to
alter the behaviour of make
clean-update:
CLEAR_DIRLIST
After make clean, do not
reconstruct the list of directories to update for
this package. Only use this if make
update successfully installed all
packages you wanted to update. Normally, this is
done automatically on make
update, but may have been suppressed by
the NOCLEAN
variable (see
above).
Update the installation of the current package. This
differs from update in that it does not replace dependent
packages. You will need to install pkgtools/pkg_tarup
for this
target to work.
Be careful when using this target! There are no guarantees that dependent packages will still work, in particular they will most certainly break if you make replace a library package whose shared library major version changed between your installed version and the new one. For this reason, this target is not officially supported and only recommended for advanced users.
This target invokes pkg_info(1) for the current package. You can use this to check which version of a package is installed.
This is a top-level command, i.e. it should be used in
the pkgsrc
directory. It creates a
database of all packages in the local pkgsrc tree, including
dependencies, comment, maintainer, and some other useful
information. Individual entries are created by running
make describe in the packages'
directories. This index file is saved as
pkgsrc/INDEX
. It can be displayed in
verbose format by running make
print-index. You can search in it with
make search
key=something
. You can
extract a list of all packages that depend on a particular
one by running make show-deps
PKG=somepackage
.
Running this command takes a very long time, some hours even on fast machines!
This target generates a
README.html
file, which can be
viewed using a browser such as www/firefox
or www/links
. The generated files
contain references to any packages which are in the
PACKAGES
directory on the local
host. The generated files can be made to refer to URLs
based on FTP_PKG_URL_HOST
and
FTP_PKG_URL_DIR
. For example, if I
wanted to generate README.html
files which pointed to binary packages on the local
machine, in the directory
/usr/packages
, set
FTP_PKG_URL_HOST=file://localhost
and
FTP_PKG_URL_DIR=/usr/packages
. The
${PACKAGES}
directory and its
subdirectories will be searched for all the binary
packages.
The target can be run at the toplevel or in category directories, in which case it descends recursively.
This is a top-level command, run it in
pkgsrc
. Use this target to create a
file README-all.html
which contains a
list of all packages currently available in the NetBSD
Packages Collection, together with the category they belong
to and a short description. This file is compiled from the
pkgsrc/*/README.html
files, so be sure
to run this after a make
readme.
This is very much the same as the
“readme” target (see above), but is to be
used when generating a pkgsrc tree to be written to a
CD-ROM. This target also produces
README.html
files, and can be made
to refer to URLs based on
CDROM_PKG_URL_HOST
and
CDROM_PKG_URL_DIR
.
This target shows which distfiles and patchfiles
are needed to build the package
(ALLFILES
, which contains all
DISTFILES
and
PATCHFILES
, but not
patches/*
).
This target shows nothing if the package is not installed. If a version of this package is installed, but is not the version provided in this version of pkgsrc, then a warning message is displayed. This target can be used to show which of your installed packages are downlevel, and so the old versions can be deleted, and the current ones added.
This target shows the directory in the pkgsrc hierarchy from which the package can be built and installed. This may not be the same directory as the one from which the package was installed. This target is intended to be used by people who may wish to upgrade many packages on a single host, and can be invoked from the top-level pkgsrc Makefile by using the “show-host-specific-pkgs” target.
This target shows which installed packages match
the current package's DEPENDS
. Useful
if out of date dependencies are causing build
problems.
After a package is installed, check all its
binaries and (on ELF platforms) shared libraries to see
if they find the shared libs they need. Run by default
if PKG_DEVELOPER
is set in
/etc/mk.conf
.
After a “make install” from a new or
upgraded pkg, this prints out an attempt to generate a
new PLIST
from a find
-newer work/.extract_done. An attempt is made
to care for shared libs etc., but it is
strongly recommended to review the
result before putting it into
PLIST
. On upgrades, it's useful to
diff the output of this command against an already
existing PLIST
file.
If the package installs files via tar(1) or
other methods that don't update file access times, be
sure to add these files manually to your
PLIST
, as the “find
-newer” command used by this target won't catch
them!
See Section 12.3, “Tweaking output of make print-PLIST” for more information on this target.
Used to do bulk builds. If an appropriate binary
package already exists, no action is taken. If not, this
target will compile, install and package it (and its
depends, if PKG_DEPENDS
is set
properly. See Section 6.3.1, “Configuration”).
After creating the binary package, the sources, the
just-installed package and its required packages are
removed, preserving free disk space.
Beware that this target may deinstall all packages installed on a system!
Used during bulk-installs to install required packages. If an up-to-date binary package is available, it will be installed via pkg_add(1). If not, make bulk-package will be executed, but the installed binary won't be removed.
A binary package is considered “up-to-date” to be installed via pkg_add(1) if:
None of the package's files
(Makefile
, ...) were modified
since it was built.
None of the package's required (binary) packages were modified since it was built.
Beware that this target may deinstall all packages installed on a system!
Table of Contents
The USE_TOOLS
definition is used both internally
by pkgsrc and also for individual packages to define what commands
are needed for building a package (like BUILD_DEPENDS
)
or for later run-time of an installed packaged (such as
DEPENDS
).
If the native system provides an adequate tool, then in many cases, a pkgsrc
package will not be used.
When building a package, the replacement tools are made available in a directory (as symlinks or wrapper scripts) that is early in the executable search path. Just like the buildlink system, this helps with consistent builds.
A tool may be needed to help build a specific package. For example, perl, GNU make (gmake) or yacc may be needed.
Also a tool may be needed, for example, because the native system's supplied tool may be inefficient for building a package with pkgsrc. For example, a package may need GNU awk, bison (instead of yacc) or a better sed.
The tools used by a package can be listed by running make show-tools.
The default set of tools used by pkgsrc is defined in
bsd.pkg.mk
. This includes standard Unix tools,
such as: cat, awk,
chmod, test, and so on.
These can be seen by running:
make show-var VARNAME=USE_TOOLS.
If a package needs a specific program to build
then the USE_TOOLS
variable can be used
to define the tools needed.
In the following examples, the :pkgsrc means to use the pkgsrc version and not the native version for a build dependency. And the :run means that it is used for a run-time dependencies also (and becomes a DEPENDS). The default is a build dependency which can be set with :build. (So in this example, it is the same as gmake:build and pkg-config:build.)
USE_TOOLS+= mktemp:pkgsrc USE_TOOLS+= gmake perl:run pkg-config
When using the tools framework, a
TOOLS_PATH.foo
variable is defined
which contains the full path to the appropriate tool. For example,
TOOLS_PATH.bash
could be “/bin/bash”
on Linux systems.
If you always need a pkgsrc version of the
tool at run-time, then just use DEPENDS
instead.
When improving or porting pkgsrc to a new platform, have a look
at (or create) the corresponding platform specific make file fragment under
pkgsrc/mk/tools/tools.${OPSYS}.mk
which defines
the name of the common tools. For example:
.if exists(/usr/bin/bzcat) TOOLS_PLATFORM.bzcat?= /usr/bin/bzcat .elif exists(/usr/bin/bzip2) TOOLS_PLATFORM.bzcat?= /usr/bin/bzip2 -cd .endif TOOLS_PLATFORM.true?= true # shell builtin
Table of Contents
mk.conf
One appealing feature of pkgsrc is that it runs on many different platforms. As a result, it is important to ensure, where possible, that packages in pkgsrc are portable. This chapter mentions some particular details you should pay attention to while working on pkgsrc.
The pkgsrc user can configure pkgsrc by overriding several
variables in the file pointed to by MAKECONF
,
which is /etc/mk.conf
by default. When you
want to use those variables in the preprocessor directives of
make(1) (for example .if
or
.for
), you need to include the file
../../mk/bsd.prefs.mk
before, which in turn
loads the user preferences.
But note that some variables may not be completely defined
after ../../mk/bsd.prefs.mk
has been
included, as they may contain references to variables that are
not yet defined. In shell commands this is no problem, since
variables are actually macros, which are only expanded when they
are used. But in the preprocessor directives mentioned above and
in dependency lines (of the form target:
dependencies
) the variables are expanded at load
time.
Currently there is no exhaustive list of all variables that tells you whether they can be used at load time or only at run time, but it is in preparation.
Occasionally, packages require interaction from the user, and this can be in a number of ways:
When fetching the distfiles, some packages require user interaction such as entering username/password or accepting a license on a web page.
When extracting the distfiles, some packages may ask for passwords.
help to configure the package before it is built
help during the build process
help during the installation of a package
The INTERACTIVE_STAGE
definition is
provided to notify the pkgsrc mechanism of an interactive stage
which will be needed, and this should be set in the package's
Makefile
, e.g.:
INTERACTIVE_STAGE= build
Multiple interactive stages can be specified:
INTERACTIVE_STAGE= configure install
The user can then decide to skip this package by setting the
BATCH
variable.
A package may be covered by a license which the user has or has not agreed to accept. For these cases, pkgsrc contains a mechanism to note that a package is covered by a particular license, and the package cannot be built unless the user has accepted the license. (Installation of binary packages are not currently subject to this mechanism.) Packages with licenses that are either Open Source according to the Open Source Initiative or Free according to the Free Software Foundation will not be marked with a license tag. Packages with licenses that have not been determined to meet either definition will be marked with a license tag referring to the license. This will prevent building unless pkgsrc is informed that the license is acceptable, and enables displaying the license.
The license tag mechanism is intended to address
copyright-related issues surrounding building, installing and
using a package, and not to address redistribution issues (see
RESTRICTED
and
NO_SRC_ON_FTP
, etc.). However, the above
definition of licenses for which tags are not needed implies
that packages with redistribution restrictions should have
tags.
Denoting that a package is covered by a particular license
is done by placing the license in
pkgsrc/licenses
and setting the
LICENSE
variable to a string identifying the
license, e.g. in graphics/xv
:
LICENSE= xv-license
When trying to build, the user will get a notice that the package is covered by a license which has not been accepted:
%
make
===> xv-3.10anb9 has an unacceptable license: xv-license. ===> To view the license, enter "/usr/bin/make show-license". ===> To indicate acceptance, add this line to your /etc/mk.conf: ===> ACCEPTABLE_LICENSES+=xv-license *** Error code 1
The license can be viewed with make
show-license, and if it is considered appropriate,
the line printed above can be added to
/etc/mk.conf
to indicate acceptance of
the particular license:
ACCEPTABLE_LICENSES+=xv-license
When adding a package with a new license, the license
text should be added to pkgsrc/licenses
for displaying. A list of known licenses can be seen in this
directory.
The use of LICENSE=shareware
,
LICENSE=no-commercial-use
, and similar
language is deprecated because it does not crisply refer to
a particular license text. Another problem with such usage
is that it does not enable a user to denote acceptance of
the license for a single package without accepting the same
license text for another package. In particular, this can
be inappropriate when e.g. one accepts a particular license to
indicate to pkgsrc that a fee has been paid.
Some licenses restrict how software may be re-distributed. In order to satisfy these restrictions, the package system defines five make variables that can be set to note these restrictions:
RESTRICTED
This variable should be set whenever a restriction exists (regardless of its kind). Set this variable to a string containing the reason for the restriction.
NO_BIN_ON_CDROM
Binaries may not be placed on CD-ROM. Set this
variable to ${RESTRICTED}
whenever a
binary package may not be included on a CD-ROM.
NO_BIN_ON_FTP
Binaries may not be placed on an FTP server. Set
this variable to ${RESTRICTED}
whenever a binary package may not be made available
on the Internet.
NO_SRC_ON_CDROM
Distfiles may not be placed on CD-ROM. Set this
variable to ${RESTRICTED}
if
re-distribution of the source code or other distfile(s) is
not allowed on CD-ROMs.
NO_SRC_ON_FTP
Distfiles may not be placed on FTP. Set this variable
to ${RESTRICTED}
if re-distribution of
the source code or other distfile(s) via the Internet is not
allowed. If this variable is not set, the distfile(s) will
be mirrored on ftp.NetBSD.org.
Please note that the use of NO_PACKAGE
,
IGNORE
, NO_CDROM
, or other
generic make variables to denote restrictions is deprecated,
because they unconditionally prevent users from generating
binary packages!
Your package may depend on some other package being present
- and there are various ways of expressing this dependency.
pkgsrc supports the BUILD_DEPENDS
and
DEPENDS
definitions, the
USE_TOOLS
definition, as well as dependencies
via buildlink3.mk
, which is the preferred way
to handle dependencies, and which uses the variables named above.
See Chapter 13, Buildlink methodology for more information.
The basic difference between the two variables is as
follows: The DEPENDS
definition registers
that pre-requisite in the binary package so it will be pulled in
when the binary package is later installed, whilst the
BUILD_DEPENDS
definition does not, marking a
dependency that is only needed for building the package.
This means that if you only need a package present whilst
you are building, it should be noted as a
BUILD_DEPENDS
.
The format for a BUILD_DEPENDS
and a
DEPENDS
definition is:
<pre-req-package-name>:../../<category>/<pre-req-package>
Please note that the “pre-req-package-name” may include any of the wildcard version numbers recognized by pkg_info(1).
If your package needs another package's binaries or
libraries to build or run, and if that package has a
buildlink3.mk
file available, use it:
.include "../../graphics/jpeg/buildlink3.mk"
If your package needs to use another package to build
itself and there is no buildlink3.mk
file available, use the BUILD_DEPENDS
definition:
BUILD_DEPENDS+= autoconf-2.13:../../devel/autoconf
If your package needs a library with which to link and
again there is no buildlink3.mk
file
available, this is specified using the
DEPENDS
definition. For example:
DEPENDS+= xpm-3.4j:../../graphics/xpm
You can also use wildcards in package dependences:
DEPENDS+= xpm-[0-9]*:../../graphics/xpm
Note that such wildcard dependencies are retained when creating binary packages. The dependency is checked when installing the binary package and any package which matches the pattern will be used. Wildcard dependencies should be used with care.
The “-[0-9]*” should be used instead of
“-*” to avoid potentially ambiguous matches
such as “tk-postgresql” matching a
“tk-*” DEPENDS
.
Wildcards can also be used to specify that a package will only build against a certain minimum version of a pre-requisite:
DEPENDS+= tiff>=3.5.4:../../graphics/tiff
This means that the package will build against version 3.5.4 of the tiff library or newer. Such a dependency may be warranted if, for example, the API of the library has changed with version 3.5.4 and a package would not compile against an earlier version of tiff.
Please note that such dependencies should only be
updated if a package requires a newer pre-requisite, but
not to denote recommendations such as
ABI changes that do not prevent a package from building
correctly. Such recommendations can be expressed using
ABI_DEPENDS
:
ABI_DEPENDS+= tiff>=3.6.1:../../graphics/tiff
In addition to the above DEPENDS
line, this denotes that while a package will build against
tiff>=3.5.4, at least version 3.6.1 is recommended.
ABI_DEPENDS
entries will be turned into
dependencies unless explicitly ignored (in which case a
warning will be printed).
To ignore these ABI dependency recommendations and just
use the required DEPENDS
, set
USE_ABI_DEPENDS=NO
. This may make
it easier and faster to update packages built using pkgsrc,
since older compatible dependencies can continue to be
used. This is useful for people who watch their rebuilds
very carefully; it is not very good as a general-purpose
hammer. If you use it, you need to be mindful of possible
ABI changes, including those from the underlying OS.
Packages that are built with recommendations ignored may not be uploaded to ftp.NetBSD.org by developers and should not be used across different systems that may have different versions of binary packages installed.
For security fixes, please update the package vulnerabilities file. See Section 18.1.10, “Handling packages with security problems” for more information.
If your package needs some executable to be able to run
correctly and if there's no
buildlink3.mk
file, this is specified
using the DEPENDS
variable. The
print/lyx
package needs to
be able to execute the latex binary from the teTeX package
when it runs, and that is specified:
DEPENDS+= teTeX-[0-9]*:../../print/teTeX
The comment about wildcard dependencies from previous paragraph applies here, too.
If your package needs files from another package to build,
add the relevant distribution files to
DISTFILES
, so they will be extracted
automatically. See the print/ghostscript
package for an example.
(It relies on the jpeg sources being present in source form
during the build.)
Your package may conflict with other packages a user might already have installed on his system, e.g. if your package installs the same set of files as another package in the pkgsrc tree.
In this case you can set CONFLICTS
to a
space-separated list of packages (including version string) your
package conflicts with.
For example, x11/Xaw3d
and x11/Xaw-Xpm
install the same shared library, thus you set in
pkgsrc/x11/Xaw3d/Makefile
:
CONFLICTS= Xaw-Xpm-[0-9]*
and in pkgsrc/x11/Xaw-Xpm/Makefile
:
CONFLICTS= Xaw3d-[0-9]*
Packages will automatically conflict with other packages with the name prefix and a different version string. “Xaw3d-1.5” e.g. will automatically conflict with the older version “Xaw3d-1.3”.
There are several reasons why a package might be
instructed to not build under certain circumstances. If the
package builds and runs on most platforms, the exceptions
should be noted with NOT_FOR_PLATFORM
. If
the package builds and runs on a small handful of platforms,
set ONLY_FOR_PLATFORM
instead.
Both ONLY_FOR_PLATFORM
and
NOT_FOR_PLATFORM
are OS triples
(OS-version-platform) that can use glob-style
wildcards.
Some packages are tightly bound to a specific version of an
operating system, e.g. LKMs or sysutils/lsof
. Such binary packages are not
backwards compatible with other versions of the OS, and should be
uploaded to a version specific directory on the FTP server. Mark
these packages by setting OSVERSION_SPECIFIC
to
“yes”. This variable is not currently used by any of
the package system internals, but may be used in the
future.
If the package should be skipped (for example, because it
provides functionality already provided by the system), set
PKG_SKIP_REASON
to a descriptive message. If
the package should fail because some preconditions are not met,
set PKG_FAIL_REASON
to a descriptive
message.
To ensure that a package may not be deleted, once it has been
installed, the PKG_PRESERVE
definition should
be set in the package Makefile. This will be carried into any
binary package that is made from this pkgsrc entry. A
“preserved” package will
not be deleted using pkg_delete(1) unless the
“-f” option is used.
When a vulnerability is found, this should be noted in
localsrc/security/advisories/pkg-vulnerabilities
,
and after committing that file, use make upload
in the same directory to update the file on ftp.NetBSD.org.
After fixing the vulnerability by a patch, its
PKGREVISION
should be increased (this
is of course not necessary if the problem is fixed by using
a newer release of the software).
Also, if the fix should be applied to the stable pkgsrc branch, be sure to submit a pullup request!
Binary packages already on ftp.NetBSD.org will be handled semi-automatically by a weekly cron job.
When making fixes to an existing package it can be useful
to change the version number in PKGNAME
. To
avoid conflicting with future versions by the original author, a
“nb1”, “nb2”, ... suffix can be used
on package versions by setting PKGREVISION=1
(2, ...). The “nb” is treated like a
“.” by the package tools. e.g.
DISTNAME= foo-17.42 PKGREVISION= 9
will result in a PKGNAME
of
“foo-17.42nb9”. If you want to use the original
value of PKGNAME
without the “nbX”
suffix, e.g. for setting DIST_SUBDIR
, use
PKGNAME_NOREV
.
When a new release of the package is released, the
PKGREVISION
should be removed, e.g. on a new
minor release of the above package, things should be like:
DISTNAME= foo-17.43
PKGREVISION
should be incremented for any
non-trivial change in the resulting binary package. Without a
PKGREVISION
bump, someone with the previous
version installed has no way of knowing that their package is out
of date. Thus, changes without increasing
PKGREVISION
are essentially labeled "this is so
trivial that no reasonable person would want to upgrade", and this
is the rough test for when increasing
PKGREVISION
is appropriate. Examples of
changes that do not merit increasing
PKGREVISION
are:
ChangingHOMEPAGE
,MAINTAINER
, or comments in Makefile. Changing build variables if the resulting binary package is the same. ChangingDESCR
. AddingPKG_OPTIONS
if the default options don't change.
Examples of changes that do merit an increase to
PKGREVISION
include:
Security fixes
Changes or additions to a patch file
Changes to the PLIST
PKGREVISION must also be incremented when dependencies have ABI changes.
When you want to replace the same text in multiple files or when the replacement text varies, patches alone cannot help. This is where the SUBST framework comes in. It provides an easy-to-use interface for replacing text in files. Example:
SUBST_CLASSES+= fix-paths SUBST_STAGE.fix-paths= pre-configure SUBST_MESSAGE.fix-paths= Fixing absolute paths. SUBST_FILES.fix-paths= src/*.c SUBST_FILES.fix-paths+= scripts/*.sh SUBST_SED.fix-paths= -e 's,"/usr/local,"${PREFIX},g' SUBST_SED.fix-paths+= -e 's,"/var/log,"${VARBASE}/log,g'
SUBST_CLASSES
is a list of identifiers
that are used to identify the different SUBST blocks that are
defined. The SUBST framework is heavily used by pkgsrc, so it is
important to always use the +=
operator with
this variable. Otherwise some substitutions may be
skipped.
The remaining variables of each SUBST block are
parameterized with the identifier from the first line
(fix-paths
in this case.) They can be seen as
parameters to a function call.
SUBST_STAGE.*
specifies the stage at
which the replacement will take place. All combinations of
pre-
, do-
and
post-
together with a phase name are
possible, though only few are actually used. Most commonly used
are post-patch
and
pre-configure
. Of these two,
pre-configure
should be preferred because
then it is possible to run bmake patch and
have the state after applying the patches but before making any
other changes. This is especially useful when you are debugging
a package in order to create new patches for it. Similarly,
post-build
is preferred over
pre-install
, because the install phase should
generally be kept as simple as possible. When you use
post-build
, you have the same files in the
working directory that will be installed later, so you can check
if the substitution has succeeded.
SUBST_MESSAGE.*
is an optional text
that is printed just before the substitution is done.
SUBST_FILES.*
is the list of shell
globbing patterns that specifies the files in which the
substitution will take place. The patterns are interpreted
relatively to the WRKSRC
directory.
SUBST_SED.*
is a list of arguments to
sed(1) that specify the actual substitution. Every sed
command should be prefixed with -e
, so that
all SUBST blocks look uniform.
There are some more variables, but they are so seldomly
used that they are only documented in the
mk/subst.mk
file.
If you need to download from a dynamic URL you can set
DYNAMIC_MASTER_SITES
and a make
fetch will call files/getsite.sh
with the name of each file to download as an argument, expecting
it to output the URL of the directory from which to download
it. graphics/ns-cult3d
is an
example of this usage.
If the download can't be automated, because the user must
submit personal information to apply for a password, or must pay
for the source, or whatever, you can set
FETCH_MESSAGE
to a list of lines that are
displayed to the user before aborting the build. Example:
FETCH_MESSAGE= "Please download the files" FETCH_MESSAGE+= " "${DISTFILES:Q} FETCH_MESSAGE+= "manually from "${MASTER_SITES:Q}"."
Sometimes authors of a software package make some
modifications after the software was released, and they put up a
new distfile without changing the package's version number. If a
package is already in pkgsrc at that time, the checksum will
no longer match. The contents of the new distfile should be
compared against the old one before changing anything, to make
sure the distfile was really updated on purpose, and that
no trojan horse or so crept in.
Please mention that the distfiles were compared and what was found
in your commit message.
Then, the correct way to work around this is to
set DIST_SUBDIR
to a unique directory name,
usually based on PKGNAME_NOREV
. All
DISTFILES
and
PATCHFILES
for this package will be put in that
subdirectory of the local distfiles directory.
(See Section 18.1.11, “How to handle incrementing versions when fixing an existing package” for more details.)
In case this
happens more often, PKGNAME
can be used (thus
including the nbX
suffix) or a date stamp
can be appended, like ${PKGNAME_NOREV}-YYYYMMDD
.
Do not forget regenerating the distinfo
file
after that, since it contains the DIST_SUBDIR
path in the filenames.
Also increase the PKGREVISION if the installed package is different.
Furthermore, a mail to the package's authors seems appropriate
telling them that changing distfiles after releases without
changing the file names is not good practice.
pkgsrc supports many different machines, with different
object formats like a.out and ELF, and varying abilities to do
shared library and dynamic loading at all. To accompany this,
varying commands and options have to be passed to the
compiler, linker, etc. to get the Right Thing, which can be
pretty annoying especially if you don't have all the machines
at your hand to test things. The
devel/libtool
pkg
can help here, as it just “knows” how to build
both static and dynamic libraries from a set of source files,
thus being platform-independent.
Here's how to use libtool in a package in seven simple steps:
Add USE_LIBTOOL=yes
to the package
Makefile.
For library objects, use “${LIBTOOL} --mode=compile
${CC}” in place of “${CC}”. You could even
add it to the definition of CC
, if only
libraries are being built in a given Makefile. This one command
will build both PIC and non-PIC library objects, so you need not
have separate shared and non-shared library rules.
For the linking of the library, remove any “ar”, “ranlib”, and “ld -Bshareable” commands, and instead use:
${LIBTOOL} --mode=link ${CC} -o ${.TARGET:.a=.la} ${OBJS:.o=.lo} \ -rpath ${PREFIX}/lib -version-info major:minor
Note that the library is changed to have a
.la
extension, and the objects are
changed to have a .lo
extension. Change OBJS
as
necessary. This automatically creates all of the
.a
,
.so.major.minor
, and ELF symlinks (if
necessary) in the build directory. Be sure to include
“-version-info”, especially when major and
minor are zero, as libtool will otherwise strip off the
shared library version.
From the libtool manual:
So, libtool library versions are described by three integers: CURRENT The most recent interface number that this library implements. REVISION The implementation number of the CURRENT interface. AGE The difference between the newest and oldest interfaces that this library implements. In other words, the library implements all the interface numbers in the range from number `CURRENT - AGE' to `CURRENT'. If two libraries have identical CURRENT and AGE numbers, then the dynamic linker chooses the library with the greater REVISION number.
The “-release” option will produce different results for a.out and ELF (excluding symlinks) in only one case. An ELF library of the form “libfoo-release.so.x.y” will have a symlink of “libfoo.so.x.y” on an a.out platform. This is handled automatically.
The “-rpath argument” is the install directory of the library being built.
In the PLIST
, include only the
.la
file, the other files will be
added automatically.
When linking shared object (.so
)
files, i.e. files that are loaded via dlopen(3), NOT
shared libraries, use “-module
-avoid-version” to prevent them getting version
tacked on.
The PLIST
file gets the
foo.so
entry.
When linking programs that depend on these libraries
before they are installed, preface
the cc(1) or ld(1) line with “${LIBTOOL}
--mode=link”, and it will find the correct
libraries (static or shared), but please be aware that
libtool will not allow you to specify a relative path in
-L (such as “-L../somelib”), because it
expects you to change that argument to be the
.la
file. e.g.
${LIBTOOL} --mode=link ${CC} -o someprog -L../somelib -lsomelib
should be changed to:
${LIBTOOL} --mode=link ${CC} -osomeprog
../somelib/somelib.la
and it will do the right thing with the libraries.
When installing libraries, preface the install(1)
or cp(1) command with “${LIBTOOL}
--mode=install”, and change the library name to
.la
. e.g.
${LIBTOOL} --mode=install ${BSD_INSTALL_DATA} ${SOMELIB:.a=.la} ${PREFIX}/lib
This will install the static .a
,
shared library, any needed symlinks, and run
ldconfig(8).
In your PLIST
, include only
the .la
file (this is a change from previous behaviour).
Add USE_LIBTOOL=yes
to the
package Makefile. This will override the package's own libtool
in most cases. For older libtool using packages, libtool is
made by ltconfig script during the do-configure step; you can
check the libtool script location by doing make
configure; find work*/ -name libtool.
LIBTOOL_OVERRIDE
specifies which libtool
scripts, relative to WRKSRC
, to override. By
default, it is set to “libtool */libtool
*/*/libtool”. If this does not match the location of the
package's libtool script(s), set it as appropriate.
If you do not need *.a
static
libraries built and installed, then use
SHLIBTOOL_OVERRIDE
instead.
If your package makes use of the platform-independent library for loading dynamic shared objects, that comes with libtool (libltdl), you should include devel/libltdl/buildlink3.mk.
Some packages use libtool incorrectly so that the package may not work or build in some circumstances. Some of the more common errors are:
The inclusion of a shared object (-module) as a dependent library in an executable or library. This in itself isn't a problem if one of two things has been done:
The shared object is named correctly, i.e.
libfoo.la
, not
foo.la
The -dlopen option is used when linking an executable.
The use of libltdl without the correct calls to initialisation routines.
The function lt_dlinit() should be called and the macro
LTDL_SET_PRELOADED_SYMBOLS
included in
executables.
If a package needs GNU autoconf or automake to be executed to regenerate the configure script and Makefile.in makefile templates, then they should be executed in a pre-configure target.
For packages that need only autoconf:
AUTOCONF_REQD= 2.50 # if default version is not good enough USE_TOOLS+= autoconf # use "autoconf213" for autoconf-2.13 ... pre-configure: cd ${WRKSRC}; autoconf ...
and for packages that need automake and autoconf:
AUTOMAKE_REQD= 1.7.1 # if default version is not good enough USE_TOOLS+= automake # use "automake14" for automake-1.4 ... pre-configure: cd ${WRKSRC}; \ aclocal; autoheader; \ automake -a --foreign -i; autoconf ...
Packages which use GNU Automake will almost certainly require GNU Make.
There are times when the configure process makes
additional changes to the generated files, which then causes
the build process to try to re-execute the automake sequence.
This is prevented by touching various files in the configure
stage. If this causes problems with your package you can set
AUTOMAKE_OVERRIDE=NO
in the package
Makefile.
Compilers for the C, C++, and Fortran languages comes with the NetBSD base system. By default, pkgsrc assumes that a package is written in C and will hide all other compilers (via the wrapper framework, see Chapter 13, Buildlink methodology).
To declare which language's compiler a package needs, set
the USE_LANGUAGES
variable. Allowed values
currently are “c”, “c++”, and
“fortran” (and any combination). The default is
“c”. Packages using GNU configure scripts, even if
written in C++, usually need a C compiler for the configure
phase.
If a program is written in Java, use the Java framework in
pkgsrc. The package must include
../../mk/java-vm.mk
. This Makefile fragment
provides the following variables:
USE_JAVA
defines if a build
dependency on the JDK is added. If
USE_JAVA
is set to “run”, then
there is only a runtime dependency on the JDK. The default is
“yes”, which also adds a build dependency on the
JDK.
Set USE_JAVA2
to declare that
a package needs a Java2 implementation. The supported values
are “yes”, “1.4”, and
“1.5”. “yes” accepts any Java2
implementation, “1.4” insists on versions 1.4 or
above, and “1.5” only accepts versions 1.5 or
above. This variable is not set by default.
If your package contains interpreted perl scripts, add
“perl” to the USE_TOOLS
variable
and set REPLACE_PERL
to ensure that the proper
interpreter path is set. REPLACE_PERL
should
contain a list of scripts, relative to WRKSRC
,
that you want adjusted. Every occurrence of
*/bin/perl
will be replaced with the full
path to the perl executable.
If a particular version of perl is needed, set the
PERL5_REQD
variable to the version number. The
default is “5.0”.
See Section 18.6.5, “Packages installing perl modules” for information about handling perl modules.
The most common failures when building a package are that some platforms do not provide certain header files, functions or libraries, or they provide the functions in a library that the original package author didn't know. To work around this, you can rewrite the source code in most cases so that it does not use the missing functions or provides a replacement function.
If a package already comes with a GNU configure script, the
preferred way to fix the build failure is to change the
configure script, not the code. In the other cases, you can
utilize the C preprocessor, which defines certain macros
depending on the operating system and hardware architecture it
compiles for. These macros can be queried using for example
#if defined(__i386)
. Almost every operating
system, hardware architecture and compiler has its own macro.
For example, if the macros __GNUC__
,
__i386__
and __NetBSD__
are all defined, you know that you are using NetBSD on an i386
compatible CPU, and your compiler is GCC.
The list of the following macros for hardware and
operating system depends on the compiler that is used. For
example, if you want to conditionally compile code on Solaris,
don't use __sun__
, as the SunPro compiler
does not define it. Use __sun
instead.
To distinguish between 4.4 BSD-derived systems and the rest of the world, you should use the following code.
#include <sys/param.h> #if (defined(BSD) && BSD >= 199306) /* BSD-specific code goes here */ #else /* non-BSD-specific code goes here */ #endif
If this distinction is not fine enough, you can also test for the following macros.
FreeBSD __FreeBSD__ DragonFly __DragonFly__ Interix __INTERIX IRIX __sgi (TODO: get a definite source for this) Linux linux, __linux, __linux__ NetBSD __NetBSD__ OpenBSD __OpenBSD__ Solaris sun, __sun
i386 i386, __i386, __i386__ MIPS __mips SPARC sparc, __sparc
Some source files trigger bugs in the compiler, based on combinations of compiler version and architecture and almost always relation to optimisation being enabled. Common symptoms are gcc internal errors or never finishing compiling a file.
Typically, a workaround involves testing the
MACHINE_ARCH
and compiler version, disabling
optimisation for that combination of file,
MACHINE_ARCH
and compiler, and documenting it
in pkgsrc/doc/HACKS
. See that file for a
number of examples.
This error message often means that a package did not link to a shared library it needs. The following functions are known to cause this error message over and over.
Function | Library | Affected platforms |
---|---|---|
accept, bind, connect | -lsocket | Solaris |
crypt | -lcrypt | DragonFly, NetBSD |
dlopen, dlsym | -ldl | Linux |
gethost* | -lnsl | Solaris |
inet_aton | -lresolv | Solaris |
nanosleep, sem_*, timer_* | -lrt | Solaris |
openpty | -lutil | Linux |
To fix these linker errors, it is often sufficient to say
LIBS.
to the package
OperatingSystem
+=
-lfoo
Makefile
and then say bmake clean;
bmake.
When you are using the SunPro compiler, there is another possibility. That compiler cannot handle the following code:
extern int extern_func(int); static inline int inline_func(int x) { return extern_func(x); } int main(void) { return 0; }
It generates the code for inline_func
even if
that function is never used. This code then refers to
extern_func
, which can usually not be resolved. To
solve this problem you can try to tell the package to disable inlining
of functions.
Sometimes packages fail to build because the compiler runs
into an operating system specific soft limit. With the
UNLIMIT_RESOURCES
variable pkgsrc can be told
to unlimit the resources. Currently, the allowed values are
“datasize” and “stacksize” (or both).
Setting this variable is similar to running the shell builtin
ulimit command to raise the maximum data
segment size or maximum stack size of a process, respectively, to
their hard limits.
The BSD-compatible install supplied
with some operating systems cannot create more than one
directory at a time. As such, you should call
${INSTALL_*_DIR}
like this:
${INSTALL_DATA_DIR} ${PREFIX}/dir1 ${INSTALL_DATA_DIR} ${PREFIX}/dir2
You can also just append “dir1
dir2
” to the
INSTALLATION_DIRS
variable, which will
automatically do the right thing.
In general, documentation should be installed into
${PREFIX}/share/doc/${PKGBASE}
or
${PREFIX}/share/doc/${PKGNAME}
(the latter
includes the version number of the package).
Many modern packages using GNU autoconf allow to set the
directory where HTML documentation is installed with the
“--with-html-dir” option. Sometimes using this flag
is needed because otherwise the documentation ends up in
${PREFIX}/share/doc/html
or other
places.
An exception to the above is that library API documentation
generated with the textproc/gtk-doc
tools, for use by special
browsers (devhelp) should be left at their default location, which
is ${PREFIX}/share/gtk-doc
. Such
documentation can be recognized from files ending in
.devhelp
or .devhelp2
.
(It is also acceptable to install such files in
${PREFIX}/share/doc/${PKGBASE}
or
${PREFIX}/share/doc/${PKGNAME}
; the
.devhelp*
file must be directly in that
directory then, no additional subdirectory level is allowed in
this case. This is usually achieved by using
“--with-html-dir=${PREFIX}/share/doc”.
${PREFIX}/share/gtk-doc
is preferred
though.)
Certain packages, most of them in the games category, install
a score file that allows all users on the system to record their
highscores. In order for this to work, the binaries need to be
installed setgid and the score files owned by the appropriate
group and/or owner (traditionally the "games" user/group). The
following variables, documented in more detail in
mk/defaults/mk.conf
, control this
behaviour: SETGIDGAME
,
GAMEDATAMODE
, GAMEGRP
,
GAMEMODE
, GAMEOWN
.
Note that per default, setgid installation of games is
disabled; setting SETGIDGAME=YES
will set all
the other variables accordingly.
A package should therefor never hard code file ownership or
access permissions but rely on INSTALL_GAME
and
INSTALL_GAME_DATA
to set these
correctly.
Your package may also contain scripts with hardcoded paths to
other interpreters besides (or as well as) perl. To correct the
full pathname to the script interpreter, you need to set the
following definitions in your Makefile
(we
shall use tclsh in this example):
REPLACE_INTERPRETER+= tcl REPLACE.tcl.old= .*/bin/tclsh REPLACE.tcl.new= ${PREFIX}/bin/tclsh REPLACE_FILES.tcl= # list of tcl scripts which need to be fixed, # relative to ${WRKSRC}, just as in REPLACE_PERL
Before March 2006, these variables were called
_REPLACE.*
and
_REPLACE_FILES.*
.
Makefiles of packages providing perl5 modules should include
the Makefile fragment
../../lang/perl5/module.mk
. It provides a
do-configure target for the standard perl
configuration for such modules as well as various hooks to tune
this configuration. See comments in this file for
details.
Perl5 modules will install into different places depending
on the version of perl used during the build process. To
address this, pkgsrc will append lines to the
PLIST
corresponding to the files listed in
the installed .packlist
file generated by
most perl5 modules. This is invoked by defining
PERL5_PACKLIST
to a space-separated list of
paths to packlist files, e.g.:
PERL5_PACKLIST= ${PERL5_SITEARCH}/auto/Pg/.packlist
The variables PERL5_SITELIB
,
PERL5_SITEARCH
, and
PERL5_ARCHLIB
represent the three locations
in which perl5 modules may be installed, and may be used by
perl5 packages that don't have a packlist. These three
variables are also substituted for in the
PLIST
.
Some packages install info files or use the
“makeinfo” or “install-info”
commands. INFO_FILES
should be defined in
the package Makefile so that INSTALL
and
DEINSTALL
scripts will be generated to
handle registration of the info files in the Info directory
file. The “install-info” command used for the info
files registration is either provided by the system, or by a
special purpose package automatically added as dependency if
needed.
PKGINFODIR
is the directory under
${PREFIX}
where info files are primarily
located. PKGINFODIR
defaults to
“info” and can be overridden by the user.
The info files for the package should be listed in the
package PLIST
; however any split info files
need not be listed.
A package which needs the “makeinfo” command
at build time must add “makeinfo” to
USE_TOOLS
in its Makefile. If a minimum
version of the “makeinfo” command is needed it
should be noted with the TEXINFO_REQD
variable in the package Makefile
. By
default, a minimum version of 3.12 is required. If the system
does not provide a makeinfo command or if it
does not match the required minimum, a build dependency on the
devel/gtexinfo
package will
be added automatically.
The build and installation process of the software provided
by the package should not use the
install-info command as the registration of
info files is the task of the package
INSTALL
script, and it must use the
appropriate makeinfo command.
To achieve this goal, the pkgsrc infrastructure creates
overriding scripts for the install-info and
makeinfo commands in a directory listed early
in PATH
.
The script overriding install-info has
no effect except the logging of a message. The script overriding
makeinfo logs a message and according to the
value of TEXINFO_REQD
either runs the appropriate
makeinfo command or exit on error.
All packages that install manual pages should install them
into the same directory, so that there is one common place to look
for them. In pkgsrc, this place is
${PREFIX}/${PKGMANDIR}
, and this expression
should be used in packages. The default for
PKGMANDIR
is
“man
”. Another often-used value
is “share/man
”.
The support for a custom PKGMANDIR
is far from complete.
The PLIST
files can just use
man/
as the top level directory for the man
page file entries, and the pkgsrc framework will convert as
needed. In all other places, the correct
PKGMANDIR
must be used.
Packages that are
configured with GNU_CONFIGURE
set as
“yes”, by default will use the
./configure
--mandir switch to set where the man pages should be installed.
The path is GNU_CONFIGURE_MANDIR
which defaults
to ${PREFIX}/${PKGMANDIR}
.
Packages that use GNU_CONFIGURE
but do not
use --mandir, can set CONFIGURE_HAS_MANDIR
to “no”.
Or if the ./configure
script uses
a non-standard use of --mandir, you can set
GNU_CONFIGURE_MANDIR
as needed.
See Section 12.5, “Man page compression” for information on installation of compressed manual pages.
If a package installs .schemas
or
.entries
files, used by GConf2,
you need to take some extra steps to make sure they get registered
in the database:
Include ../../devel/GConf2/schemas.mk
instead of its buildlink3.mk
file. This
takes care of rebuilding the GConf2 database at installation and
deinstallation time, and tells the package where to install
GConf2 data files using some standard configure arguments. It
also disallows any access to the database directly from the
package.
Ensure that the package installs its
.schemas
files under
${PREFIX}/share/gconf/schemas
. If they get
installed under ${PREFIX}/etc
, you will
need to manually patch the package.
Check the PLIST and remove any entries under the etc/gconf directory, as they will be handled automatically. See Section 8.14, “How do I change the location of configuration files?” for more information.
Define the GCONF2_SCHEMAS
variable in
your Makefile
with a list of all
.schemas
files installed by the package, if
any. Names must not contain any directories in them.
Define the GCONF2_ENTRIES
variable in
your Makefile
with a
list of all .entries
files installed by the
package, if any. Names must not contain any directories in
them.
If a package installs .omf
files, used by
scrollkeeper, you need to take some extra steps to make sure they
get registered in the database:
Include
../../textproc/scrollkeeper/omf.mk
instead of its buildlink3.mk
file. This
takes care of rebuilding the scrollkeeper database at
installation and deinstallation time, and disallows any access
to it directly from the package.
Check the PLIST and remove any entries under the
libdata/scrollkeeper
directory, as they
will be handled automatically.
Remove the share/omf
directory from
the PLIST. It will be handled by scrollkeeper.
If a package installs font files, you will need to rebuild the fonts database in the directory where they get installed at installation and deinstallation time. This can be automatically done by using the pkginstall framework.
You can list the directories where fonts are installed in the
FONTS_DIRS.
variables, where type
type
can be one of
“ttf”, “type1” or “x11”.
Also make sure that the database file
fonts.dir
is not listed in the PLIST.
Note that you should not create new directories for fonts; instead use the standard ones to avoid that the user needs to manually configure his X server to find them.
If a package installs GTK2 immodules or loaders, you need to take some extra steps to get them registered in the GTK2 database properly:
Include
../../x11/gtk2/modules.mk
instead of its
buildlink3.mk
file. This takes care of
rebuilding the database at installation and deinstallation time.
Set GTK2_IMMODULES=YES
if
your package installs GTK2 immodules.
Set GTK2_LOADERS=YES
if your package installs
GTK2 loaders.
Patch the package to not touch any of the GTK2 databases directly. These are:
libdata/gtk-2.0/gdk-pixbuf.loaders
libdata/gtk-2.0/gtk.immodules
Check the PLIST
and remove
any entries under the libdata/gtk-2.0
directory, as they will be handled automatically.
If a package installs SGML or XML data files that need to be registered in system-wide catalogs (like DTDs, sub-catalogs, etc.), you need to take some extra steps:
Include
../../textproc/xmlcatmgr/catalogs.mk
in
your Makefile
, which takes care of
registering those files in system-wide catalogs at
installation and deinstallation time.
Set SGML_CATALOGS
to the full path of
any SGML catalogs installed by the package.
Set XML_CATALOGS
to the full path of
any XML catalogs installed by the package.
Set SGML_ENTRIES
to individual entries
to be added to the SGML catalog. These come in groups of
three strings; see xmlcatmgr(1) for more information
(specifically, arguments recognized by the 'add' action).
Note that you will normally not use this variable.
Set XML_ENTRIES
to individual entries
to be added to the XML catalog. These come in groups of three
strings; see xmlcatmgr(1) for more information (specifically,
arguments recognized by the 'add' action). Note that you will
normally not use this variable.
If a package provides extensions to the MIME database by
installing .xml
files inside
${PREFIX}/share/mime/packages
, you
need to take some extra steps to ensure that the database is kept
consistent with respect to these new files:
Include
../../databases/shared-mime-info/mimedb.mk
(avoid using the buildlink3.mk
file from
this same directory, which is reserved for inclusion from
other buildlink3.mk
files). It takes
care of rebuilding the MIME database at installation and
deinstallation time, and disallows any access to it directly
from the package.
Check the PLIST and remove any entries under the
share/mime
directory,
except for files saved under
share/mime/packages
. The former are
handled automatically by
the update-mime-database program, but the latter are
package-dependent and must be removed by the package that
installed them in the first place.
Remove any share/mime/*
directories
from the PLIST. They will be handled by the shared-mime-info
package.
If a package uses intltool during its build, add
intltool
to the USE_TOOLS
,
which forces it to use the intltool package provided by pkgsrc,
instead of the one bundled with the distribution file.
This tracks intltool's build-time dependencies and uses the latest available version; this way, the package benefits of any bug fixes that may have appeared since it was released.
If a package contains a rc.d script, it won't be copied into
the startup directory by default, but you can enable it, by adding
the option PKG_RCD_SCRIPTS=YES
in
/etc/mk.conf
. This option will copy the scripts
into /etc/rc.d
when a package is installed, and
it will automatically remove the scripts when the package is
deinstalled.
If a package installs TeX packages into the texmf tree,
the ls-R
database of the tree needs to be
updated.
Except the main TeX packages such as teTeX-texmf,
packages should install files
into PKG_LOCALTEXMFPREFIX
,
not PKG_TEXMFPREFIX
.
Include
../../print/teTeX/module.mk
instead
of ../../mk/tex.buildlink3.mk
. This
takes care of rebuilding the ls-R
database at installation and deinstallation time.
If your package installs files into a texmf
tree other than the one
at PKG_LOCALTEXMFPREFIX
,
set TEXMFDIRS
to the list of all texmf
trees that need database update.
If your package also installs font map files that need
to be registered using updmap,
set TEX_FONTMAPS
to the list of all
such font map files. Then updmap will
be run automatically at installation/deinstallation to
enable/disable font map files for TeX output
drivers.
Make sure that none of ls-R
databases are included in PLIST
, as
they will be removed only by the teTeX-bin package.
There are some packages that provide libraries and executables for running binaries from a one operating system on a different one (if the latter supports it). One example is running Linux binaries on NetBSD.
The pkgtools/rpm2pkg
helps in extracting and packaging Linux rpm packages.
The CHECK_SHLIBS
can be set to no to
avoid the check-shlibs target, which tests
if all libraries for each installed executable can be found by
the dynamic linker. Since the standard dynamic linker is run,
this fails for emulation packages, because the libraries used
by the emulation are not in the standard directories.
If a package installs images under the
share/icons/hicolor
and/or updates the
share/icons/hicolor/icon-theme.cache
database, you need to take some extra steps to make sure that the
shared theme directory is handled appropriately and that the cache
database is rebuilt:
Include
../../graphics/hicolor-icon-theme/buildlink3.mk
.
Check the PLIST
and remove the
entry that refers to the theme cache.
Ensure that the PLIST does not remove the shared icon
directories from the share/icons/hicolor
hierarchy because they will be handled automatically.
The best way to verify that the PLIST is correct with respect to the last two points is to regenerate it using make print-PLIST.
If a package installs .desktop
files
under share/applications
and these include
MIME information, you need to take extra steps to ensure that they
are registered into the MIME database:
Include
../../sysutils/desktop-file-utils/desktopdb.mk
.
Check the PLIST and remove the entry that refers to the
share/applications/mimeinfo.cache
file.
It will be handled automatically.
The best way to verify that the PLIST is correct with respect to the last point is to regenerate it using make print-PLIST.
In some cases one does not have the time to solve a problem immediately. There are currently two ways to declare that one knows that a package has problems.
The first way is to plainly mark it as broken. For
this, one just sets the variable BROKEN
to the
reason why the package is broken (similar to the
RESTRICTED
variable). A user trying to build
the package will immediately be shown this message, and the build
will not be even tried.
After each pkgsrc freeze period (a time when the
tree is stabilized and a new pkgsrc branch is cut), the packages
that were not building in the official branch build on the latest
NetBSD release will be marked as broken on that branch. This is
done by setting the BROKEN_IN
variable to the
branch name (or appending the branch name to it). If a user tries
to build such a package and the build fails, the user gets a
message that says that the package was broken on the respective
branch(es).
Both types of packages are removed from pkgsrc in irregular intervals.
To check out all the gotchas when building a package, here are the steps that I do in order to get a package working. Please note this is basically the same as what was explained in the previous sections, only with some debugging aids.
Be sure to set PKG_DEVELOPER=1
in
/etc/mk.conf
Install pkgtools/url2pkg
,
create a directory for a new package, change into it, then run
url2pkg:
%
mkdir /usr/pkgsrc/
category
/examplepkg
%
cd /usr/pkgsrc/
category
/examplepkg
%
url2pkg http://www.example.com/path/to/distfile.tar.gz
Edit the Makefile
as requested.
Fill in the DESCR
file
Run make configure
Add any dependencies glimpsed from documentation and the
configure step to the package's
Makefile
.
Make the package compile, doing multiple rounds of
%
make
%
pkgvi ${WRKSRC}/some/file/that/does/not/compile
%
mkpatches
%
patchdiff
%
mv ${WRKDIR}/.newpatches/* patches
%
make mps
%
make clean
Doing this step as non-root user will ensure that no files
are modified that shouldn't be, especially during the build
phase. mkpatches,
patchdiff and pkgvi are
from the pkgtools/pkgdiff
package.
Look at the Makefile
, fix if
necessary; see Section 10.1, “Makefile
”.
Generate a PLIST
:
#
make install
#
make print-PLIST >PLIST
#
make deinstall
#
make install
#
make deinstall
You usually need to be root
to do
this. Look if there are any files left:
#
make print-PLIST
If this reveals any files that are missing in
PLIST
, add them.
Now that the PLIST
is OK, install the
package again and make a binary package:
#
make reinstall
#
make package
Delete the installed package:
#
pkg_delete
examplepkg
Repeat the above make print-PLIST command, which shouldn't find anything now:
#
make print-PLIST
Reinstall the binary package:
#
pkg_add .../
examplepkg
.tgz
Play with it. Make sure everything works.
Run pkglint from pkgtools/pkglint
, and fix the problems it
reports:
#
pkglint
Submit (or commit, if you have cvs access); see Chapter 20, Submitting and Committing.
Table of Contents
Our policy is that we accept binaries only from pkgsrc developers to guarantee that the packages don't contain any trojan horses etc. This is not to annoy anyone but rather to protect our users! You're still free to put up your home-made binary packages and tell the world where to get them. NetBSD developers doing bulk builds and wanting to upload them please see Section 6.3.8, “Uploading results of a bulk build”.
First, check that your package is complete, compiles and runs well; see Chapter 19, Debugging and the rest of this document. Next, generate an uuencoded gzipped tar(1) archive that contains all files that make up the package. Finally, send this package to the pkgsrc bug tracking system, either with the send-pr(1) command, or if you don't have that, go to the web page http://www.NetBSD.org/Misc/send-pr.html, which contains some instructions and a link to a form, where you can submit packages.
In the form of the problem report, the category should be “pkg”, the synopsis should include the package name and version number, and the description field should contain a short description of your package (contents of the COMMENT variable or DESCR file are OK). The uuencoded package data should go into the “fix” field.
If you want to submit several packages, please send a separate PR for each one, it's easier for us to track things that way.
Alternatively, you can also import new packages into pkgsrc-wip (“pkgsrc work-in-progress”); see the homepage at http://pkgsrc-wip.sourceforge.net/ for details.
Please note all package additions, updates, moves, and
removals in pkgsrc/doc/CHANGES-
. It's very
important to keep this file up to date and conforming to the
existing format, because it will be used by scripts to
automatically update pages on www.NetBSD.org and other
sites. Additionally, check the
YYYY
pkgsrc/doc/TODO
file and remove the entry
for the package you updated or removed, in case it was mentioned
there.
When the PKGREVISION
of a package is
bumped, the change should appear in
pkgsrc/doc/CHANGES-
if it is security
related or otherwise relevant. Mass bumps that result from a
dependency being updated should not be mentioned. In all other
cases it's the developer's decision.YYYY
There is a make target that helps in creating proper
CHANGES-
entries: make
changes-entry. It uses the optional YYYY
CTYPE
and NETBSD_LOGIN_NAME
variables. The general
usage is to first make sure that your CHANGES-
file is up-to-date (to avoid having to resolve conflicts later-on)
and then to cd to the package directory. For
package updates, make changes-entry is enough.
For new packages, or package moves or removals, set the
YYYY
CTYPE
variable on the command line to "Added",
"Moved", or "Removed". You can set NETBSD_LOGIN_NAME
in /etc/mk.conf
if your local login name is
not the same as your NetBSD login name. Don't forget to commit
the changes to pkgsrc/doc/CHANGES-
!YYYY
This section is only of interest for pkgsrc developers with write access to the pkgsrc repository. Please remember that cvs imports files relative to the current working directory, and that the pathname that you give the cvs import command is so that it knows where to place the files in the repository. Newly created packages should be imported with a vendor tag of “TNF” and a release tag of “pkgsrc-base”, e.g:
$
cd .../pkgsrc/category/pkgname$
cvs import pkgsrc/category/pkgname TNF pkgsrc-base
Remember to move the directory from which you imported out of
the way, or cvs will complain the next time you “cvs
update” your source tree. Also don't forget to add the new
package to the category's Makefile
.
The commit message of the initial import should include part of the
DESCR
file, so people reading the mailing lists know
what the package is/does.
For new packages, “cvs import” is preferred to “cvs add” because the former gets everything with a single command, and provides a consistent tag.
Please always put a concise, appropriate and relevant summary of the changes between old and new versions into the commit log when updating a package. There are various reasons for this:
A URL is volatile, and can change over time. It may go away completely or its information may be overwritten by newer information.
Having the change information between old and new versions in our CVS repository is very useful for people who use either cvs or anoncvs.
Having the change information between old and new versions in our CVS repository is very useful for people who read the pkgsrc-changes mailing list, so that they can make tactical decisions about when to upgrade the package.
Please also recognize that, just because a new version of a package has been released, it should not automatically be upgraded in the CVS repository. We prefer to be conservative in the packages that are included in pkgsrc - development or beta packages are not really the best thing for most places in which pkgsrc is used. Please use your judgement about what should go into pkgsrc, and bear in mind that stability is to be preferred above new and possibly untested features.
Make a copy of the directory somewhere else.
Remove all CVS dirs.
Alternatively to the first two steps you can also do:
%
cvs -d user@cvs.NetBSD.org:/cvsroot export -D today pkgsrc/category/package
and use that for further work.
Fix CATEGORIES
and any
DEPENDS
paths that just did “../package”
instead of “../../category/package”.
cvs import the modified package in the new place.
Check if any package depends on it:
%
cd /usr/pkgsrc
%
grep /package */*/Makefile* */*/buildlink*
Fix paths in packages from step 5 to point to new location.
cvs rm (-f) the package at the old location.
Remove from oldcategory/Makefile
.
Add to newcategory/Makefile
.
Commit the changed and removed files:
%
cvs commit oldcategory/package oldcategory/Makefile newcategory/Makefile
(and any packages from step 5, of course).
This section contains the answers to questions that may
arise when you are writing a package. If you don't find your
question answered here, first have a look in the other chapters,
and if you still don't have the answer, ask on the
pkgsrc-users
mailing list.
21.1. | What is the difference between
|
|
|
21.2. | What is the difference between
|
|
|
21.3. | What is the difference between
|
|
|
21.4. | What is the difference between
|
[FIXME] |
|
21.5. | Why does make show-var
VARNAME=BUILDLINK_PREFIX. |
For optimization reasons, some variables are only available in the “wrapper” phase and later. To “simulate” the wrapper phase, append PKG_PHASE=wrapper to the above command. |
|
21.6. | What does
|
The |
|
21.7. | Which mailing lists are there for package developers? |
|
|
21.8. | Where is the pkgsrc documentation? |
There are many places where you can find documentation about pkgsrc:
|
|
21.9. | I have a little time to kill. What shall I do? |
This is not really an FAQ yet, but here's the answer anyway.
|
Table of Contents
Quoting GNOME's web site:
The GNOME project provides two things: The GNOME desktop environment, an intuitive and attractive desktop for users, and the GNOME development platform, an extensive framework for building applications that integrate into the rest of the desktop.
pkgsrc provides a seamless way to automatically build and install a complete GNOME environment under many different platforms. We can say with confidence that pkgsrc is one of the most advanced build and packaging systems for GNOME due to its included technologies buildlink3, the wrappers and tools framework and automatic configuration file management. Lots of efforts are put into achieving a completely clean deinstallation of installed software components.
Given that pkgsrc is NetBSD's official packaging system, the above also means that great efforts are put into making GNOME work under this operating system. Recently, DragonFly BSD also adopted pkgsrc as its preferred packaging system, contributing lots of portability fixes to make GNOME build and install under it.
This chapter is aimed at pkgsrc developers and other people interested in helping our GNOME porting and packaging efforts. It provides instructions on how to manage the existing packages and some important information regarding their internals.
Should you have some spare cycles to devote to NetBSD, pkgsrc and GNOME and are willing to learn new exciting stuff, please jump straight to the pending work list! There is still a long way to go to get a fully-functional GNOME desktop under NetBSD and we need your help to achieve it!
pkgsrc includes three GNOME-related meta packages:
meta-pkgs/gnome-base
: Provides
the core GNOME desktop environment. It only includes the necessary
bits to get it to boot correctly, although it may lack important
functionality for daily operation. The idea behind this package is
to let end users build their own configurations on top of this one,
first installing this meta package to achieve a functional setup and
then adding individual applications.
meta-pkgs/gnome
: Provides a
complete installation of the GNOME platform and desktop as defined
by the GNOME project; this is based on the components distributed in
the platform/x.y/x.y.z/sources
and
desktop/x.y/x.y.z/sources
directories of the
official FTP server. Developer-only tools found in those
directories are not installed unless required by some other
component to work properly. Similarly, packages from the bindings
set (bindings/x.y/x.y.z/sources
) are not pulled
in unless required as a dependency for an end-user component. This
package "extends" meta-pkgs/gnome-base
.
meta-pkgs/gnome-devel
:
Installs all the tools required to build a GNOME component when
fetched from the CVS repository. These are required to let the
autogen.sh scripts work appropriately.
In all these packages, the DEPENDS
lines are
sorted in a way that eases updates: a package may depend on other
packages listed before it but not on any listed after it. It is very
important to keep this order to ease updates so... do not
change it to alphabetical sorting!
Almost all GNOME applications are written in C and use a common set of tools as their build system. Things get different with the new bindings to other languages (such as Python), but the following will give you a general idea on the minimum required tools:
Almost all GNOME applications use the GNU Autotools as their build system. As a general rule you will need to tell this to your package:
GNU_CONFIGURE=yes USE_LIBTOOL=yes USE_TOOLS+=gmake
If the package uses pkg-config to detect dependencies, add this tool to the list of required utilities:
USE_TOOLS+=pkg-config
Also use pkgtools/verifypc
at
the end of the build process to ensure that you did not miss to
specify any dependency in your package and that the version
requirements are all correct.
If the package uses intltool, be sure to add
intltool
to the USE_TOOLS
to handle dependencies and to force the package to use the latest
available version.
If the package uses gtk-doc (a documentation generation utility), do not add a dependency on it. The tool is rather big and the distfile should come with pregenerated documentation anyway; if it does not, it is a bug that you ought to report. For such packages you should disable gtk-doc (unless it is the default):
CONFIGURE_ARGS+=--disable-gtk-doc
The default location of installed HTML files
(share/gtk-doc/<package-name>
) is correct
and should not be changed unless the package insists on installing
them somewhere else. Otherwise programs as
devhelp will not be able to open them. You can
do that with an entry similar to:
CONFIGURE_ARGS+=--with-html-dir=${PREFIX}/share/gtk-doc/...
GNOME uses multiple shared directories and
files under the installation prefix to maintain databases. In this
context, shared means that those exact same directories and files are
used among several different packages, leading to conflicts in the
PLIST
. pkgsrc currently includes functionality to
handle the most common cases, so you have to forget about using
@unexec ${RMDIR}
lines in your file lists and
omitting shared files from them. If you find yourself doing those,
your package is most likely incorrect.
The following table lists the common situations that result in using shared directories or files. For each of them, the appropriate solution is given. After applying the solution be sure to regenerate the package's file list with make print-PLIST and ensure it is correct.
Table 22.1. PLIST handling for GNOME packages
If the package... | Then... |
---|---|
Installs OMF files under share/omf . |
See Section 18.6.9, “Packages installing scrollkeeper data files”. |
Installs icons under the
share/icons/hicolor hierarchy or updates
share/icons/hicolor/icon-theme.cache . |
See Section 18.6.18, “Packages installing hicolor theme icons”. |
Installs files under
share/mime/packages . |
See Section 18.6.13, “Packages installing extensions to the MIME database”. |
Installs .desktop files under
share/applications and these include MIME
information. |
See Section 18.6.19, “Packages installing desktop files”. |
When seeing GNOME as a whole, there are two kinds of updates:
Given that there is still a very long way for GNOME 3 (if it
ever appears), we consider a major update one that goes from a
2.X
version to a 2.Y
one,
where Y
is even and greater than
X
. These are hard to achieve because they
introduce lots of changes in the components' code and almost all
GNOME distfiles are updated to newer versions. Some of them can
even break API and ABI compatibility with the previous major
version series. As a result, the update needs to be done all at
once to minimize breakage.
A major update typically consists of around 80 package updates and the addition of some new ones.
We consider a minor update one that goes from a
2.A.X
version to a 2.A.Y
one where Y
is greater than
X
. These are easy to achieve because they do
not update all GNOME components, can be done in an incremental way
and do not break API nor ABI compatibility.
A minor update typically consists of around 50 package updates, although the numbers here may vary a lot.
In order to update the GNOME components in pkgsrc to a new stable release (either major or minor), the following steps should be followed:
Get a list of all the tarballs that form the new release by
using the following commands. These will leave the full list of the
components's distfiles into the list.txt
file:
%
echo ls "*.tar.bz2" | \ ftp -V ftp://ftp.gnome.org/pub/gnome/platform/x.y/x.y.z/sources/ | \ awk '{ print $9 }' >list.txt
%
echo ls "*.tar.bz2" | \ ftp -V ftp://ftp.gnome.org/pub/gnome/desktop/x.y/x.y.z/sources/ | \ awk '{ print $9 }' >>list.txt
Open each meta package's Makefile
and
bump their version to the release you are updating them to. The
three meta packages should be always consistent with versioning.
Obviously remove any PKGREVISION
s that might be
in them.
For each meta package, update all its
DEPENDS
lines to match the latest versions as
shown by the above commands. Do not list any
newer version (even if found in the FTP) because the meta packages
are supposed to list the exact versions that form a specific GNOME
release. Exceptions are permitted here if a newer version solves a
serious issue in the overall desktop experience; these typically
come in the form of a revision bump in pkgsrc, not in newer versions
from the developers.
Packages not listed in the list.txt
file
should be updated to the latest version available (if found in
pkgsrc). This is the case, for example, of the dependencies on the
GNU Autotools in the meta-pkgs/gnome-devel
meta package.
Generate a patch from the modified meta packages and extract the list of "new" lines. This will provide you an outline on what packages need to be updated in pkgsrc and in what order:
%
cvs diff -u gnome-devel gnome-base gnome | grep '^+D' >todo.txt
For major desktop updates it is recommended to zap all your installed packages and start over from scratch at this point.
Now comes the longest step by far: iterate over the contents
of todo.txt
and update the packages listed in
it in order. For major desktop updates none of these should be
committed until the entire set is completed because there are chances
of breaking not-yet-updated packages.
Once the packages are up to date and working, commit them to
the tree one by one with appropriate log messages. At the end,
commit the three meta package updates and all the corresponding
changes to the doc/CHANGES-<YEAR>
and
pkgsrc/doc/TODO
files.
GNOME is a very big component in pkgsrc which approaches 100 packages. Please, it is very important that you always, always, always feed back any portability fixes you do to a GNOME package to the mainstream developers (see Section 10.3.5, “Feedback to the author”). This is the only way to get their attention on portability issues and to ensure that future versions can be built out-of-the box on NetBSD. The less custom patches in pkgsrc, the easier further updates are. Those developers in charge of issuing major GNOME updates will be grateful if you do that.
The most common places to report bugs are the GNOME's Bugzilla and the freedesktop.org's Bugzilla. Not all components use these to track bugs, but most of them do. Do not be short on your reports: always provide detailed explanations of the current failure, how it can be improved to achieve maximum portability and, if at all possible, provide a patch against CVS head. The more verbose you are, the higher chances of your patch being accepted.
Also, please avoid using preprocessor magic to fix portability
issues. While the FreeBSD GNOME people are doing a great job in porting
GNOME to their operating system, the official GNOME sources are now
plagued by conditionals that check for __FreeBSD__
and similar macros. This hurts portability. Please see our patching
guidelines (Section 10.3.4, “Patching guidelines”) for more
details.
This part of the guide deals with everything from the infrastructure that is behind the interfaces described in the developer's guide. A casual package maintainer should not need anything from this part.
Table of Contents
Table of Contents
The pkgsrc infrastructure consists of many small Makefile fragments. Each such fragment needs a properly specified interface. This chapter explains how such an interface looks like.
Whenever a variable is defined in the pkgsrc infrastructure, the location and the way of definition provide much information about the intended use of that variable. Additionally, more documentation may be found in a header comment or in this pkgsrc guide.
A special file is
mk/defaults/mk.conf
, which lists all
variables that are intended to be user-defined. They are either
defined using the ?=
operator or they are
left undefined because defining them to anything would
effectively mean “yes”. All these variables may be
overridden by the pkgsrc user in the MAKECONF
file.
Outside this file, the following conventions apply:
Variables that are defined using the ?=
operator may be overridden by a package.
Variables that are defined using the =
operator may be used read-only at run-time.
Variables whose name starts with an underscore must not be accessed outside the pkgsrc infrastructure at all. They may change without further notice.
These conventions are currently not applied consistently to the complete pkgsrc infrastructure.
All variables that contain lists of things should default
to being empty. Two examples that do not follow this rule are
USE_LANGUAGES
and
DISTFILES
. These variables cannot simply be
modified using the +=
operator in package
Makefile
s (or other files included by
them), since there is no guarantee whether the variable is
already set or not, and what its value is. In the case of
DISTFILES
, the packages “know”
the default value and just define it as in the following
example.
DISTFILES= ${DISTNAME}${EXTRACT_SUFX} additional-files.tar.gz
Because of the selection of this default value, the same
value appears in many package Makefiles. Similarly for
USE_LANGUAGES
, but in this case the default
value (“c
”) is so short that it
doesn't stand out. Nevertheless it is mentioned in many
files.
Variable evaluation takes place either at load time or at runtime, depending on the context in which they occur. The contexts where variables are evaluated at load time are:
The right hand side of the :=
and !=
operators,
Make directives like .if
or
.for
,
Dependency lines.
A special exception are references to the iteration
variables of .for
loops, which are expanded
inline, no matter in which context they appear.
As the values of variables may change during load time,
care must be taken not to evaluate them by accident. Typical
examples for variables that should not be evaluated at load time
are DEPENDS
and
CONFIGURE_ARGS
. To make the effect more
clear, here is an example:
CONFIGURE_ARGS= # none CFLAGS= -O CONFIGURE_ARGS+= CFLAGS=${CFLAGS:Q} CONFIGURE_ARGS:= ${CONFIGURE_ARGS} CFLAGS+= -Wall
This code shows how the use of the :=
operator can quickly lead to unexpected results. The first
paragraph is fairly common code. The second paragraph evaluates
the CONFIGURE_ARGS
variable, which results in
CFLAGS=-O
. In the third paragraph, the
-Wall
is appended to the
CFLAGS
, but this addition will not appear in
CONFIGURE_ARGS
. In actual code, the three
paragraphs from above typically occur in completely unrelated
files.
There are many ways in which the definition and use of a
variable can be restricted in order to detect bugs and
violations of the (mostly unwritten) policies. See the
pkglint
developer's documentation for further
details.
Most of the .mk
files fall into one
of the following classes. Cases where a file falls into more
than one class should be avoided as it often leads to subtle
bugs.
In a traditional imperative programming language some of
the .mk
files could be described as
procedures. They take some input parameters and—after
inclusion—provide a result in output parameters. Since all
variables in Makefile
s have global scope
care must be taken not to use parameter names that have already
another meaning. For example, PKGNAME
is a
bad choice for a parameter name.
Procedures are completely evaluated at preprocessing time.
That is, when calling a procedure all input parameters must be
completely resolvable. For example,
CONFIGURE_ARGS
should never be an input
parameter since it is very likely that further text will be
added after calling the procedure, which would effectively apply
the procedure to only a part of the variable. Also, references
to other variables wit will be modified after calling the
procedure.
A procedure can declare its output parameters either as suitable for use in preprocessing directives or as only available at runtime. The latter alternative is for variables that contain references to other runtime variables.
Procedures shall be written such that it is possible to call the procedure more than once. That is, the file must not contain multiple-inclusion guards.
Examples for procedures are
mk/bsd.options.mk
and
mk/buildlink3/bsd.builtin.mk
. To express
that the parameters are evaluated at load time, they should be
assigned using the :=
operator, which should
be used only for this purpose.
Action files take some input parameters and may define runtime variables. They shall not define loadtime variables. There are action files that are included implicitly by the pkgsrc infrastructure, while other must be included explicitly.
An example for action files is
mk/subst.mk
.
Package Makefile
s usually consist of
a set of variable definitions, and include the file
../../mk/bsd.pkg.mk
in the very last line.
Before that, they may also include various other
*.mk
files if they need to query the
availability of certain features like the type of compiler or
the X11 implementation. Due to the heavy use of preprocessor
directives like .if
and
.for
, the order in which the files are loaded
matters.
This section describes at which point the various files are loaded and gives reasons for that order.
The very first action in bsd.prefs.mk
is to define some essential variables like
OPSYS
, OS_VERSION
and
MACHINE_ARCH
.
Then, the user settings are loaded from the file specified
in MAKECONF
. If the bmake command from pkgsrc
is used, MAKECONF
defaults to
.
With the native make(1) command on NetBSD, it defaults to
${prefix}
/etc/mk.conf/etc/mk.conf
. After that, those variables
that have not been overridden by the user are loaded from
mk/defaults/mk.conf
.
After the user settings, the system settings and platform settings are loaded, which may override the user settings.
Then, the tool definitions are loaded. The tool wrappers are not yet in effect. This only happens when building a package, so the proper variables must be used instead of the direct tool names.
As the last steps, some essential variables from the wrapper and the package system flavor are loaded, as well as the variables that have been cached in earlier phases of a package build.
First, bsd.prefs.mk
is loaded.
Then, the various *-vars.mk
files are
loaded, which fill default values for those variables that have
not been defined by the package. These variables may later
be used even in unrelated files.
Then, the file bsd.pkg.error.mk
provides the target error-check
that is added
as a special dependency to all other targets that use
DELAYED_ERROR_MSG
or
DELAYED_WARNING_MSG
.
Then, the package-specific hacks from
hacks.mk
are included.
Then, various other files follow. Most of them don't have any dependencies on what they need to have included before or after them, though some do.
The code to check PKG_FAIL_REASON
and
PKG_SKIP_REASON
is then executed, which
restricts the use of these variables to all the files that have
been included before. Appearances in later files will be
silently ignored.
Then, the files for the main targets are included, in the order of later execution, though the actual order should not matter.
At last, some more files are included that don't set any interesting variables but rather just define make targets to be executed.
Table of Contents
The pkgsrc infrastructure consists of a large codebase, and there are many corners where every little bit of a file is well thought out, making pkgsrc likely to fail as soon as anything is changed near those parts. To prevent most changes from breaking anything, a suite of regression tests should go along with every important part of the pkgsrc infrastructure. This chapter describes how regression tests work in pkgsrc and how you can add new tests.
You first need to install the pkgtools/pkg_regress
package, which
provides the pkg_regress command. Then you
can simply run that command, which will run all tests in the
regress
category.
Every directory in the regress
category that contains a file called spec
is considered a regression test. This file is a shell program
that is included by the pkg_regress command.
The following functions can be overridden to suit your
needs.
These functions do not take any parameters. They are all called in “set -e” mode, so you should be careful to check the exitcodes of any commands you run in the test.
do_setup()
This function prepares the environment for the test. By default it does nothing.
do_test()
This function runs the actual test. By default,
it calls TEST_MAKE
with the arguments
MAKEARGS_TEST
and writes its output including
error messages into the file
TEST_OUTFILE
.
check_result()
This function is run after the test and is typically used to compare the actual output from the one that is expected. It can make use of the various helper functions from the next section.
do_cleanup()
This function cleans everything up after the test has been run. By default it does nothing.
exit_status(expected)
This function compares the exitcode of the do_test() function with its first parameter. If they differ, the test will fail.
output_require(regex...)
This function checks for each of its parameters if the output from do_test() matches the extended regular expression. If it does not, the test will fail.
output_prohibit(regex...)
This function checks for each of its parameters if the output from do_test() does not match the extended regular expression. If any of the regular expressions matches, the test will fail.
Table of Contents
The pkgsrc system has already been ported to many operating systems, hardware architectures and compilers. This chapter explains the necessary steps to make pkgsrc even more portable.
To port pkgsrc to a new operating system (called
MyOS
in this example), you need to touch the
following files:
pkgtools/bootstrap-mk-files/files/mods/MyOS
.sys.mk
This file contains some basic definitions, for example the name of the C compiler.
mk/bsd.prefs.mk
Insert code that defines the variables
OPSYS
, OS_VERSION
,
LOWER_OS_VERSION
,
LOWER_VENDOR
,
MACHINE_ARCH
, OBJECT_FMT
,
APPEND_ELF
, and the other variables that
appear in this file.
mk/platform/MyOS
.mk
This file contains the platform-specific definitions that are used by pkgsrc. Start by copying one of the other files and edit it to your needs.
mk/platform/MyOS
.pkg.dist
This file contains a list of directories,
together with their permission bits and ownership. These
directories will be created automatically with every package
that does not explicitly set NO_MTREE
. There
have been some discussions about whether this file is needed at
all, but with no result.
mk/platform/MyOS
.x11.dist
Just copy one of the pre-existing x11.dist files
to your
.MyOS
.x11.dist
mk/tools/bootstrap.mk
On some operating systems, the tools that are provided with the base system are not good enough for pkgsrc. For example, there are many versions of sed(1) that have a narrow limit on the line length they can process. Therefore pkgsrc brings its own tools, which can be enabled here.
mk/tools/tools.MyOS
.mk
This file defines the paths to all the tools that are needed by one or the other package in pkgsrc, as well as by pkgsrc itself. Find out where these tools are on your platform and add them.
Now, you should be able to build some basic packages, like
lang/perl5
, shells/bash
.
Table of Contents
We checked to find a piece of software that wasn't in the packages collection, and picked GNU bison. Quite why someone would want to have bison when Berkeley yacc is already present in the tree is beyond us, but it's useful for the purposes of this exercise.
# $NetBSD$ # DISTNAME= bison-1.25 CATEGORIES= devel MASTER_SITES= ${MASTER_SITE_GNU} MAINTAINER= thorpej@NetBSD.org HOMEPAGE= http://www.gnu.org/software/bison/bison.html COMMENT= GNU yacc clone GNU_CONFIGURE= yes INFO_FILES= bison.info .include "../../mk/bsd.pkg.mk"
GNU version of yacc. Can make re-entrant parsers, and numerous other improvements. Why you would want this when Berkeley yacc(1) is part of the NetBSD source tree is beyond me.
The NetBSD package system comes with
pkgtools/pkglint
which helps to check the contents of these
files. After installation it is quite easy to use, just change to the
directory of the package you wish to examine and execute
pkglint:
$
pkglint
looks fine.
Depending on the supplied command line arguments (see pkglint(1)), more checks will be performed. Use e.g. pkglint -Call -Wall for a very thorough check.
Create the directory where the package lives, plus any auxiliary directories:
#
cd /usr/pkgsrc/lang
#
mkdir bison
#
cd bison
#
mkdir patches
Create Makefile
, DESCR
and
PLIST
(see Chapter 10, Package components - files, directories and contents)
then continue with fetching the distfile:
#
make fetch
>> bison-1.25.tar.gz doesn't seem to exist on this system. >> Attempting to fetch from ftp://prep.ai.mit.edu/pub/gnu//. Requesting ftp://prep.ai.mit.edu/pub/gnu//bison-1.25.tar.gz (via ftp://orpheus.amdahl.com:80/) ftp: Error retrieving file: 500 Internal error >> Attempting to fetch from ftp://wuarchive.wustl.edu/systems/gnu//. Requesting ftp://wuarchive.wustl.edu/systems/gnu//bison-1.25.tar.gz (via ftp://orpheus.amdahl.com:80/) ftp: Error retrieving file: 500 Internal error >> Attempting to fetch from ftp://ftp.freebsd.org/pub/FreeBSD/distfiles//. Requesting ftp://ftp.freebsd.org/pub/FreeBSD/distfiles//bison-1.25.tar.gz (via ftp://orpheus.amdahl.com:80/) Successfully retrieved file.
Generate the checksum of the distfile into
distinfo
:
#
make makedistinfo
Now compile:
#
make
>> Checksum OK for bison-1.25.tar.gz. ===> Extracting for bison-1.25 ===> Patching for bison-1.25 ===> Ignoring empty patch directory ===> Configuring for bison-1.25 creating cache ./config.cache checking for gcc... cc checking whether we are using GNU C... yes checking for a BSD compatible install... /usr/bin/install -c -o bin -g bin checking how to run the C preprocessor... cc -E checking for minix/config.h... no checking for POSIXized ISC... no checking whether cross-compiling... no checking for ANSI C header files... yes checking for string.h... yes checking for stdlib.h... yes checking for memory.h... yes checking for working const... yes checking for working alloca.h... no checking for alloca... yes checking for strerror... yes updating cache ./config.cache creating ./config.status creating Makefile ===> Building for bison-1.25 cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g LR0.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g allocate.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g closure.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g conflicts.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g derives.c cc -c -DXPFILE=\"/usr/pkg/share/bison.simple\" -DXPFILE1=\"/usr/pkg/share/bison.hairy\" -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -g ./files.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g getargs.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g gram.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g lalr.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g lex.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g main.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g nullable.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g output.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g print.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g reader.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g reduce.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g symtab.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g warshall.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g version.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g getopt.c cc -c -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_STDLIB_H=1 -DHAVE_MEMORY_H=1 -DHAVE_ALLOCA=1 -DHAVE_STRERROR=1 -I./../include -g getopt1.c cc -g -o bison LR0.o allocate.o closure.o conflicts.o derives.o files.o getargs.o gram.o lalr.o lex.o main.o nullable.o output.o print.o reader.o reduce.o symtab.o warshall.o version.o getopt.o getopt1.o ./files.c:240: warning: mktemp() possibly used unsafely, consider using mkstemp() rm -f bison.s1 sed -e "/^#line/ s|bison|/usr/pkg/share/bison|" < ./bison.simple > bison.s1
Everything seems OK, so install the files:
#
make install
>> Checksum OK for bison-1.25.tar.gz. ===> Installing for bison-1.25 sh ./mkinstalldirs /usr/pkg/bin /usr/pkg/share /usr/pkg/info /usr/pkg/man/man1 rm -f /usr/pkg/bin/bison cd /usr/pkg/share; rm -f bison.simple bison.hairy rm -f /usr/pkg/man/man1/bison.1 /usr/pkg/info/bison.info* install -c -o bin -g bin -m 555 bison /usr/pkg/bin/bison /usr/bin/install -c -o bin -g bin -m 644 bison.s1 /usr/pkg/share/bison.simple /usr/bin/install -c -o bin -g bin -m 644 ./bison.hairy /usr/pkg/share/bison.hairy cd .; for f in bison.info*; do /usr/bin/install -c -o bin -g bin -m 644 $f /usr/pkg/info/$f; done /usr/bin/install -c -o bin -g bin -m 644 ./bison.1 /usr/pkg/man/man1/bison.1 ===> Registering installation for bison-1.25
You can now use bison, and also - if you decide so - remove it with pkg_delete bison. Should you decide that you want a binary package, do this now:
#
make package
>> Checksum OK for bison-1.25.tar.gz. ===> Building package for bison-1.25 Creating package bison-1.25.tgz Registering depends:. Creating gzip'd tar ball in '/u/pkgsrc/lang/bison/bison-1.25.tgz'
Now that you don't need the source and object files any more, clean up:
#
make clean
===> Cleaning for bison-1.25
Table of Contents
#
make
===> Checking for vulnerabilities in figlet-2.2.1nb2 => figlet221.tar.gz doesn't seem to exist on this system. => Attempting to fetch figlet221.tar.gz from ftp://ftp.figlet.org/pub/figlet/program/unix/. => [172219 bytes] Connected to ftp.plig.net. 220 ftp.plig.org NcFTPd Server (licensed copy) ready. 331 Guest login ok, send your complete e-mail address as password. 230-You are user #5 of 500 simultaneous users allowed. 230- 230- ___ _ _ _ 230- | _| |_ ___ ___| |_|___ ___ ___ ___ 230- | _| _| . |_| . | | | . |_| . | _| . | 230- |_| |_| | _|_| _|_|_|_ |_|___|_| |_ | 230- |_| |_| |___| |___| 230- 230-** Welcome to ftp.plig.org ** 230- 230-Please note that all transfers from this FTP site are logged. If you 230-do not like this, please disconnect now. 230- 230-This arhive is available via 230- 230-HTTP: http://ftp.plig.org/ 230-FTP: ftp://ftp.plig.org/ (max 500 connections) 230-RSYNC: rsync://ftp.plig.org/ (max 30 connections) 230- 230-Please email comments, bug reports and requests for packages to be 230-mirrored to ftp-admin@plig.org. 230- 230- 230 Logged in anonymously. Remote system type is UNIX. Using binary mode to transfer files. 200 Type okay. 250 "/pub" is new cwd. 250-"/pub/figlet" is new cwd. 250- 250-Welcome to the figlet archive at ftp.figlet.org 250- 250- ftp://ftp.figlet.org/pub/figlet/ 250- 250-The official FIGlet web page is: 250- http://www.figlet.org/ 250- 250-If you have questions, please mailto:info@figlet.org. If you want to 250-contribute a font or something else, you can email us. 250 250 "/pub/figlet/program" is new cwd. 250 "/pub/figlet/program/unix" is new cwd. local: figlet221.tar.gz remote: figlet221.tar.gz 502 Unimplemented command. 227 Entering Passive Mode (195,40,6,41,246,104) 150 Data connection accepted from 84.128.86.72:65131; transfer starting for figlet221.tar.gz (172219 bytes). 38% |************** | 65800 64.16 KB/s 00:01 ETA 226 Transfer completed. 172219 bytes received in 00:02 (75.99 KB/s) 221 Goodbye. => Checksum OK for figlet221.tar.gz. ===> Extracting for figlet-2.2.1nb2 ===> Required installed package ccache-[0-9]*: ccache-2.3nb1 found ===> Patching for figlet-2.2.1nb2 ===> Applying pkgsrc patches for figlet-2.2.1nb2 ===> Overriding tools for figlet-2.2.1nb2 ===> Creating toolchain wrappers for figlet-2.2.1nb2 ===> Configuring for figlet-2.2.1nb2 ===> Building for figlet-2.2.1nb2 gcc -O2 -DDEFAULTFONTDIR=\"/usr/pkg/share/figlet\" -DDEFAULTFONTFILE=\"standard.flf\" figlet.c zipio.c crc.c inflate.c -o figlet chmod a+x figlet gcc -O2 -o chkfont chkfont.c => Unwrapping files-to-be-installed.#
#
make install
===> Checking for vulnerabilities in figlet-2.2.1nb2 ===> Installing for figlet-2.2.1nb2 install -d -o root -g wheel -m 755 /usr/pkg/bin install -d -o root -g wheel -m 755 /usr/pkg/man/man6 mkdir -p /usr/pkg/share/figlet cp figlet /usr/pkg/bin cp chkfont /usr/pkg/bin chmod 555 figlist showfigfonts cp figlist /usr/pkg/bin cp showfigfonts /usr/pkg/bin cp fonts/*.flf /usr/pkg/share/figlet cp fonts/*.flc /usr/pkg/share/figlet cp figlet.6 /usr/pkg/man/man6 ===> Registering installation for figlet-2.2.1nb2#
Table of Contents
As in other big projects, the directory layout of pkgsrc
is quite complex for newbies. This chapter explains where you
find things on the FTP server. The base directory on
ftp.NetBSD.org
is /pub/pkgsrc/
.
On other servers it may be different, but inside this directory,
everything should look the same, no matter on which server you
are. This directory contains some subdirectories, which are
explained below.
For those who only want to manage binary packages on systems other than NetBSD, we provide the package management tools in a separate, small tar file.
The directory distfiles
contains lots
of archive files from all pkgsrc packages, which are mirrored
here. The subdirectories are called after their package names
and are used when the distributed files have names that don't
explicitly contain a version number or are otherwise too generic
(for example release.tar.gz
).
This directory contains things that individual pkgsrc developers find worth publishing.
These directories contain binary packages. Those
directories that have a branch name
(200x
Qy
)
contain packages from that particular branch. The directory
packages
contains binary packages from
pkgsrc-current. (However, this does not necessarily mean that
the packages are still current anymore.)
Below the packages*
directories are
directories that distinguish the packages by operating system
and version, the next directory level specifies the hardware
architecture.
In each of the platform-specific directories, there is a
whole binary packages collection. It has a directory called
All
which contains all binary packages.
Besides that, there are various category directories that
contain symbolic links to the real binary packages.
These directories contain the “real” pkgsrc, that is the files that define how to create binary packages from source archives.
The directory pkgsrc
contains a
snapshot of the CVS repository, which is updated regularly. The
file pkgsrc.tar.gz
contains the same as the
directory, ready to be downloaded as a whole.
In the directories for the quarterly branches, there is an
additional file called
pkgsrc-200
,
which contains the state of pkgsrc when it was branched.x
Qy
.tar.gz
Table of Contents
This section contains information on editing the pkgsrc guide itself.
The pkgsrc guide's source code is stored in
pkgsrc/doc/guide/files
, and several files
are created from it:
pkgsrc/doc/pkgsrc.txt
pkgsrc/doc/pkgsrc.html
http://www.NetBSD.org/Documentation/pkgsrc/pkgsrc.pdf: The PDF version of the pkgsrc guide.
http://www.NetBSD.org/Documentation/pkgsrc/pkgsrc.ps: PostScript version of the pkgsrc guide.
The procedure to edit the pkgsrc guide is:
Make sure you have the packages needed to
regenerate the pkgsrc guide (and other XML-based NetBSD
documentation) installed. These are meta-pkgs/netbsd-doc
for creating the
ASCII and HTML versions, and meta-pkgs/netbsd-doc-print
for the
PostScript and PDF versions. You will need both packages
installed, to make sure documentation is consistent across all
formats.
Run cd doc/guide to get to the right directory. All further steps will take place here.
Edit the XML file(s) in
files/
.
Run bmake to check the pkgsrc
guide for valid XML and to build the final output files. If you
get any errors at this stage, you can just edit the files, as
there are only symbolic links in the working directory, pointing
to the files in files/
.
(cd files && cvs commit)
Run bmake clean && bmake to regenerate the output files with the proper RCS Ids.
Run bmake regen to install and
commit the files in both pkgsrc/doc
and
htdocs
.
If you have added, removed or renamed some chapters, you need to synchronize them using cvs add or cvs delete in the htdocs directory.