doc/en_US.ISO8859-1/articles/committers-guide/article.xml
2013-08-03 14:10:45 +00:00

4658 lines
171 KiB
XML

<?xml version="1.0" encoding="iso-8859-1"?>
<!DOCTYPE article PUBLIC "-//FreeBSD//DTD DocBook XML V4.5-Based Extension//EN"
"../../../share/xml/freebsd45.dtd" [
<!ENTITY ga "Google Analytics">
]>
<article lang='en'>
<articleinfo>
<title>Committer's Guide</title>
<corpauthor>The &os; Documentation Project</corpauthor>
<copyright>
<year>1999</year>
<year>2000</year>
<year>2001</year>
<year>2002</year>
<year>2003</year>
<year>2004</year>
<year>2005</year>
<year>2006</year>
<year>2007</year>
<year>2008</year>
<year>2009</year>
<year>2010</year>
<year>2011</year>
<year>2012</year>
<year>2013</year>
<holder>The &os; Documentation Project</holder>
</copyright>
<legalnotice id="trademarks" role="trademarks">
&tm-attrib.freebsd;
&tm-attrib.coverity;
&tm-attrib.ibm;
&tm-attrib.intel;
&tm-attrib.sparc;
&tm-attrib.general;
</legalnotice>
<pubdate>$FreeBSD$</pubdate>
<releaseinfo>$FreeBSD$</releaseinfo>
<abstract>
<para>This document provides information for the &os;
committer community. All new committers should read this
document before they start, and existing committers are
strongly encouraged to review it from time to time.</para>
<para>Almost all &os; developers have commit rights to one or
more repositories. However, a few developers do not, and some
of the information here applies to them as well. (For
instance, some people only have rights to work with the
Problem Report database). Please see <xref
linkend="non-committers"/> for more information.</para>
<para>This document may also be of interest to members of the
&os; community who want to learn more about how the project
works.</para>
</abstract>
</articleinfo>
<sect1 id="admin">
<title>Administrative Details</title>
<informaltable frame="none" orient="port" pgwide="1">
<tgroup cols="2">
<colspec colwidth="20*"/>
<colspec colwidth="80*"/>
<tbody>
<row>
<entry><emphasis>Login Methods</emphasis></entry>
<entry>&man.ssh.1;, protocol 2 only</entry>
</row>
<row>
<entry><emphasis>Main Shell Host</emphasis></entry>
<entry><hostid
role="fqdn">freefall.FreeBSD.org</hostid></entry>
</row>
<row>
<entry><emphasis><literal>src/</literal> Subversion
Root</emphasis></entry>
<entry><literal>svn+ssh://</literal><hostid
role="fqdn">svn.FreeBSD.org</hostid><filename>/base</filename>
(see also <xref
linkend="svn-getting-started-base-layout"/>).</entry>
</row>
<row>
<entry><emphasis><literal>doc/</literal> Subversion
Root</emphasis></entry>
<entry><literal>svn+ssh://</literal><hostid
role="fqdn">svn.FreeBSD.org</hostid><filename>/doc</filename>
(see also <xref
linkend="svn-getting-started-doc-layout"/>).</entry>
</row>
<row>
<entry><emphasis><literal>ports/</literal> Subversion
Root</emphasis></entry>
<entry><literal>svn+ssh://</literal><hostid
role="fqdn">svn.FreeBSD.org</hostid><filename>/ports</filename>
(see also <xref
linkend="svn-getting-started-ports-layout"/>).</entry>
</row>
<row>
<entry><emphasis>Internal Mailing Lists</emphasis></entry>
<entry>developers (technically called all-developers),
doc-developers, doc-committers, ports-developers,
ports-committers, src-developers, src-committers. (Each
project repository has its own -developers and
-committers mailing lists. Archives for these lists may
be found in files
<filename>/home/mail/<replaceable>repository-name</replaceable>-developers-archive</filename>
and
<filename>/home/mail/<replaceable>repository-name</replaceable>-committers-archive</filename>
on the <hostid role="domainname">FreeBSD.org</hostid>
cluster.)</entry>
</row>
<row>
<entry><emphasis>Core Team monthly
reports</emphasis></entry>
<entry><filename>/home/core/public/monthly-reports</filename>
on the <hostid role="domainname">FreeBSD.org</hostid>
cluster.</entry>
</row>
<row>
<entry><emphasis>Ports Management Team monthly
reports</emphasis></entry>
<entry><filename>/home/portmgr/public/monthly-reports</filename>
on the <hostid role="domainname">FreeBSD.org</hostid>
cluster.</entry>
</row>
<row>
<entry><emphasis>Noteworthy <literal>src/</literal> SVN
Branches</emphasis></entry>
<entry>
<literal>stable/8</literal> (8.X-STABLE),
<literal>stable/9</literal> (9.X-STABLE),
<literal>head</literal> (-CURRENT)</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>&man.ssh.1; is required to connect to the project hosts.
For more information, see <xref linkend="ssh.guide"/>.</para>
<para>Useful links:</para>
<itemizedlist>
<listitem>
<para><ulink url="&url.base;/internal/">&os;
Project Internal Pages</ulink></para>
</listitem>
<listitem>
<para><ulink
url="&url.base;/internal/machines.html">&os; Project
Hosts</ulink></para>
</listitem>
<listitem>
<para><ulink
url="&url.base;/administration.html">&os; Project
Administrative Groups</ulink></para>
</listitem>
</itemizedlist>
</sect1>
<sect1 id="committer.types">
<title>Commit Bit Types</title>
<para>The &os; repository has a number of components which,
when combined, support the basic operating system source,
documentation, third party application ports infrastructure, and
various maintained utilities. When &os; commit bits are
allocated, the areas of the tree where the bit may be used are
specified. Generally, the areas associated with a bit reflect
who authorized the allocation of the commit bit. Additional
areas of authority may be added at a later date: when this
occurs, the committer should follow normal commit bit allocation
procedures for that area of the tree, seeking approval from the
appropriate entity and possibly getting a mentor for that area
for some period of time.</para>
<informaltable frame="none" pgwide="1">
<tgroup cols="3">
<tbody>
<row>
<entry><emphasis>Committer Type</emphasis></entry>
<entry><emphasis>Responsible</emphasis></entry>
<entry><emphasis>Tree Components</emphasis></entry>
</row>
<row>
<entry>src</entry>
<entry>core@</entry>
<entry>src/, doc/ subject to appropriate review</entry>
</row>
<row>
<entry>doc</entry>
<entry>doceng@</entry>
<entry>doc/, src/ documentation</entry>
</row>
<row>
<entry>ports</entry>
<entry>portmgr@</entry>
<entry>ports/</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>Commit bits allocated prior to the development of the notion
of areas of authority may be appropriate for use in many parts
of the tree. However, common sense dictates that a committer
who has not previously worked in an area of the tree seek review
prior to committing, seek approval from the appropriate
responsible party, and/or work with a mentor. Since the rules
regarding code maintenance differ by area of the tree, this is
as much for the benefit of the committer working in an area of
less familiarity as it is for others working on the tree.</para>
<para>Committers are encouraged to seek review for their work as
part of the normal development process, regardless of the area
of the tree where the work is occurring.</para>
<sect2>
<title>Policy for <filename>doc/</filename> Committer Activity
in <filename>src/</filename></title>
<itemizedlist>
<listitem>
<para>doc committers may commit documentation changes to src
files, such as man pages, READMEs, fortune databases,
calendar files, and comment fixes without approval from a
src committer, subject to the normal care and tending of
commits.</para>
</listitem>
<listitem>
<para>doc committers may commit minor src changes and fixes,
such as build fixes, small features, etc, with an
"Approved by" from a src committer.</para>
</listitem>
<listitem>
<para>doc committers may seek an upgrade to a src commit bit
by acquiring a mentor, who will propose the doc committer
to core. When approved, they will be added to 'access'
and the normal mentoring period will ensue, which will
involve a continuing of <quote>Approved by</quote> for
some period.</para>
</listitem>
<listitem>
<para>"Approved by" is only acceptable from non-mentored src
committers -- mentored committers can provide a "Reviewed
by" but not an "Approved by".</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="subversion-primer">
<title>Subversion Primer</title>
<para>It is assumed that you are already familiar with the basic
operation of the version control systems in use. Traditionally
this was CVS. Subversion is used for the <literal>src</literal>
tree as of May 2008, the <literal>doc/www</literal> tree as of
May 2012 and the <literal>ports</literal> tree as of July
2012.</para>
<para><ulink url="http://wiki.freebsd.org/SubversionMissing">There
is a list of things missing in Subversion when compared to
CVS</ulink>. The notes at <ulink
url="http://people.freebsd.org/~peter/svn_notes.txt"></ulink>
might also be useful.</para>
<sect2 id="svn-intro">
<title>Introduction</title>
<para>The &os; source repository switched from
<acronym>CVS</acronym> to Subversion on May 31st, 2008. The
first real <acronym>SVN</acronym> commit is
<emphasis>r179447</emphasis>.</para>
<para>The &os; <literal>doc/www</literal> repository switched
from <acronym>CVS</acronym> to Subversion on May 19th, 2012.
The first real <acronym>SVN</acronym> commit is
<emphasis>r38821</emphasis>.</para>
<note>
<para>Part of the <literal>doc/www</literal>
<acronym>CVS</acronym> to <acronym>SVN</acronym> conversion
included an infrastructural change to the build process.
The most notable change is the location of the
&os;&nbsp;website <literal>www</literal> tree, which has
been moved from
<literal>www/<replaceable>lang</replaceable>/</literal> to
<literal>head/<replaceable>lang</replaceable>/htdocs/</literal>.</para>
</note>
<para>The &os; <literal>ports</literal> repository switched
from <acronym>CVS</acronym> to Subversion on July 14th, 2012.
The first real <acronym>SVN</acronym> commit is
<emphasis>r300894</emphasis>.</para>
<para>There are mechanisms in place to automatically merge
changes back from the Subversion <literal>src</literal>
repository to the <acronym>CVS</acronym> repository for
some &os; branches (<literal>releng/6</literal> through
<literal>releng/9</literal>), however this is purely to
support pre-existing end-user installs and should not be
relied upon, recommended or advertised. Future branches
will not be exported to CVS at all. The
<literal>ports</literal> repository was exported to CVS
for a period of time to aid end user migration, but as of
28th February 2013 is no longer exported.</para>
<para>Subversion is not that different from
<acronym>CVS</acronym> when it comes to daily use, but there
are differences. Subversion has a number of features that
should make developers' lives easier. The most important
advantage to Subversion (and the reason why &os; switched) is
that it handles branches and merging much better than CVS
does. Some of the principal differences are:</para>
<itemizedlist>
<listitem>
<para>Commits are atomic.</para>
</listitem>
<listitem>
<para>Revision numbers apply across the repository&mdash;all
files that were modified in the same commit have the same
revision number.</para>
</listitem>
<listitem>
<para>Branching and tagging are namespace operations.</para>
</listitem>
<listitem>
<para>Directories are versioned.</para>
</listitem>
<listitem>
<para>Files and directories can have arbitrary, versioned
metadata attached to them.</para>
</listitem>
<listitem>
<para>Files and directories can be copied, with full history
tracking.</para>
</listitem>
<listitem>
<para>No more contortions due to <acronym>CVS</acronym>
weakness such as applying &man.patch.1; files at compile
time in order to avoid touching vendor branch code.</para>
</listitem>
<listitem>
<para>No more repo-copies.</para>
</listitem>
</itemizedlist>
<para>Subversion can be installed from the &os; Ports
Collection by issuing these commands:</para>
<screen>&prompt.root; <userinput>cd /usr/ports/devel/subversion</userinput>
&prompt.root; <userinput>make clean install</userinput></screen>
</sect2>
<sect2 id="svn-getting-started">
<title>Getting Started</title>
<para>There are a few ways to obtain a working copy of the tree
from Subversion. This section will explain them.</para>
<sect3 id="svn-getting-started-direct-checkout">
<title>Direct Checkout</title>
<para>The first is to check out directly from the main
repository. For the <literal>src</literal> tree,
use:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/base/head /usr/src</userinput></screen>
<para>For the <literal>doc</literal> tree, use:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/doc/head /usr/doc</userinput></screen>
<para>For the <literal>ports</literal> tree, use:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/ports/head /usr/ports</userinput></screen>
<note>
<para>Though the remaining examples in this document are
written with the workflow of working with the
<literal>src</literal> tree in mind, the underlying
concepts are the same for working with the
<literal>doc</literal> and the <literal>ports</literal>
tree.
Ports related Subversion operations are listed in
<xref linkend="ports"/>.</para>
</note>
<para>The above command will check out a
<literal>CURRENT</literal> source tree as <filename
class="directory"><replaceable>/usr/src/</replaceable></filename>,
which can be any target directory on the local filesystem.
Omitting the final argument of that command causes the
working copy, in this case, to be named <quote>head</quote>,
but that can be renamed safely.</para>
<para><literal>svn+ssh</literal> means the
<acronym>SVN</acronym> protocol tunnelled over
<acronym>SSH</acronym>. The name of the server is
<literal>svn.freebsd.org</literal>, <literal>base</literal>
is the path to the repository, and <literal>head</literal>
is the subdirectory within the repository.</para>
<para>If your &os; login name is different from your login
name on your local machine, you must either include it in
the <acronym>URL</acronym> (for example
<literal>svn+ssh://jarjar@svn.freebsd.org/base/head</literal>),
or add an entry to your <filename>~/.ssh/config</filename>
in the form:</para>
<programlisting>Host svn.freebsd.org
User jarjar</programlisting>
<para>This is the simplest method, but it's hard to tell just
yet how much load it will place on the repository.
Subversion is much faster than <acronym>CVS</acronym>,
however.</para>
<note>
<para>The <command>svn diff</command> does not require
access to the server as <acronym>SVN</acronym> stores a
reference copy of every file in the working copy. This,
however, means that Subversion working copies are very
large in size.</para>
</note>
</sect3>
<sect3 id="svn-getting-started-checkout-from-a-mirror">
<title>Checkout from a Mirror</title>
<para>Check out a working copy from a mirror by
substituting the mirror's <acronym>URL</acronym> for
<literal>svn+ssh://svn.freebsd.org/base</literal>. This can
be an official mirror or a mirror maintained by
using <command>svnsync</command>.</para>
<para>There is a serious disadvantage to this method: every
time something is to be committed, a
<command>svn relocate</command> to the master repository has
to be done, remembering to <command>svn relocate</command>
back to the mirror after the commit. Also, since
<command>svn relocate</command> only works between
repositories that have the same UUID, some hacking of the
local repository's UUID has to occur before it is possible
to start using it.</para>
<para>Unlike with <acronym>CVS</acronym>,
the hassle of a local
<command>svnsync</command> mirror probably is not worth it
unless the network connectivity situation or other factors
demand it. If it is needed, see the end of this chapter for
information on how to set one up.</para>
</sect3>
<sect3 id="svn-getting-started-base-layout">
<title><literal>RELENG_*</literal> Branches and General
Layout</title>
<para>In <literal>svn+ssh://svn.freebsd.org/base</literal>,
<emphasis>base</emphasis> refers to the source tree.
Similarly, <emphasis>ports</emphasis> refers to the ports
tree, and so on. These are separate repositories with their
own change number sequences, access controls and commit
mail.</para>
<para>For the base repository, HEAD refers to the -CURRENT
tree. For example, <filename>head/bin/ls</filename> is what
would go into <filename>/usr/src/bin/ls</filename> in a
release. Some key locations are:</para>
<itemizedlist>
<listitem>
<para><emphasis>/head/</emphasis> which corresponds to
<literal>HEAD</literal>, also known as
<literal>-CURRENT</literal>.</para>
</listitem>
<listitem>
<para><emphasis>/stable/<replaceable>n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n</replaceable></literal>.</para>
</listitem>
<listitem>
<para><emphasis>/releng/<replaceable>n.n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n_n</replaceable></literal>.</para>
</listitem>
<listitem>
<para><emphasis>/release/<replaceable>n.n.n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n_n_n</replaceable>_RELEASE</literal>.</para>
</listitem>
<listitem>
<para><emphasis>/vendor*</emphasis> is the vendor branch
import work area. This directory itself does not
contain branches, however its subdirectories do. This
contrasts with the <emphasis>stable</emphasis>,
<emphasis>releng</emphasis> and
<emphasis>release</emphasis> directories.</para>
</listitem>
<listitem>
<para><emphasis>/projects</emphasis> and
<emphasis>/user</emphasis> feature a branch work area,
like in Perforce. As above, the
<emphasis>/user</emphasis> directory does not contain
branches itself.</para>
</listitem>
</itemizedlist>
</sect3>
<sect3 id="svn-getting-started-doc-layout">
<title>&os; Documentation Project Branches and
Layout</title>
<para>In <literal>svn+ssh://svn.freebsd.org/doc</literal>,
<emphasis>doc</emphasis> refers to the repository root of
the source tree.</para>
<para>In general, most &os; Documentation Project work will be
done within the <filename>head/</filename> branch of the
documentation source tree.</para>
<para>&os; documentation is written and/or translated to
various languages, each in a separate
directory in the <filename>head/</filename>
branch.</para>
<para>Each translation set contains several subdirectories for
the various parts of the &os; Documentation Project. A few
noteworthy directories are:</para>
<itemizedlist>
<listitem>
<para><emphasis>/articles/</emphasis> contains the source
code for articles written by various &os;
contributors.</para>
</listitem>
<listitem>
<para><emphasis>/books/</emphasis> contains the source
code for the different books, such as the
&os;&nbsp;Handbook.</para>
</listitem>
<listitem>
<para><emphasis>/htdocs/</emphasis> contains the source
code for the &os;&nbsp;website.</para>
</listitem>
</itemizedlist>
</sect3>
<sect3 id="svn-getting-started-ports-layout">
<title>&os; Ports Tree Branches and Layout</title>
<para>In <literal>svn+ssh://svn.freebsd.org/ports</literal>,
<emphasis>ports</emphasis> refers to the repository root of the
ports tree.</para>
<para>In general, most &os; port work will be done within
the <filename>head/</filename> branch of the ports tree
which is the actual ports tree used to install software.
Some other key locations are:</para>
<itemizedlist>
<listitem>
<para><emphasis>/branches/RELENG_<replaceable>n_n_n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n_n_n</replaceable></literal>
is used to merge back security updates in preparation
for a release.</para>
</listitem>
<listitem>
<para><emphasis>/tags/RELEASE_<replaceable>n_n_n</replaceable></emphasis>
which corresponds to
<literal>RELEASE_<replaceable>n_n_n</replaceable></literal>
represents a release tag of the ports tree.</para>
</listitem>
<listitem>
<para><emphasis>/tags/RELEASE_<replaceable>n</replaceable>_EOL</emphasis>
represents the end of life tag of a specific &os;
branch.</para>
</listitem>
</itemizedlist>
</sect3>
</sect2>
<sect2 id="svn-daily-use">
<title>Daily Use</title>
<para>This section will explain how to perform common day-to-day
operations with Subversion.</para>
<sect3 id="svn-daily-use-help">
<title>Help</title>
<para><acronym>SVN</acronym> has built in help documentation.
It can be accessed by typing the following command:</para>
<screen>&prompt.user; <userinput>svn help</userinput></screen>
<para>Additional information can be found in the
<ulink url="http://svnbook.red-bean.com/">Subversion
Book</ulink>.</para>
</sect3>
<sect3 id="svn-daily-use-checkout">
<title>Checkout</title>
<para>As seen earlier, to check out the &os; head
branch:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/base/head /usr/src</userinput></screen>
<para>At some point, more than just <literal>HEAD</literal>
will probably be useful, for instance when merging changes
to stable/7. Therefore, it may be useful to have a partial
checkout of the complete tree (a full checkout would be very
painful).</para>
<para>To do this, first check out the root of the
repository:</para>
<screen>&prompt.user; <userinput>svn checkout --depth=immediates svn+ssh://svn.freebsd.org/base</userinput></screen>
<para>This will give <literal>base</literal> with all the
files it contains (at the time of writing, just
<filename>ROADMAP.txt</filename>) and empty subdirectories
for <literal>head</literal>, <literal>stable</literal>,
<literal>vendor</literal> and so on.</para>
<para>Expanding the working copy is possible. Just change the
depth of the various subdirectories:</para>
<screen>&prompt.user; <userinput>svn up --set-depth=infinity base/head</userinput>
&prompt.user; <userinput>svn up --set-depth=immediates base/release base/releng base/stable</userinput></screen>
<para>The above command will pull down a full copy of
<literal>head</literal>, plus empty copies of every
<literal>release</literal> tag, every
<literal>releng</literal> branch, and every
<literal>stable</literal> branch.</para>
<para>If at a later date merging to
<literal>7-STABLE</literal> is required, expand the working
copy:</para>
<screen>&prompt.user; <userinput>svn up --set-depth=infinity base/stable/7</userinput></screen>
<para>Subtrees do not have to be expanded completely. For
instance, expanding only <literal>stable/7/sys</literal> and
then later expand the rest of
<literal>stable/7</literal>:</para>
<screen>&prompt.user; <userinput>svn up --set-depth=infinity base/stable/7/sys</userinput>
&prompt.user; <userinput>svn up --set-depth=infinity base/stable/7</userinput></screen>
<para>Updating the tree with <command>svn update</command>
will only update what was previously asked for (in this
case, <literal>head</literal> and
<literal>stable/7</literal>; it will not pull down the whole
tree.</para>
<note>
<para>Decreasing the depth of a working copy is not
possible.</para>
</note>
</sect3>
<sect3 id="svn-daily-use-anonymous-checkout">
<title>Anonymous Checkout</title>
<para>It is possible to anonymously check out the &os;
repository with Subversion. This will give access to a
read-only tree that can be updated, but not committed back
to the main repository. To do this, use the following command:</para>
<screen>&prompt.user; <userinput>svn co <replaceable>https://svn0.us-west.FreeBSD.org</replaceable>/base/head /usr/src</userinput></screen>
<para>Select the closest mirror and verify the mirror server
certificate from the list of <ulink
url="&url.books.handbook;/svn-mirrors.html">Subversion
mirror sites</ulink>.</para>
</sect3>
<sect3 id="svn-daily-use-updating-the-tree">
<title>Updating the Tree</title>
<para>To update a working copy to either the latest revision,
or a specific revision:</para>
<screen>&prompt.user; <userinput>svn update</userinput>
&prompt.user; <userinput>svn update -<replaceable>r12345</replaceable></userinput></screen>
</sect3>
<sect3 id="svn-daily-use-status">
<title>Status</title>
<para>To view the local changes that have been made to the
working copy:</para>
<screen>&prompt.user; <userinput>svn status</userinput></screen>
<para>To show local changes and files that are out-of-date
do:</para>
<screen>&prompt.user; <userinput>svn status --show-updates</userinput></screen>
</sect3>
<sect3 id="svn-daily-use-editing-and-committing">
<title>Editing and Committing</title>
<para>Unlike Perforce, <acronym>SVN</acronym> does not need to
be told in advance about file editing.</para>
<para><command>svn commit</command> works like the equivalent
<acronym>CVS</acronym> command. To commit all changes in
the current directory and all subdirectories:</para>
<screen>&prompt.user; <userinput>svn commit</userinput></screen>
<para>To commit all changes in, for example, <filename
class="directory"><replaceable>lib/libfetch/</replaceable></filename>
and <filename
class="directory"><replaceable>usr/bin/fetch/</replaceable></filename>
in a single operation:</para>
<screen>&prompt.user; <userinput>svn commit <replaceable>lib/libfetch</replaceable> <replaceable>usr/bin/fetch</replaceable></userinput></screen>
<para>There is also a commit wrapper for the ports tree to
handle the properties and sanity checking your
changes:</para>
<screen>&prompt.user; <userinput>/usr/ports/Tools/scripts/psvn commit
</userinput></screen>
</sect3>
<sect3 id="svn-daily-use-adding-and-removing">
<title>Adding and Removing Files</title>
<note>
<para>Before adding files, get a copy of <ulink
url="http://people.freebsd.org/~peter/auto-props.txt">auto-props.txt</ulink>
(there is also a <ulink
url="http://people.freebsd.org/~beat/cvs2svn/auto-props.txt">
ports tree specific version</ulink>)
and add it to <filename>~/.subversion/config</filename>
according to the instructions in the file. If you added
something before reading this, use
<command>svn rm --keep-local</command> for just added
files, fix your config file and re-add them again. The
initial config file is created when you first run a svn
command, even something as simple as
<command>svn help</command>.</para>
</note>
<para>Files are added to a
<acronym>SVN</acronym> repository with <command>svn
add</command>. To add a file named
<emphasis>foo</emphasis>, edit it, then:</para>
<screen>&prompt.user; <userinput>svn add <replaceable>foo</replaceable></userinput></screen>
<note>
<para>Most new source files should include a
<literal>&dollar;&os;&dollar;</literal> string near the start of the
file. On commit, <command>svn</command> will expand
the <literal>&dollar;&os;&dollar;</literal> string,
adding the file path, revision number, date and time of
commit, and the username of the committer. Files which
cannot be modified may be committed without the
<literal>&dollar;&os;&dollar;</literal> string.</para>
</note>
<para>Files can be removed with <command>svn
remove</command>:</para>
<screen>&prompt.user; <userinput>svn remove <replaceable>foo</replaceable></userinput></screen>
<para>Subversion does not require deleting the file before
using <command>svn rm</command>, and indeed complains if
that happens.</para>
<para>It is possible to add directories with
<command>svn add</command>:</para>
<screen>&prompt.user; <userinput>mkdir <replaceable>bar</replaceable></userinput>
&prompt.user; <userinput>svn add <replaceable>bar</replaceable></userinput></screen>
<para>Although <command>svn mkdir</command> makes this easier
by combining the creation of the directory and the adding of
it:</para>
<screen>&prompt.user; <userinput>svn mkdir <replaceable>bar</replaceable></userinput></screen>
<para>Like files, directories are removed with
<command>svn rm</command>. There is no separate command
specifically for removing directories.</para>
<screen>&prompt.user; <userinput>svn rm <replaceable>bar</replaceable></userinput></screen>
</sect3>
<sect3 id="svn-daily-use-copying-and-moving">
<title>Copying and Moving Files</title>
<para>This command creates a copy of
<filename>foo.c</filename> named <filename>bar.c</filename>,
with the new file also under version control:</para>
<screen>&prompt.user; <userinput>svn copy <replaceable>foo.c</replaceable> <replaceable>bar.c</replaceable></userinput></screen>
<para>The example above is equivalent to:</para>
<screen>&prompt.user; <userinput>cp foo.c bar.c</userinput>
&prompt.user; <userinput>svn add bar.c</userinput></screen>
<para>To move and rename a file:</para>
<screen>&prompt.user; <userinput>svn move <replaceable>foo.c</replaceable> <replaceable>bar.c</replaceable></userinput></screen>
</sect3>
<sect3 id="svn-daily-use-log-and-annotate">
<title>Log and Annotate</title>
<para><command>svn log</command> shows revisions and commit
messages, most recent first, for files or directories. When
used on a directory, all revisions that affected the
directory and files within that directory are shown.</para>
<para><command>svn annotate</command>, or equally <command>svn
praise</command> or <command>svn blame</command>, shows
the most recent revision number and who committed that
revision for each line of a file.</para>
</sect3>
<sect3 id="svn-daily-use-diffs">
<title>Diffs</title>
<para><command>svn diff</command> displays changes to the
working copy. Diffs generated by <acronym>SVN</acronym> are
unified and include new files by default in the diff
output.</para>
<para><command>svn diff</command> can show the changes between
two revisions of the same file:</para>
<screen>&prompt.user; <userinput>svn diff -r179453:179454 ROADMAP.txt</userinput></screen>
<para>It can also show all changes for a specific changeset.
The following will show what changes were made to the
current directory and all subdirectories in changeset
179454:</para>
<screen>&prompt.user; <userinput>svn diff -c179454 .</userinput></screen>
</sect3>
<sect3 id="svn-daily-use-reverting">
<title>Reverting</title>
<para>Local changes (including additions and deletions) can be
reverted using <command>svn revert</command>. It does not
update out-of-date files, but just replaces them with
pristine copies of the original version.</para>
</sect3>
<sect3 id="svn-daily-use-conflicts">
<title>Conflicts</title>
<para>If an <command>svn update</command> resulted in a merge
conflict, Subversion will remember which files have
conflicts and refuse to commit any changes to those files
until explicitly told that the conflicts have been resolved.
The simple, not yet deprecated procedure is the
following:</para>
<screen>&prompt.user; <userinput>svn resolved <replaceable>foo</replaceable></userinput></screen>
<para>However, the preferred procedure is:</para>
<screen>&prompt.user; <userinput>svn resolve --accept=working <replaceable>foo</replaceable></userinput></screen>
<para>The two examples are equivalent. Possible values for
<literal>--accept</literal> are:</para>
<itemizedlist>
<listitem>
<para><literal>working</literal>: use the version in your
working directory (which one presumes has been edited to
resolve the conflicts).</para>
</listitem>
<listitem>
<para><literal>base</literal>: use a pristine copy of the
version you had before <command>svn update</command>,
discarding your own changes, the conflicting changes,
and possibly other intervening changes as well.</para>
</listitem>
<listitem>
<para><literal>mine-full</literal>: use what you had
before <command>svn update</command>, including your own
changes, but discarding the conflicting changes, and
possibly other intervening changes as well.</para>
</listitem>
<listitem>
<para><literal>theirs-full</literal>: use the version that
was retrieved when you did
<command>svn update</command>, discarding your own
changes.</para>
</listitem>
</itemizedlist>
</sect3>
</sect2>
<sect2>
<title>Advanced Use</title>
<sect3 id="svn-advanced-use-sparse-checkouts">
<title>Sparse Checkouts</title>
<para><acronym>SVN</acronym> allows
<emphasis>sparse</emphasis>, or partial checkouts of a
directory by adding <option>--depth</option> to a
<command>svn checkout</command>.</para>
<para>Valid arguments to <option>--depth</option>
are:</para>
<itemizedlist>
<listitem>
<para><literal>empty</literal>: the directory itself
without any of its contents.</para>
</listitem>
<listitem>
<para><literal>files</literal>: the directory and any
files it contains.</para>
</listitem>
<listitem>
<para><literal>immediates</literal>: the directory and any
files and directories it contains, but none of the
subdirectories' contents.</para>
</listitem>
<listitem>
<para><literal>infinity</literal>: anything.</para>
</listitem>
</itemizedlist>
<para>The <literal>--depth</literal> option applies to many
other commands, including <command>svn commit</command>,
<command>svn revert</command>, and <command>svn
diff</command>.</para>
<para>Since <literal>--depth</literal> is sticky, there is a
<literal>--set-depth</literal> option for <command>svn
update</command> that will change the selected depth.
Thus, given the working copy produced by the previous
example:</para>
<screen>&prompt.user; <userinput>cd <replaceable>~/freebsd</replaceable></userinput>
&prompt.user; <userinput>svn update --set-depth=immediates .</userinput></screen>
<para>The above command will populate the working copy in
<replaceable>~/freebsd</replaceable> with
<filename>ROADMAP.txt</filename> and empty subdirectories,
and nothing will happen when <command>svn update</command>
is executed on the subdirectories. However, the following
command will set the depth for
<replaceable>head</replaceable> (in this case) to infinity,
and fully populate it:</para>
<screen>&prompt.user; <userinput>svn update --set-depth=infinity <replaceable>head</replaceable></userinput></screen>
</sect3>
<sect3 id="svn-advanced-use-direct-operation">
<title>Direct Operation</title>
<para>Certain operations can be performed directly on the
repository without touching the working copy. Specifically,
this applies to any operation that does not require editing
a file, including:</para>
<itemizedlist>
<listitem>
<para><literal>log</literal>,
<literal>diff</literal></para>
</listitem>
<listitem>
<para><literal>mkdir</literal></para>
</listitem>
<listitem>
<para><literal>remove</literal>, <literal>copy</literal>,
<literal>rename</literal></para>
</listitem>
<listitem>
<para><literal>propset</literal>,
<literal>propedit</literal>,
<literal>propdel</literal></para>
</listitem>
<listitem>
<para><literal>merge</literal></para>
</listitem>
</itemizedlist>
<para>Branching is very fast. The following command would be
used to branch <literal>RELENG_8</literal>:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/head svn+ssh://svn.freebsd.org/base/stable/8</userinput></screen>
<para>This is equivalent to the following set of commands
which take minutes and hours as opposed to seconds,
depending on your network connection:</para>
<screen>&prompt.user; <userinput>svn checkout --depth=immediates svn+ssh://svn.freebsd.org/base</userinput>
&prompt.user; <userinput>cd base</userinput>
&prompt.user; <userinput>svn update --depth=infinity head</userinput>
&prompt.user; <userinput>svn copy head stable/8</userinput>
&prompt.user; <userinput>svn commit stable/8</userinput></screen>
</sect3>
<sect3 id="svn-advanced-use-merging">
<title>Merging with <acronym>SVN</acronym></title>
<para>This section deals with merging code from one branch to
another (typically, from head to a stable branch).</para>
<note>
<para>In all examples below, <literal>&dollar;FSVN</literal>
refers to the location of the &os; Subversion repository,
<literal>svn+ssh://svn.freebsd.org/base/</literal>.</para>
</note>
<sect4>
<title>About Merge Tracking</title>
<para>From the user's perspective, merge tracking
information (or mergeinfo) is stored in a property called
<literal>svn:mergeinfo</literal>, which is a
comma-separated list of revisions and ranges of revisions
that have been merged. When set on a file, it applies
only to that file. When set on a directory, it applies to
that directory and its descendants (files and directories)
except for those that have their own
<literal>svn:mergeinfo</literal>.</para>
<para>It is <emphasis>not</emphasis> inherited. For
instance, <filename
class="directory">stable/6/contrib/openpam/</filename>
does not implicitly inherit mergeinfo from
<filename class="directory">stable/6/</filename>, or
<filename class="directory">stable/6/contrib/</filename>.
Doing so would make partial checkouts very hard to manage.
Instead, mergeinfo is explicitly propagated down the tree.
For merging something into
<filename class="directory">branch/foo/bar/</filename>,
the following rules apply:</para>
<orderedlist>
<listitem>
<para>If
<filename class="directory">branch/foo/bar/</filename>
does not already have a mergeinfo record, but a direct
ancestor (for instance,
<filename class="directory">branch/foo/</filename>)
does, then that record will be propagated down to
<filename class="directory">branch/foo/bar/</filename>
before information about the current merge is
recorded.</para>
</listitem>
<listitem>
<para>Information about the current merge will
<emphasis>not</emphasis> be propagated back up that
ancestor.</para>
</listitem>
<listitem>
<para>If a direct descendant of
<filename class="directory">branch/foo/bar/</filename>
(for instance, <filename
class="directory">branch/foo/bar/baz/</filename>)
already has a mergeinfo record, information about the
current merge will be propagated down to it.</para>
</listitem>
</orderedlist>
<para>If you consider the case where a revision changes
several separate parts of the tree (for example, <filename
class="directory">branch/foo/bar/</filename> and
<filename class="directory">branch/foo/quux/</filename>),
but you only want to merge some of it (for example,
<filename class="directory">branch/foo/bar/</filename>),
you will see that these rules make sense. If mergeinfo
was propagated up, it would seem like that revision had
also been merged to <filename
class="directory">branch/foo/quux/</filename>, when in
fact it had not been.</para>
</sect4>
<sect4>
<title>Selecting the Source and Target</title>
<para>Because of mergeinfo propagation, it is important to
choose the source and target for the merge carefully to
minimise property changes on unrelated directories.</para>
<para>The rules for selecting the merge target (the
directory that you will merge the changes to) can be
summarized as follows:</para>
<orderedlist>
<listitem>
<para>Never merge directly to a file.</para>
</listitem>
<listitem>
<para>Never, ever merge directly to a file.</para>
</listitem>
<listitem>
<para><emphasis>Never, ever, ever</emphasis> merge
directly to a file.</para>
</listitem>
<listitem>
<para>Changes to kernel code should be merged to
<filename class="directory">sys/</filename>. For
instance, a change to the &man.ichwd.4; driver should
be merged to
<filename class="directory">sys/</filename>, not
<filename class="directory">sys/dev/ichwd/</filename>.
Likewise, a change to the TCP/IP stack should be
merged to <filename class="directory">sys/</filename>,
not <filename
class="directory">sys/netinet/</filename>.</para>
</listitem>
<listitem>
<para>Changes to code under
<filename class="directory">etc/</filename> should be
merged at <filename class="directory">etc/</filename>,
not below it.</para>
</listitem>
<listitem>
<para>Changes to vendor code (code in
<filename class="directory">contrib/</filename>,
<filename class="directory">crypto/</filename> and so
on) should be merged to the directory where vendor
imports happen. For instance, a change to <filename
class="directory">crypto/openssl/util/</filename>
should be merged to <filename
class="directory">crypto/openssl/</filename>. This
is rarely an issue, however, since changes to vendor
code are usually merged wholesale.</para>
</listitem>
<listitem>
<para>Changes to userland programs should as a general
rule be merged to the directory that contains the
Makefile for that program. For instance, a change to
<filename
class="directory">usr.bin/xlint/arch/i386/</filename>
should be merged to <filename
class="directory">usr.bin/xlint/</filename>.</para>
</listitem>
<listitem>
<para>Changes to userland libraries should as a general
rule be merged to the directory that contains the
Makefile for that library. For instance, a change to
<filename class="directory">lib/libc/gen/</filename>
should be merged to <filename
class="directory">lib/libc/</filename>.</para>
</listitem>
<listitem>
<para>There may be cases where it makes sense to deviate
from the rules for userland programs and libraries.
For instance, everything under <filename
class="directory">lib/libpam/</filename> is merged
to <filename class="directory">lib/libpam/</filename>,
even though the library itself and all of the modules
each have their own Makefile.</para>
</listitem>
<listitem>
<para>Changes to manual pages should be merged to
<filename
class="directory">share/man/man<replaceable>N</replaceable>/</filename>,
for the appropriate value of
<literal>N</literal>.</para>
</listitem>
<listitem>
<para>Other changes to
<filename class="directory">share/</filename> should
be merged to the appropriate subdirectory and not to
<filename class="directory">share/</filename>
directly.</para>
</listitem>
<listitem>
<para>Changes to a top-level file in the source tree
such as <filename>UPDATING</filename> or
<filename>Makefile.inc1</filename> should be merged
directly to that file rather than to the root of the
whole tree. Yes, this is an exception to the first
three rules.</para>
</listitem>
<listitem>
<para>When in doubt, ask.</para>
</listitem>
</orderedlist>
<para>If you need to merge changes to several places at once
(for instance, changing a kernel interface and every
userland program that uses it), merge each target
separately, then commit them together. For instance, if
you merge a revision that changed a kernel
<acronym>API</acronym> and updated all the userland bits
that used that <acronym>API</acronym>, you would merge the
kernel change to sys, and the userland bits to the
appropriate userland directories, then commit all of these
in one go.</para>
<para>The source will almost invariably be the same as the
target. For instance, you will always merge <filename
class="directory">stable/7/lib/libc/</filename> from
<filename class="directory">head/lib/libc/</filename>.
The only exception would be when merging changes to code
that has moved in the source branch but not in the parent
branch. For instance, a change to &man.pkill.1; would be
merged from <filename
class="directory">bin/pkill/</filename> in head to
<filename class="directory">usr.bin/pkill/</filename> in
stable/7.</para>
</sect4>
<sect4>
<title>Preparing the Merge Target</title>
<para>Because of the mergeinfo propagation issues described
earlier, it is very important that you never merge changes
into a sparse working copy. You must always have a full
checkout of the branch you will merge into. For instance,
when merging from HEAD to 7, you must have a full checkout
of stable/7:</para>
<screen>&prompt.user; <userinput>cd stable/7</userinput>
&prompt.user; <userinput>svn up --set-depth=infinity</userinput></screen>
<para>The target directory must also be up-to-date and must
not contain any uncommitted changes or stray files.</para>
</sect4>
<sect4>
<title>Identifying Revisions</title>
<para>Identifying revisions to be merged is a must. If the
target already has complete mergeinfo, ask
<acronym>SVN</acronym> for a list:</para>
<screen>&prompt.user; <userinput>cd stable/6/contrib/openpam</userinput>
&prompt.user; <userinput>svn mergeinfo --show-revs=eligible $FSVN/head/contrib/openpam</userinput></screen>
<para>If the target does not have complete mergeinfo, check
the log for the merge source.</para>
</sect4>
<sect4>
<title>Merging</title>
<para>Now, let us start merging!</para>
<sect5>
<title>The Principles</title>
<para>Say you would like to merge:</para>
<itemizedlist>
<listitem>
<para>revision <literal>&dollar;R</literal></para>
</listitem>
<listitem>
<para>in directory &dollar;target in stable branch
&dollar;B</para>
</listitem>
<listitem>
<para>from directory &dollar;source in head</para>
</listitem>
<listitem>
<para>&dollar;FSVN is
<literal>svn+ssh://svn.freebsd.org/base</literal></para>
</listitem>
</itemizedlist>
<para>Assuming that revisions &dollar;P and &dollar;Q have
already been merged, and that the current directory is
an up-to-date working copy of stable/&dollar;B, the
existing mergeinfo looks like this:</para>
<screen>&prompt.user; <userinput>svn propget svn:mergeinfo -R $target</userinput>
$target - /head/$source:$P,$Q</screen>
<para>Merging is done like so:</para>
<screen>&prompt.user; <userinput>svn merge -c$R $FSVN/head/$source $target</userinput></screen>
<para>Checking the results of this is possible with
<command>svn diff</command>.</para>
<para>The svn:mergeinfo now looks like:</para>
<screen>&prompt.user; <userinput>svn propget svn:mergeinfo -R $target</userinput>
$target - head/$source:$P,$Q,$R</screen>
<para>If the results are not exactly as shown, assistance
may be required before committing as mistakes may have
been made, or there may be something wrong with the
existing mergeinfo, or there may be a bug in
Subversion.</para>
</sect5>
<sect5>
<title>Practical Example</title>
<para>As a practical example, consider the following
scenario: The changes to <filename>netmap.4</filename>
in r238987 is to be merged from CURRENT to 9-STABLE.
The file resides in <filename
class="directory">head/share/man/man4</filename> and
according to <xref linkend="svn-advanced-use-merging"/>
this is also where to do the merge. Note that in this
example all paths are relative to the top of the svn
repository. For more information on the directory
layout, see <xref
linkend="svn-getting-started-base-layout"/>.</para>
<para>The first step is to inspect the existing
mergeinfo.</para>
<screen>&prompt.user; <userinput>svn propget svn:mergeinfo -R stable/9/share/man/man4</userinput></screen>
<para>Take a quick note of how it looks before moving on
to the next step; doing the actual merge:</para>
<screen>&prompt.user; <userinput>svn merge -c r238987 svn+ssh://svn.freebsd.org/base/head/share/man/man4 stable/9/share/man/man4</userinput>
--- Merging r238987 into 'stable/9/share/man/man4':
U stable/9/share/man/man4/netmap.4
--- Recording mergeinfo for merge of r238987 into
'stable/9/share/man/man4':
U stable/9/share/man/man4</screen>
<para>Check that the revision number of the merged
revision has been added. Once this is verified, the
only thing left is the actual commit.</para>
<screen>&prompt.user; <userinput>svn commit stable/9/share/man/man4</userinput></screen>
</sect5>
<sect5>
<title>Merging into the Kernel
(<filename class="directory">sys/</filename>)</title>
<para>As stated above, merging into the kernel is
different from merging in the rest of the tree. In many
ways merging to the kernel is simpler because there is
always the same merge target
(<filename class="directory">sys/</filename>).</para>
<para>Once <command>svn merge</command> has been executed,
<command>svn diff</command> has to be run on the
directory to check the changes. This may show some
unrelated property changes, but these can be ignored.
Next, build and test the kernel, and, once the tests are
complete, commit the code as normal, making sure that
the commit message starts with <quote>Merge
<replaceable>r226222</replaceable> from head</quote>,
or similar.</para>
</sect5>
</sect4>
<sect4>
<title>Precautions Before Committing</title>
<para>As always, build world (or appropriate parts of
it).</para>
<para>Check the changes with <command>svn diff</command> and
<command>svn stat</command>. Make sure all the files that
should have been added or deleted were in fact added or
deleted.</para>
<para>Take a closer look at any property change (marked by a
<literal>M</literal> in the second column of <command>svn
stat</command>). Normally, no svn:mergeinfo properties
should be anywhere except the target directory (or
directories).</para>
<para>If something looks fishy, ask for help.</para>
</sect4>
<sect4>
<title>Committing</title>
<para>Make sure to commit a top level directory to have the
mergeinfo included as well. Do not specify individual
files on the command line. For more information about
committing files in general, see the relevant section of
this primer.</para>
</sect4>
</sect3>
<sect3 id="svn-advanced-use-vendor-imports">
<title>Vendor Imports with <acronym>SVN</acronym></title>
<important>
<para>Please read this entire section before starting a
vendor import.</para>
</important>
<note>
<para>Patches to vendor code fall into two
categories:</para>
<itemizedlist>
<listitem>
<para>Vendor patches: these are patches that have been
issued by the vendor, or that have been extracted from
the vendor's version control system, which address
issues which in your opinion cannot wait until the
next vendor release.</para>
</listitem>
<listitem>
<para>&os; patches: these are patches that modify the
vendor code to address &os;-specific issues.</para>
</listitem>
</itemizedlist>
<para>The nature of a patch dictates where it should be
committed:</para>
<itemizedlist>
<listitem>
<para>Vendor patches should be committed to the vendor
branch, and merged from there to head. If the patch
addresses an issue in a new release that is currently
being imported, it <emphasis>must not</emphasis> be
committed along with the new release: the release must
be imported and tagged first, then the patch can be
applied and committed. There is no need to re-tag the
vendor sources after committing the patch.</para>
</listitem>
<listitem>
<para>&os; patches should be committed directly to
head.</para>
</listitem>
</itemizedlist>
</note>
<sect4>
<title>Preparing the Tree</title>
<para>If importing for the first time after the switch to
Subversion, flattening and cleaning up the vendor tree is
necessary, as well as bootstrapping the merge history in
the main tree.</para>
<sect5>
<title>Flattening</title>
<para>During the conversion from <acronym>CVS</acronym> to
Subversion, vendor branches were imported with the same
layout as the main tree. This means that the
<literal>pf</literal> vendor sources ended up in
<filename>vendor/pf/dist/contrib/pf</filename>. The
vendor source is best directly in
<filename>vendor/pf/dist</filename>.</para>
<para>To flatten the <literal>pf</literal> tree:</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf/dist/contrib/pf</replaceable></userinput>
&prompt.user; <userinput>svn mv $(svn list) ../..</userinput>
&prompt.user; <userinput>cd ../..</userinput>
&prompt.user; <userinput>svn rm contrib</userinput>
&prompt.user; <userinput>svn propdel -R svn:mergeinfo .</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>The <literal>propdel</literal> bit is necessary
because starting with 1.5, Subversion will automatically
add <literal>svn:mergeinfo</literal> to any directory
that is copied or moved. In this case, as nothing is
being merged from the deleted tree, they just get in the
way.</para>
<para>Tags may be flattened as well (3, 4, 3.5 etc.); the
procedure is exactly the same, only changing
<literal>dist</literal> to <literal>3.5</literal> or
similar, and putting the <command>svn commit</command>
off until the end of the process.</para>
</sect5>
<sect5>
<title>Cleaning Up</title>
<para>The <literal>dist</literal> tree can be cleaned up
as necessary. Disabling keyword expansion is
recommended, as it makes no sense on unmodified vendor
code and in some cases it can even be harmful.
<application>OpenSSH</application>, for example,
includes two files that originated with &os; and still
contain the original version tags. To do this:</para>
<screen>&prompt.user; <userinput>svn propdel svn:keywords -R .</userinput>
&prompt.root; <userinput>svn commit</userinput></screen>
</sect5>
<sect5>
<title>Bootstrapping Merge History</title>
<para>If importing for the first time after the switch to
Subversion, bootstrap <literal>svn:mergeinfo</literal>
on the target directory in the main tree to the revision
that corresponds to the last related change to the
vendor tree, prior to importing new sources:</para>
<screen>&prompt.user; <userinput>cd <replaceable>head/contrib/pf</replaceable></userinput>
&prompt.user; <userinput>svn merge --record-only svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist@180876</replaceable> .</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
</sect5>
</sect4>
<sect4>
<title>Importing New Sources</title>
<para>With two commits&mdash;one for the import itself and
one for the tag&mdash;this step can optionally be repeated
for every upstream release between the last import and the
current import.</para>
<sect5>
<title>Preparing the Vendor Sources</title>
<para>Unlike in <acronym>CVS</acronym> where only the
needed parts were imported into the vendor tree to avoid
bloating the main tree, Subversion is able to store a
full distribution in the vendor tree. So, import
everything, but merge only what is required.</para>
<para>A <command>svn add</command> is required to add any
files that were added since the last vendor import, and
<command>svn rm</command> is required to remove any that
were removed since. Preparing sorted lists of the
contents of the vendor tree and of the sources that are
about to be imported is recommended, to facilitate the
process.</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf/dist</replaceable></userinput>
&prompt.user; <userinput>svn list -R | grep -v '/$' | sort >../old</userinput>
&prompt.user; <userinput>cd <replaceable>../pf-4.3</replaceable></userinput>
&prompt.user; <userinput>find . -type f | cut -c 3- | sort >../new</userinput></screen>
<para>With these two files,
<command>comm -23 ../old ../new</command> will list
removed files (files only in <filename>old</filename>),
while <command>comm -13 ../old ../new</command> will
list added files only in
<filename>new</filename>.</para>
</sect5>
<sect5>
<title>Importing into the Vendor Tree</title>
<para>Now, the sources must be copied into
<filename><replaceable>dist</replaceable></filename> and
the <command>svn add</command> and
<command>svn rm</command> commands should be used as
needed:</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf/pf-4.3</replaceable></userinput>
&prompt.user; <userinput>tar cf - . | tar xf - -C ../dist</userinput>
&prompt.user; <userinput>cd <replaceable>../dist</replaceable></userinput>
&prompt.user; <userinput>comm -23 ../old ../new | xargs svn rm</userinput>
&prompt.user; <userinput>comm -13 ../old ../new | xargs svn --parents add</userinput></screen>
<para>If any directories were removed, they will have to
be <command>svn rm</command>ed manually. Nothing will
break if they are not, but they will remain in the
tree.</para>
<para>Check properties on any new files. All text files
should have <literal>svn:eol-style</literal> set to
<literal>native</literal>. All binary files should have
<literal>svn:mime-type</literal> set to
<literal>application/octet-stream</literal> unless there
is a more appropriate media type. Executable files
should have <literal>svn:executable</literal> set to
<literal>*</literal>. No other properties should exist
on any file in the tree.</para>
<para>Committing is now possible, however it is good
practice to make sure that everything is OK by using the
<command>svn stat</command> and
<command>svn diff</command> commands.</para>
</sect5>
<sect5>
<title>Tagging</title>
<para>Once committed, vendor releases should be tagged for
future reference. The best and quickest way to do this
is directly in the repository:</para>
<screen>&prompt.user; <userinput>svn cp svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist</replaceable> svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/4.3</replaceable></userinput></screen>
<para>Once that is complete, <command>svn up</command> the
working copy of
<filename><replaceable>vendor/pf</replaceable></filename>
to get the new tag, although this is rarely
needed.</para>
<para>If creating the tag in the working copy of the tree,
<command>svn:mergeinfo</command> results must be
removed:</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf</replaceable></userinput>
&prompt.user; <userinput>svn cp dist 4.3</userinput>
&prompt.user; <userinput>svn propdel svn:mergeinfo -R 4.3</userinput></screen>
</sect5>
</sect4>
<sect4>
<title>Merging to Head</title>
<screen>&prompt.user; <userinput>cd <replaceable>head/contrib/pf</replaceable></userinput>
&prompt.user; <userinput>svn up</userinput>
&prompt.user; <userinput>svn merge --accept=postpone svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist</replaceable> .</userinput></screen>
<para>The <literal>--accept=postpone</literal> tells
Subversion that it should not complain because merge
conflicts will be taken care of manually.</para>
<para>It is necessary to resolve any merge conflicts.
This process is the same in <acronym>SVN</acronym> as in
<acronym>CVS</acronym>.</para>
<para>Make sure that any files that were added or removed in
the vendor tree have been properly added or removed in the
main tree. To check diffs against the vendor
branch:</para>
<screen>&prompt.user; <userinput>svn diff --no-diff-deleted --old=svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist</replaceable> --new=.</userinput></screen>
<para>The <literal>--no-diff-deleted</literal> tells
Subversion not to complain about files that are in the
vendor tree but not in the main tree, i.e., things that
would have previously been removed before the vendor
import, like for example the vendor's makefiles
and configure scripts.</para>
<para>Using <acronym>CVS</acronym>, once a file was off the
vendor branch, it was not able to be put back. With
Subversion, there is no concept of on or off the vendor
branch. If a file that previously had local
modifications, to make it not show up in diffs in the
vendor tree, all that has to be done is remove any
left-over cruft like &os; version tags, which is much
easier.</para>
<para>If any changes are required for the world to build
with the new sources, make them now, and keep testing
until everything builds and runs perfectly.</para>
</sect4>
<sect4>
<title>Committing the Vendor Import</title>
<para>Committing is now possible! Everything must be
committed in one go. If done properly, the tree will move
from a consistent state with old code, to a consistent
state with new code.</para>
</sect4>
<sect4>
<title>From Scratch</title>
<sect5>
<title>Importing into the Vendor Tree</title>
<para>This section is an example of importing and tagging
<application>byacc</application> into
<filename class="directory">head</filename>.</para>
<para>First, prepare the directory in
<filename class="directory">vendor</filename>:</para>
<screen>&prompt.user; <userinput>svn co --depth immediates <replaceable>$FSVN/vendor</replaceable></userinput>
&prompt.user; <userinput>cd <replaceable>vendor</replaceable></userinput>
&prompt.user; <userinput>svn mkdir <replaceable>byacc</replaceable></userinput>
&prompt.user; <userinput>svn mkdir <replaceable>byacc/dist</replaceable></userinput></screen>
<para>Now, import the sources into the
<filename class="directory">dist</filename> directory.
Once the files are in place, <command>svn add</command>
the new ones, then <command>svn commit</command> and tag
the imported version. To save time and bandwidth,
direct remote committing and tagging is possible:</para>
<screen>&prompt.user; <userinput>svn cp -m <replaceable>"Tag byacc 20120115"</replaceable> <replaceable>$FSVN/vendor/byacc/dist</replaceable> <replaceable>$FSVN/vendor/byacc/20120115</replaceable></userinput></screen>
</sect5>
<sect5>
<title>Merging to <literal>head</literal></title>
<para>Due to this being a new file, copy it for the
merge:</para>
<screen>&prompt.user; <userinput>svn cp -m <replaceable>"Import byacc to contrib"</replaceable> <replaceable>$FSVN/vendor/byacc/dist</replaceable> <replaceable>$FSVN/head/contrib/byacc</replaceable></userinput></screen>
<para>Working normally on newly imported sources is still
possible.</para>
</sect5>
</sect4>
</sect3>
<sect3 id="svn-advanced-use-reverting-a-commit">
<title>Reverting a Commit</title>
<para>Reverting a commit to a previous version is fairly
easy:</para>
<screen>&prompt.user; <userinput>svn merge -r179454:179453 ROADMAP.txt</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>Change number syntax, with negative meaning a reverse
change, can also be used:</para>
<screen>&prompt.user; <userinput>svn merge -c -179454 ROADMAP.txt</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>This can also be done directly in the repository:</para>
<screen>&prompt.user; <userinput>svn merge -r179454:179453 svn+ssh://svn.freebsd.org/base/ROADMAP.txt</userinput></screen>
<note>
<para>It is important to ensure that the mergeinfo
is correct when reverting a file in order to permit
<command>svn mergeinfo --eligible</command> to work as
expected.</para>
</note>
<para>Reverting the deletion of a file is slightly different.
Copying the version of the file that predates the deletion
is required. For example, to restore a file that was
deleted in revision N, restore version N-1:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/ROADMAP.txt@179454</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>or, equally:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/ROADMAP.txt@179454 svn+ssh://svn.freebsd.org/base</userinput></screen>
<para>Do <emphasis>not</emphasis> simply recreate the file
manually and <command>svn add</command> it&mdash;this will
cause history to be lost.</para>
</sect3>
<sect3 id="svn-advanced-use-fixing-mistakes">
<title>Fixing Mistakes</title>
<para>While we can do surgery in an emergency, do not plan on
having mistakes fixed behind the scenes. Plan on mistakes
remaining in the logs forever. Be sure to check the output
of <command>svn status</command> and <command>svn
diff</command> before committing.</para>
<para>Mistakes will happen but,
they can generally be fixed without
disruption.</para>
<para>Take a case of adding a file in the wrong location. The
right thing to do is to <command>svn move</command> the file
to the correct location and commit. This causes just a
couple of lines of metadata in the repository journal, and
the logs are all linked up correctly.</para>
<para>The wrong thing to do is to delete the file and then
<command>svn add</command> an independent copy in the
correct location. Instead of a couple of lines of text, the
repository journal grows an entire new copy of the file.
This is a waste.</para>
</sect3>
<sect3 id="svn-advanced-use-setting-up-svnsync">
<title>Setting up a <application>svnsync</application>
Mirror</title>
<para>You probably do not want to do this unless there is a
good reason for it. Such reasons might be to support many
multiple local read-only client machines, or if your network
bandwidth is limited. Starting a fresh mirror from empty
would take a very long time. Expect a minimum of 10 hours
for high speed connectivity. If you have international
links, expect this to take 4 to 10 times longer.</para>
<para>A far better option is to grab a seed file. It is large
(~1GB) but will consume less network traffic and take less
time to fetch than a svnsync will. This is possible in one
of the following three ways:</para>
<screen>&prompt.user; <userinput>rsync -va --partial --progress freefall:/home/peter/svnmirror-base-r179637.tbz2 .</userinput></screen>
<screen>&prompt.user; <userinput>rsync -va --partial --progress rsync://repoman.freebsd.org:50873/svnseed/svnmirror-base-r215629.tar.xz .</userinput></screen>
<screen>&prompt.user; <userinput>fetch ftp://ftp.freebsd.org/pub/FreeBSD/development/subversion/svnmirror-base-r221445.tar.xz</userinput></screen>
<para>Once you have the file, extract it to somewhere like
<filename class="directory">home/svnmirror/base/</filename>.
Then, update it, so that it fetches changes since the last
revision in the archive:</para>
<screen>&prompt.user; <userinput>svnsync sync file:///home/svnmirror/base</userinput></screen>
<para>You can then set that up to run from &man.cron.8;, do
checkouts locally, set up a svnserve server for your local
machines to talk to, etc.</para>
<para>The seed mirror is set to fetch from
<literal>svn://svn.freebsd.org/base</literal>. The
configuration for the mirror is stored in
<literal>revprop 0</literal> on the local mirror. To see
the configuration, try:</para>
<screen>&prompt.user; <userinput>svn proplist -v --revprop -r 0 file:///home/svnmirror/base</userinput></screen>
<para>Use <literal>propset</literal> to change things.</para>
</sect3>
<sect3 id="svn-advanced-use-committing-high-ascii-data">
<title>Committing High-<acronym>ASCII</acronym> Data</title>
<para>Files that have high-<acronym>ASCII</acronym> bits are
considered binary files in <acronym>SVN</acronym>, so the
pre-commit checks fail and indicate that the
<literal>mime-type</literal> property should be set to
<literal>application/octet-stream</literal>. However, the
use of this is discouraged, so please do not set it. The
best way is always avoiding high-<acronym>ASCII</acronym>
data, so that it can be read everywhere with any text editor
but if it is not avoidable, instead of changing the
mime-type, set the <literal>fbsd:notbinary</literal>
property with <literal>propset</literal>:</para>
<screen>&prompt.user; <userinput>svn propset fbsd:notbinary yes foo.data</userinput></screen>
</sect3>
<sect3 id="svn-advanced-use-maintaining-a-project-branch">
<title>Maintaining a Project Branch</title>
<para>A project branch is one that is synced to head (or
another branch) is used to develop a project then commit it
back to head. In <acronym>SVN</acronym>,
<quote>dolphin</quote> branching is used for this. A
<quote>dolphin</quote> branch is one that diverges for a
while and is finally committed back to the original branch.
During development code migration in one direction (from
head to the branch only). No code is committed back to head
until the end. Once you commit back at the end, the branch
is dead (although you can have a new branch with the same
name after you delete the branch if you want).</para>
<para>As per <ulink
url="http://people.freebsd.org/~peter/svn_notes.txt">http://people.freebsd.org/~peter/svn_notes.txt</ulink>,
work that is intended to be merged back into HEAD should be
in <filename class="directory">base/projects/</filename>.
If you are doing work that is beneficial to the &os;
community in some way but not intended to be merged directly
back into HEAD then the proper location is <filename
class="directory">base/user/<replaceable>your-name</replaceable>/</filename>.
<ulink
url="http://svnweb.freebsd.org/base/projects/GUIDELINES.txt">This
page</ulink> contains further details.</para>
<para>To create a project branch:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/head svn+ssh://svn.freebsd.org/base/projects/spif</userinput></screen>
<para>To merge changes from HEAD back into the project
branch:</para>
<screen>&prompt.user; <userinput>cd copy_of_spif</userinput>
&prompt.user; <userinput>svn merge svn+ssh://svn.freebsd.org/base/head</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>It is important to resolve any merge conflicts before
committing.</para>
<!--
<para>To collapse everything back at the end:</para>
<screen>&prompt.user; <userinput>svn write me</userinput></screen>
-->
</sect3>
</sect2>
<sect2>
<title>Some Tips</title>
<para>In commit logs etc., <quote>rev 179872</quote> should be
spelled <quote>r179872</quote> as per convention.</para>
<para>Do not remove and re-add the same file in a single commit
as this will break the CVS exporter.</para>
<para>Speeding up svn is possible by adding the following to
<filename>~/.ssh/config</filename>:</para>
<screen>Host *
ControlPath ~/.ssh/sockets/master-%l-%r@%h:%p
ControlMaster auto
ControlPersist yes</screen>
<para>and then typing</para>
<screen><userinput>mkdir ~/.ssh/sockets</userinput></screen>
<para>Checking out a working copy with a stock Subversion client
without &os;-specific patches
(<makevar>OPTIONS_SET=FREEBSD_TEMPLATE</makevar>) will mean
that <literal>&dollar;FreeBSD&dollar;</literal> tags will not
be expanded. Once the correct version has been installed,
trick Subversion into expanding them like so:</para>
<screen>&prompt.user; <userinput>svn propdel -R svn:keywords .</userinput>
&prompt.user; <userinput>svn revert -R .</userinput></screen>
<para>This will wipe out uncommitted patches.</para>
</sect2>
</sect1>
<sect1 id="conventions">
<title>Conventions and Traditions</title>
<para>As a new developer there are a number of things you should
do first. The first set is specific to committers only. (If
you are not a committer, e.g., have GNATS-only access, then your
mentor needs to do these things for you.)</para>
<sect2 id="conventions-committers">
<title>Guidelines for Committers</title>
<note>
<para>The <literal>.ent</literal>, <literal>.xml</literal>,
and <literal>.xml</literal> files listed below exist in the
&os; Documentation Project SVN repository at
<hostid role="fqdn">svn.FreeBSD.org/doc/</hostid>.</para>
</note>
<para>If you have been given commit rights to one or more of the
repositories:</para>
<itemizedlist>
<listitem>
<para>Add your author entity to
<filename>head/share/xml/authors.ent</filename>; this
should be done first since an omission of this commit will
cause the next commits to break the doc/ build.</para>
<para>This is a relatively easy task, but remains a good
first test of your version control skills.</para>
<important>
<para>New files that do not have the
<literal>FreeBSD=%H</literal>
<command>svn:keywords</command> property will be
rejected when attempting to commit them to the
repository. Be sure to read
<xref linkend="svn-daily-use-adding-and-removing"/>
regarding adding and removing files, in addition to
verifying that <filename>~/.subversion/config</filename>
contains the necessary &quot;auto-props&quot; entries
from <filename>auto-props.txt</filename> mentioned
there.</para>
</important>
<note>
<para>Do not forget to get mentor approval for these
patches!</para>
</note>
</listitem>
<listitem>
<para>Add yourself to the <quote>Developers</quote> section
of the <ulink
url="&url.articles.contributors;/index.html">Contributors
List</ulink>
(<filename>head/en_US.ISO8859-1/articles/contributors/contrib.committers.xml</filename>)
and remove yourself from the
<quote>Additional Contributors</quote> section
(<filename>head/en_US.ISO8859-1/articles/contributors/contrib.additional.xml</filename>).
Please note that entries are sorted by last name.</para>
</listitem>
<listitem>
<para>Add an entry for yourself to
<filename>head/share/xml/news.xml</filename>. Look for
the other entries that look like
<quote>A new committer</quote> and follow the
format.</para>
</listitem>
<listitem>
<para>You should add your PGP or GnuPG key to
<filename>head/share/pgpkeys</filename> (and if you do not
have a key, you should create one). Do not forget to
commit the updated
<filename>head/share/pgpkeys/pgpkeys.ent</filename> and
<filename>head/share/pgpkeys/pgpkeys-developers.xml</filename>.
Please note that entries are sorted by last name.</para>
<para>&a.des.email; has written a shell script
(<filename>head/share/pgpkeys/addkey.sh</filename>) to
make this extremely simple. See the <ulink
url="http://svnweb.FreeBSD.org/doc/head/share/pgpkeys/README">README</ulink>
file for more information.</para>
<note>
<para>It is important to have an up-to-date PGP/GnuPG key
in the Handbook, since the key may be required for
positive identification of a committer, e.g., by the
&a.admins; for account recovery. A complete keyring of
<hostid role="domainname">FreeBSD.org</hostid> users is
available for download from <ulink
url="&url.base;/doc/pgpkeyring.txt">http://www.FreeBSD.org/doc/pgpkeyring.txt</ulink>.</para>
</note>
</listitem>
<listitem>
<para>Add an entry for yourself to
<filename>src/share/misc/committers-<replaceable>repository</replaceable>.dot</filename>,
where repository is either doc, ports or src, depending on
the commit privileges you obtained.</para>
</listitem>
<listitem>
<para>Some people add an entry for themselves to
<filename>ports/astro/xearth/files/freebsd.committers.markers</filename>.</para>
</listitem>
<listitem>
<para>Some people add an entry for themselves to
<filename>src/usr.bin/calendar/calendars/calendar.freebsd</filename>.</para>
</listitem>
<listitem>
<para>If you already have an account at the
<ulink url="http://wiki.freebsd.org">&os; wiki</ulink>,
make sure your mentor moves you from the <ulink
url="http://wiki.freebsd.org/ContributorsGroup">Contributors
group</ulink> to the <ulink
url="http://wiki.freebsd.org/DevelopersGroup">Developers
group</ulink>. Otherwise, consider signing up for an
account so you can publish projects and ideas you are
working on.</para>
</listitem>
<listitem>
<para>Once you get access to the wiki, you may add yourself
to the
<ulink url="http://wiki.freebsd.org/HowWeGotHere">How We
Got Here</ulink> and
<ulink url="http://wiki.freebsd.org/IrcNicks">Irc
Nicks</ulink> pages.</para>
</listitem>
<listitem>
<para>If you subscribe to &a.svn-src-all.name;,
&a.svn-ports-all.name; or &a.svn-doc-all.name;, you will
probably want to unsubscribe to avoid receiving duplicate
copies of commit messages and their followups.</para>
</listitem>
</itemizedlist>
<note>
<para>All <filename>src</filename> commits should go to
&os.current; first before being merged to &os.stable;. No
major new features or high-risk modifications should be made
to the &os.stable; branch.</para>
</note>
</sect2>
<sect2 id="conventions-everyone">
<title>Guidelines for Everyone</title>
<para>Whether or not you have commit rights:</para>
<itemizedlist>
<listitem>
<para>Introduce yourself to the other developers, otherwise
no one will have any idea who you are or what you are
working on. You do not have to write a comprehensive
biography, just write a paragraph or two about who you are
and what you plan to be working on as a developer in
&os;. (You should also mention who your mentor will
be). Email this to the &a.developers; and you will be on
your way!</para>
</listitem>
<listitem>
<para>Log into <hostid>hub.FreeBSD.org</hostid> and create a
<filename>/var/forward/<replaceable>user</replaceable></filename>
(where <replaceable>user</replaceable> is your username)
file containing the e-mail address where you want mail
addressed to
<replaceable>yourusername</replaceable>@FreeBSD.org to be
forwarded. This includes all of the commit messages as
well as any other mail addressed to the &a.committers; and
the &a.developers;. Really large mailboxes which have
taken up permanent residence on <hostid>hub</hostid> often
get <quote>accidentally</quote> truncated without warning,
so forward it or read it and you will not lose it.</para>
<para>Due to the severe load dealing with SPAM places on the
central mail servers that do the mailing list processing
the front-end server does do some basic checks and will
drop some messages based on these checks. At the moment
proper DNS information for the connecting host is the only
check in place but that may change. Some people blame
these checks for bouncing valid email. If you want these
checks turned off for your email you can place a file
named <filename>.spam_lover</filename> in your home
directory on
<hostid role="fqdn">freefall.FreeBSD.org</hostid> to
disable the checks for your email.</para>
</listitem>
</itemizedlist>
<note>
<para>If you are a developer but not a committer, you will
not be subscribed to the committers or developers mailing
lists; the subscriptions are derived from the access
rights.</para>
</note>
</sect2>
<sect2 id="mentors">
<title>Mentors</title>
<para>All new developers also have a mentor assigned to them for
the first few months. Your mentor is responsible for teaching
you the rules and conventions of the project and guiding your
first steps in the developer community. Your mentor is also
personally responsible for your actions during this initial
period.</para>
<para>For committers: until your mentor decides (and announces
with a forced commit to <filename>access</filename>) that you
have learned the ropes and are ready to commit on your own,
you should not commit anything without first getting your
mentor's review and approval, and you should document that
approval with an <literal>Approved by:</literal> line in the
commit message.</para>
</sect2>
</sect1>
<sect1 id="pref-license">
<title>Preferred License for New Files</title>
<para>Currently the &os; Project suggests and uses the following
text as the preferred license scheme:</para>
<programlisting>/*-
* Copyright (c) [year] [your name]
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* [id for your version control system, if any]
*/</programlisting>
<para>The &os; project strongly discourages the so-called
"advertising clause" in new code. Due to the large number of
contributors to the &os; project, complying with this clause for
many commercial vendors has become difficult. If you have code
in the tree with the advertising clause, please consider
removing it. In fact, please consider using the above license
for your code.</para>
<para>The &os; project discourages completely new licenses and
variations on the standard licenses. New licenses require the
approval of the &a.core; to reside in the
main repository. The more different licenses that are used in
the tree, the more problems that this causes to those wishing to
utilize this code, typically from unintended consequences from a
poorly worded license.</para>
<para>Project policy dictates that code under some non-BSD
licenses must be placed only in specific sections of the
repository, and in some cases, compilation must be conditional
or even disabled by default. For example, the GENERIC kernel
must be compiled under only licenses identical to or
substantially similar to the BSD license. GPL, APSL, CDDL, etc,
licensed software must not be compiled into GENERIC.</para>
<para>Developers are reminded that in open source, getting "open"
right is just as important as getting "source" right, as
improper handling of intellectual property has serious
consequences. Any questions or concerns should immediately be
brought to the attention of the core team.</para>
</sect1>
<sect1 id="developer.relations">
<title>Developer Relations</title>
<para>If you are working directly on your own code or on code
which is already well established as your responsibility, then
there is probably little need to check with other committers
before jumping in with a commit. If you see a bug in an area of
the system which is clearly orphaned (and there are a few such
areas, to our shame), the same applies. If, however, you are
about to modify something which is clearly being actively
maintained by someone else (and it is only by watching the
<literal><replaceable>repository</replaceable>-committers</literal>
mailing list that you can really get a feel for just what is and
is not) then consider sending the change to them instead, just
as you would have before becoming a committer. For ports, you
should contact the listed <makevar>MAINTAINER</makevar> in the
<filename>Makefile</filename>. For other parts of the
repository, if you are unsure who the active maintainer might
be, it may help to scan the revision history to see who has
committed changes in the past. &a.fenner.email; has written a nice
shell script that can help determine who the active maintainer
might be. It lists each person who has committed to a given
file along with the number of commits each person has made. It
can be found on <hostid>freefall</hostid> at
<filename>~fenner/bin/whodid</filename>. If your queries go
unanswered or the committer otherwise indicates a lack of
interest in the area affected, go ahead and commit it.</para>
<para>If you are unsure about a commit for any reason at
all, have it reviewed by <literal>-hackers</literal>
before committing. Better to have it flamed then and there
rather than when it is part of the repository. If you do
happen to commit something which results in controversy
erupting, you may also wish to consider backing the change out
again until the matter is settled. Remember &ndash; with a
version control system we can always change it back.</para>
<para>Do not impugn the intentions of someone you disagree with.
If they see a different solution to a problem than you, or even
a different problem, it is not because they are stupid, because
they have questionable parentage, or because they are trying to
destroy your hard work, personal image, or &os;, but simply
because they have a different outlook on the world. Different
is good.</para>
<para>Disagree honestly. Argue your position from its merits,
be honest about any shortcomings it may have, and be open to
seeing their solution, or even their vision of the problem,
with an open mind.</para>
<para>Accept correction. We are all fallible. When you have made
a mistake, apologize and get on with life. Do not beat up
yourself, and certainly do not beat up others for your mistake.
Do not waste time on embarrassment or recrimination, just fix
the problem and move on.</para>
<para>Ask for help. Seek out (and give) peer reviews. One of
the ways open source software is supposed to excel is in the
number of eyeballs applied to it; this does not apply if nobody
will review code.</para>
</sect1>
<sect1 id="gnats">
<title>GNATS</title>
<para>The &os; Project utilizes
<application>GNATS</application> for tracking bugs and change
requests. Be sure that if you commit a fix or suggestion found
in a <application>GNATS</application> PR, you use
<command>edit-pr <replaceable>pr-number</replaceable></command>
on <hostid>freefall</hostid> to close it. It is also considered
nice if you take time to close any PRs associated with your
commits, if appropriate. You can also make use of
&man.send-pr.1; yourself for proposing any change which you feel
should probably be made, pending a more extensive peer-review
first.</para>
<para>You can find out more about <application>GNATS</application>
at:</para>
<itemizedlist>
<listitem>
<para><ulink
url="&url.articles.pr-guidelines;/index.html">&os;
Problem Report Handling Guidelines</ulink></para>
</listitem>
<listitem>
<para><ulink
url="http://www.cs.utah.edu/csinfo/texinfo/gnats/gnats.html"></ulink></para>
</listitem>
<listitem>
<para><ulink
url="&url.base;/support.html">http://www.FreeBSD.org/support.html</ulink></para>
</listitem>
<listitem>
<para>&man.send-pr.1;</para>
</listitem>
</itemizedlist>
<para>You can run a local copy of GNATS, and then integrate the
&os; GNATS tree by creating an
<application>rsync</application> mirror. Then you can run GNATS
commands locally, allowing you to query the PR database without
an Internet connection.</para>
<sect2>
<title>Mirroring the GNATS Tree</title>
<para>It is possible to mirror the GNATS database by installing
<filename role="package">net/rsync</filename>, and
executing:</para>
<screen>&prompt.user; <userinput>rsync -va rsync://bit0.us-west.freebsd.org/FreeBSD-bit/gnats .</userinput></screen>
</sect2>
<sect2 id="gnatstools">
<title>Useful Tools</title>
<para>Other than <command>edit-pr</command> there are a
collection of tools in <filename>~gnats/tools/</filename>
on <hostid>freefall</hostid> which can make working with PRs
much easier.</para>
<para><command>open-pr</command>, <command>close-pr</command>,
<command>take-pr</command>, and <command>feedback-pr</command>
take PR numbers as arguments and then ask you to select from a
preexisting list of change reasons or let you type in your
own.</para>
<para><command>change-pr</command> is a multi purpose tool
that lets you make multiple changes at the same time with one
command.</para>
<para>For example, to assign PR 123456 to yourself type
<command>take-pr <replaceable>123456</replaceable></command>.
If you want to set the PR to patched awaiting an MFC at
the same time use:
<command>change-pr -t -p -m "awaiting MFC"
<replaceable>123456</replaceable></command></para>
</sect2>
</sect1>
<sect1 id="people">
<title>Who's Who</title>
<para>Besides the repository meisters, there are other &os;
project members and teams whom you will probably get to know in
your role as a committer. Briefly, and by no means
all-inclusively, these are:</para>
<variablelist>
<varlistentry>
<term>&a.doceng;</term>
<listitem>
<para>doceng is the group responsible for the documentation
build infrastructure, approving new documentation
committers, and ensuring that the &os; website and
documentation on the FTP site is up to date with respect
to the CVS tree. It is not a conflict resolution body.
The vast majority of documentation related discussion
takes place on the &a.doc;. More details regarding the
doceng team can be found in its <ulink
url="http://www.FreeBSD.org/internal/doceng.html">charter</ulink>.
Committers interested in contributing to the documentation
should familiarize themselves with the <ulink
url="&url.books.fdp-primer;/index.html">Documentation
Project Primer</ulink>.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.ru.email;</term>
<listitem>
<para>Ruslan is Mister &man.mdoc.7;. If you are writing a
manual page and need some advice on the structure, or the
markup, ask Ruslan.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.bde.email;</term>
<listitem>
<para>Bruce is the Style Police-Meister. When you do a
commit that could have been done better, Bruce will be
there to tell you. Be thankful that someone is. Bruce is
also very knowledgeable on the various standards
applicable to &os;.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.re.members.email;</term>
<listitem>
<para>These are the members of the &a.re;. This team is
responsible for setting release deadlines and controlling
the release process. During code freezes, the release
engineers have final authority on all changes to the
system for whichever branch is pending release status. If
there is something you want merged from &os.current; to
&os.stable; (whatever values those may have at any given
time), these are the people to talk to about it.</para>
<para>Hiroki is also the keeper of the release documentation
(<filename>src/release/doc/*</filename>). If you commit a
change that you think is worthy of mention in the release
notes, please make sure he knows about it. Better still,
send him a patch with your suggested commentary.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.des.email;</term>
<listitem>
<para>Dag-Erling is the
<ulink url="&url.base;/security/">&os; Security
Officer</ulink> and oversees the
&a.security-officer;.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.wollman.email;</term>
<listitem>
<para>If you need advice on obscure network internals or
are not sure of some potential change to the networking
subsystem you have in mind, Garrett is someone to talk
to. Garrett is also very knowledgeable on the various
standards applicable to &os;.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.committers;</term>
<listitem>
<para>&a.svn-src-all.name;, &a.svn-ports-all.name; and
&a.svn-doc-all.name; are the mailing lists that the
version control system uses to send commit messages to.
You should <emphasis>never</emphasis> send email directly
to these lists. You should only send replies to this list
when they are short and are directly related to a
commit.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.developers;</term>
<listitem>
<para>All committers are subscribed to -developers. This
list was created to be a forum for the committers
<quote>community</quote> issues. Examples are Core
voting, announcements, etc.</para>
<para>The &a.developers; is for the exclusive use of &os;
committers. In order to develop &os;, committers must
have the ability to openly discuss matters that will be
resolved before they are publicly announced. Frank
discussions of work in progress are not suitable for open
publication and may harm &os;.</para>
<para>All &os; committers are reminded to obey the
copyright of the original author(s) of &a.developers;
mail. Do not publish or forward messages from the
&a.developers; outside the list membership without
permission of all of the authors.</para>
<para>Copyright violators will be removed from the
&a.developers;, resulting in a suspension of commit
privileges. Repeated or flagrant violations may result in
permanent revocation of commit privileges.</para>
<para>This list is <emphasis>not</emphasis> intended as a
place for code reviews or a replacement for the &a.arch;.
In fact using it as such hurts the &os; Project as it
gives a sense of a closed list where general decisions
affecting all of the &os; using community are made
without being <quote>open</quote>. Last, but not least
<emphasis>never, never ever, email the &a.developers; and
CC:/BCC: another &os; list</emphasis>. Never, ever
email another &os; email list and CC:/BCC: the
&a.developers;. Doing so can greatly diminish the
benefits of this list.</para>
</listitem>
</varlistentry>
</variablelist>
</sect1>
<sect1 id="ssh.guide">
<title>SSH Quick-Start Guide</title>
<procedure>
<step>
<para>If you do not wish to type your password in every
time you use &man.ssh.1;, and you use RSA or DSA keys to
authenticate, &man.ssh-agent.1; is there for your
convenience. If you want to use &man.ssh-agent.1;, make
sure that you run it before running other applications. X
users, for example, usually do this from their
<filename>.xsession</filename> or
<filename>.xinitrc</filename>. See &man.ssh-agent.1;
for details.</para>
</step>
<step>
<para>Generate a key pair using &man.ssh-keygen.1;. The key
pair will wind up in your
<filename><envar>$HOME</envar>/.ssh/</filename>
directory.</para>
</step>
<step>
<para>Send your public key
(<filename><envar>$HOME</envar>/.ssh/id_dsa.pub</filename>
or
<filename><envar>$HOME</envar>/.ssh/id_rsa.pub</filename>)
to the person setting you up as a committer so it can be put
into the
<filename><replaceable>yourlogin</replaceable></filename>
file in
<filename class="directory">/etc/ssh-keys/</filename> on
<hostid>freefall</hostid>.</para>
</step>
</procedure>
<para>Now you should be able to use &man.ssh-add.1; for
authentication once per session. This will prompt you for
your private key's pass phrase, and then store it in your
authentication agent (&man.ssh-agent.1;). If you no longer
wish to have your key stored in the agent, issuing
<command>ssh-add -d</command> will remove it.</para>
<para>Test by doing something such as <command>ssh
freefall.FreeBSD.org ls /usr</command>.</para>
<para>For more information, see
<filename role="package">security/openssh</filename>,
&man.ssh.1;, &man.ssh-add.1;, &man.ssh-agent.1;,
&man.ssh-keygen.1;, and &man.scp.1;.</para>
</sect1>
<sect1 id="coverity">
<title>&coverity.prevent; Availability for &os; Committers</title>
<para>In January 2006, the &os;&nbsp;Foundation obtained a license
for &coverity.prevent; from &coverity;&nbsp;Ltd. With this
donation, all &os; developers can obtain access to
<application>Coverity Prevent</application> analysis results of
all &os; Project software.</para>
<para>&os; developers who are interested in obtaining access to
the analysis results of the automated
<application>Coverity Prevent</application> runs, can find out
more by logging into <hostid>freefall</hostid> and reading the
relevant bits of the files:</para>
<variablelist>
<varlistentry>
<term><filename>/usr/local/coverity/coverity_license.txt</filename></term>
<listitem>
<para>The license terms to which the &os; developers will
have to agree in order to use &coverity.prevent; analysis
results.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><filename>/usr/local/coverity/coverity_announcement.txt</filename></term>
<listitem>
<para>The announcement posted to the developers' mailing
list of the &os; Project. It contains useful information
about the &os; Foundation and &coverity;&nbsp;Ltd., as
well as signup information for registering with the
&coverity.prevent; installation of the &os;
Cluster.</para>
<para>After reading and understanding the license terms
of <filename>coverity_license.txt</filename>, all &os;
developers who are interested in using the analysis
results of &coverity.prevent; should read this
file.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><filename>/usr/local/coverity/coverity_readme.txt</filename></term>
<listitem>
<para>A short guide about fixes which are committed to the
&os; source tree after being detected by
&coverity.prevent; and analyzed by a &os;
developer.</para>
</listitem>
</varlistentry>
</variablelist>
<para>The &os; Wiki includes a mini-guide for developers who are
interested in working with the &coverity.prevent; analysis
reports:
<ulink url="http://wiki.freebsd.org/CoverityPrevent"></ulink>.
Please note that this mini-guide is only readable by &os;
developers, so if you cannot access this page, you will have to
ask someone to add you to the appropriate Wiki access
list.</para>
<para>Finally, all &os; developers who are going to use
&coverity.prevent; are always encouraged to ask for more details
and usage information, by posting any questions to the mailing
list of the &os; developers.</para>
</sect1>
<sect1 id="rules">
<title>The &os; Committers' Big List of Rules</title>
<orderedlist>
<listitem>
<para>Respect other committers.</para>
</listitem>
<listitem>
<para>Respect other contributors.</para>
</listitem>
<listitem>
<para>Discuss any significant change
<emphasis>before</emphasis> committing.</para>
</listitem>
<listitem>
<para>Respect existing maintainers (if listed in the
<makevar>MAINTAINER</makevar> field in
<filename>Makefile</filename> or in the
<filename>MAINTAINER</filename> file in the top-level
directory).</para>
</listitem>
<listitem>
<para>Any disputed change must be backed out pending
resolution of the dispute if requested by a maintainer.
Security related changes may
override a maintainer's wishes at the Security Officer's
discretion.</para>
</listitem>
<listitem>
<para>Changes go to &os.current; before
&os.stable; unless specifically permitted by
the release engineer or unless they are not applicable to
&os.current;. Any non-trivial or non-urgent
change which is applicable should also be allowed to sit in
&os.current; for at least 3 days before
merging so that it can be given sufficient testing. The
release engineer has the same authority over the
&os.stable; branch as outlined for the
maintainer in rule #5.</para>
</listitem>
<listitem>
<para>Do not fight in public with other committers; it looks
bad. If you must <quote>strongly disagree</quote> about
something, do so only in private.</para>
</listitem>
<listitem>
<para>Respect all code freezes and read the
<literal>committers</literal> and
<literal>developers</literal> mailing lists in a timely
manner so you know when a code freeze is in effect.</para>
</listitem>
<listitem>
<para>When in doubt on any procedure, ask first!</para>
</listitem>
<listitem>
<para>Test your changes before committing them.</para>
</listitem>
<listitem>
<para>Do not commit to anything under the
<filename>src/contrib</filename>,
<filename>src/crypto</filename>, or
<filename>src/sys/contrib</filename> trees without
<emphasis>explicit</emphasis> approval from the respective
maintainer(s).</para>
</listitem>
</orderedlist>
<para>As noted, breaking some of these rules can be grounds for
suspension or, upon repeated offense, permanent removal of
commit privileges. Individual members of core have the power to
temporarily suspend commit privileges until core as a whole has
the chance to review the issue. In case of an
<quote>emergency</quote> (a committer doing damage to the
repository), a temporary suspension may also be done by the
repository meisters. Only a 2/3 majority of core has the
authority to suspend commit privileges for longer than a week or
to remove them permanently. This rule does not exist to set
core up as a bunch of cruel dictators who can dispose of
committers as casually as empty soda cans, but to give the
project a kind of safety fuse. If someone is out of control, it
is important to be able to deal with this immediately rather
than be paralyzed by debate. In all cases, a committer whose
privileges are suspended or revoked is entitled to a
<quote>hearing</quote> by core, the total duration of the
suspension being determined at that time. A committer whose
privileges are suspended may also request a review of the
decision after 30 days and every 30 days thereafter (unless the
total suspension period is less than 30 days). A committer
whose privileges have been revoked entirely may request a review
after a period of 6 months has elapsed. This review policy is
<emphasis>strictly informal</emphasis> and, in all cases, core
reserves the right to either act on or disregard requests for
review if they feel their original decision to be the right
one.</para>
<para>In all other aspects of project operation, core is a subset
of committers and is bound by the
<emphasis>same rules</emphasis>. Just because someone is in
core this does not mean that they have special dispensation to
step outside any of the lines painted here; core's
<quote>special powers</quote> only kick in when it acts as a
group, not on an individual basis. As individuals, the core
team members are all committers first and core second.</para>
<sect2>
<title>Details</title>
<orderedlist>
<listitem id="respect">
<para>Respect other committers.</para>
<para>This means that you need to treat other committers as
the peer-group developers that they are. Despite our
occasional attempts to prove the contrary, one does not
get to be a committer by being stupid and nothing rankles
more than being treated that way by one of your peers.
Whether we always feel respect for one another or not (and
everyone has off days), we still have to
<emphasis>treat</emphasis> other committers with respect
at all times, on public forums and in private
email.</para>
<para>Being able to work together long term is this
project's greatest asset, one far more important than any
set of changes to the code, and turning arguments about
code into issues that affect our long-term ability to work
harmoniously together is just not worth the trade-off by
any conceivable stretch of the imagination.</para>
<para>To comply with this rule, do not send email when you
are angry or otherwise behave in a manner which is likely
to strike others as needlessly confrontational. First
calm down, then think about how to communicate in the most
effective fashion for convincing the other person(s) that
your side of the argument is correct, do not just blow off
some steam so you can feel better in the short term at the
cost of a long-term flame war. Not only is this very bad
<quote>energy economics</quote>, but repeated displays of
public aggression which impair our ability to work well
together will be dealt with severely by the project
leadership and may result in suspension or termination of
your commit privileges. The project leadership will take
into account both public and private communications
brought before it. It will not seek the disclosure of
private communications, but it will take it into account
if it is volunteered by the committers involved in the
complaint.</para>
<para>All of this is never an option which the project's
leadership enjoys in the slightest, but unity comes first.
No amount of code or good advice is worth trading that
away.</para>
</listitem>
<listitem>
<para>Respect other contributors.</para>
<para>You were not always a committer. At one time you were
a contributor. Remember that at all times. Remember what
it was like trying to get help and attention. Do not
forget that your work as a contributor was very important
to you. Remember what it was like. Do not discourage,
belittle, or demean contributors. Treat them with
respect. They are our committers in waiting. They are
every bit as important to the project as committers.
Their contributions are as valid and as important as your
own. After all, you made many contributions before you
became a committer. Always remember that.</para>
<para>Consider the points raised under
<xref linkend="respect"/> and apply them also to
contributors.</para>
</listitem>
<listitem>
<para>Discuss any significant change
<emphasis>before</emphasis> committing.</para>
<para>The repository is not where changes should be
initially submitted for correctness or argued over, that
should happen first in the mailing lists and the commit
should only happen once something resembling consensus has
been reached. This does not mean that you have to ask
permission before correcting every obvious syntax error or
manual page misspelling, simply that you should try to
develop a feel for when a proposed change is not quite
such a no-brainer and requires some feedback first.
People really do not mind sweeping changes if the result
is something clearly better than what they had before,
they just do not like being <emphasis>surprized</emphasis>
by those changes. The very best way of making sure that
you are on the right track is to have your code reviewed
by one or more other committers.</para>
<para>When in doubt, ask for review!</para>
</listitem>
<listitem>
<para>Respect existing maintainers if listed.</para>
<para>Many parts of &os; are not <quote>owned</quote> in
the sense that any specific individual will jump up and
yell if you commit a change to <quote>their</quote> area,
but it still pays to check first. One convention we use
is to put a maintainer line in the
<filename>Makefile</filename> for any package or subtree
which is being actively maintained by one or more people;
see <ulink
url="&url.books.developers-handbook;/policies.html">http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/developers-handbook/policies.html</ulink>
for documentation on this. Where sections of code have
several maintainers, commits to affected areas by one
maintainer need to be reviewed by at least one other
maintainer. In cases where the
<quote>maintainer-ship</quote> of something is not clear,
you can also look at the repository logs for the file(s)
in question and see if someone has been working recently
or predominantly in that area.</para>
<para>Other areas of &os; fall under the control of
someone who manages an overall category of &os;
evolution, such as internationalization or networking.
See <ulink
url="&url.base;/administration.html">http://www.FreeBSD.org/administration.html</ulink>
for more information on this.</para>
</listitem>
<listitem>
<para>Any disputed change must be backed out pending
resolution of the dispute if requested by a maintainer.
Security related changes may
override a maintainer's wishes at the Security Officer's
discretion.</para>
<para>This may be hard to swallow in times of conflict (when
each side is convinced that they are in the right, of
course) but a version control system makes it unnecessary
to have an ongoing dispute raging when it is far easier to
simply reverse the disputed change, get everyone calmed
down again and then try to figure out what is the best way
to proceed. If the change turns out to be the best thing
after all, it can be easily brought back. If it turns out
not to be, then the users did not have to live with the
bogus change in the tree while everyone was busily
debating its merits. People <emphasis>very</emphasis>
rarely call for back-outs in the repository since
discussion generally exposes bad or controversial changes
before the commit even happens, but on such rare occasions
the back-out should be done without argument so that we
can get immediately on to the topic of figuring out
whether it was bogus or not.</para>
</listitem>
<listitem>
<para>Changes go to &os.current; before &os.stable; unless
specifically permitted by the release engineer or unless
they are not applicable to &os.current;. Any non-trivial
or non-urgent change which is applicable should also be
allowed to sit in &os.current; for at least 3 days before
merging so that it can be given sufficient testing. The
release engineer has the same authority over the
&os.stable; branch as outlined in rule #5.</para>
<para>This is another <quote>do not argue about it</quote>
issue since it is the release engineer who is ultimately
responsible (and gets beaten up) if a change turns out to
be bad. Please respect this and give the release engineer
your full cooperation when it comes to the &os.stable;
branch. The management of &os.stable; may frequently seem
to be overly conservative to the casual observer, but also
bear in mind the fact that conservatism is supposed to be
the hallmark of &os.stable; and different rules apply
there than in &os.current;. There is also really no point
in having &os.current; be a testing ground if changes are
merged over to &os.stable; immediately. Changes need a
chance to be tested by the &os.current; developers, so
allow some time to elapse before merging unless the
&os.stable; fix is critical, time sensitive or so obvious
as to make further testing unnecessary (spelling fixes to
manual pages, obvious bug/typo fixes, etc.) In other
words, apply common sense.</para>
<para>Changes to the security branches (for example,
<literal>RELENG_7_0</literal>) must be approved by a
member of the &a.security-officer;, or in some cases, by a
member of the &a.re;.</para>
</listitem>
<listitem>
<para>Do not fight in public with other committers; it looks
bad. If you must <quote>strongly disagree</quote> about
something, do so only in private.</para>
<para>This project has a public image to uphold and that
image is very important to all of us, especially if we are
to continue to attract new members. There will be
occasions when, despite everyone's very best attempts at
self-control, tempers are lost and angry words are
exchanged. The best thing that can be done in such cases
is to minimize the effects of this until everyone has
cooled back down. That means that you should not air your
angry words in public and you should not forward private
correspondence to public mailing lists or aliases. What
people say one-to-one is often much less sugar-coated than
what they would say in public, and such communications
therefore have no place there - they only serve to inflame
an already bad situation. If the person sending you a
flame-o-gram at least had the grace to send it privately,
then have the grace to keep it private yourself. If you
feel you are being unfairly treated by another developer,
and it is causing you anguish, bring the matter up with
core rather than taking it public. Core will do its best
to play peace makers and get things back to sanity. In
cases where the dispute involves a change to the codebase
and the participants do not appear to be reaching an
amicable agreement, core may appoint a mutually-agreeable
3rd party to resolve the dispute. All parties involved
must then agree to be bound by the decision reached by
this 3rd party.</para>
</listitem>
<listitem>
<para>Respect all code freezes and read the
<literal>committers</literal> and
<literal>developers</literal> mailing list on a timely
basis so you know when a code freeze is in effect.</para>
<para>Committing unapproved changes during a code freeze is
a really big mistake and committers are expected to keep
up-to-date on what is going on before jumping in after a
long absence and committing 10 megabytes worth of
accumulated stuff. People who abuse this on a regular
basis will have their commit privileges suspended until
they get back from the &os; Happy Reeducation Camp we
run in Greenland.</para>
</listitem>
<listitem>
<para>When in doubt on any procedure, ask first!</para>
<para>Many mistakes are made because someone is in a hurry
and just assumes they know the right way of doing
something. If you have not done it before, chances are
good that you do not actually know the way we do things
and really need to ask first or you are going to
completely embarrass yourself in public. There is no
shame in asking
<quote>how in the heck do I do this?</quote> We already
know you are an intelligent person; otherwise, you would
not be a committer.</para>
</listitem>
<listitem>
<para>Test your changes before committing them.</para>
<!-- XXX Needs update re sparc64 + pc98
Also, needs more details on which machines are available for testing
-->
<para>This may sound obvious, but if it really were so
obvious then we probably would not see so many cases of
people clearly not doing this. If your changes are to the
kernel, make sure you can still compile both GENERIC and
LINT. If your changes are anywhere else, make sure you
can still make world. If your changes are to a branch,
make sure your testing occurs with a machine which is
running that code. If you have a change which also may
break another architecture, be sure and test on all
supported architectures. Please refer to the
<ulink url="http://www.FreeBSD.org/internal/">&os;
Internal Page</ulink> for a list of available resources.
As other architectures are added to the &os; supported
platforms list, the appropriate shared testing resources
will be made available.</para>
</listitem>
<listitem>
<para>Do not commit to anything under the
<filename>src/contrib</filename>,
<filename>src/crypto</filename>, and
<filename>src/sys/contrib</filename> trees without
<emphasis>explicit</emphasis> approval from the respective
maintainer(s).</para>
<para>The trees mentioned above are for contributed software
usually imported onto a vendor branch. Committing
something there, even if it does not take the file off the
vendor branch, may cause unnecessary headaches for those
responsible for maintaining that particular piece of
software. Thus, unless you have
<emphasis>explicit</emphasis> approval from the maintainer
(or you are the maintainer), do <emphasis>not</emphasis>
commit there!</para>
<para>Please note that this does not mean you should not try
to improve the software in question; you are still more
than welcome to do so. Ideally, you should submit your
patches to the vendor. If your changes are
&os;-specific, talk to the maintainer; they may be
willing to apply them locally. But whatever you do, do
<emphasis>not</emphasis> commit there by yourself!</para>
<para>Contact the &a.core; if you wish to take up
maintainership of an unmaintained part of the tree.</para>
</listitem>
</orderedlist>
</sect2>
<sect2>
<title>Policy on Multiple Architectures</title>
<para>&os; has added several new architecture ports during
recent release cycles and is truly no longer an &i386; centric
operating system. In an effort to make it easier to keep
&os; portable across the platforms we support, core has
developed the following mandate:</para>
<blockquote>
<para>Our 32-bit reference platform is &arch.i386;, and our
64-bit reference platform is &arch.sparc64;. Major design
work (including major API and ABI changes) must prove
itself on at least one 32-bit and at least one 64-bit
platform, preferably the primary reference platforms,
before it may be committed to the source tree.</para>
</blockquote>
<para>The &arch.i386; and &arch.sparc64; platforms were chosen
due to being more readily available to developers and as
representatives of more diverse processor and system designs -
big versus little endian, register file versus register stack,
different DMA and cache implementations, hardware page tables
versus software TLB management etc.</para>
<para>The &arch.ia64; platform has many of the same
complications that &arch.sparc64; has, but is still limited in
availability to developers.</para>
<para>We will continue to re-evaluate this policy as cost and
availability of the 64-bit platforms change.</para>
<para>Developers should also be aware of our Tier Policy for
the long term support of hardware architectures. The rules
here are intended to provide guidance during the development
process, and are distinct from the requirements for features
and architectures listed in that section. The Tier rules for
feature support on architectures at release-time are more
strict than the rules for changes during the development
process.</para>
</sect2>
<sect2>
<title>Other Suggestions</title>
<para>When committing documentation changes, use a spell checker
before committing. For all SGML docs, you should also
verify that your formatting directives are correct by running
<command>make lint</command>.</para>
<para>For all on-line manual pages, run <command>manck</command>
(from ports) over the manual page to verify all of the cross
references and file references are correct and that the man
page has all of the appropriate <makevar>MLINK</makevar>s
installed.</para>
<para>Do not mix style fixes with new functionality. A style
fix is any change which does not modify the functionality of
the code. Mixing the changes obfuscates the functionality
change when asking for differences between revisions, which
can hide any new bugs. Do not include whitespace changes with
content changes in commits to <filename>doc/</filename> or
<filename>www/</filename>. The extra clutter in the diffs
makes the translators' job much more difficult. Instead, make
any style or whitespace changes in separate commits that are
clearly labeled as such in the commit message.</para>
</sect2>
<sect2>
<title>Deprecating Features</title>
<para>When it is necessary to remove functionality from software
in the base system the following guidelines should be followed
whenever possible:</para>
<orderedlist>
<listitem>
<para>Mention is made in the manual page and possibly the
release notes that the option, utility, or interface is
deprecated. Use of the deprecated feature generates a
warning.</para>
</listitem>
<listitem>
<para>The option, utility, or interface is preserved until
the next major (point zero) release.</para>
</listitem>
<listitem>
<para>The option, utility, or interface is removed and no
longer documented. It is now obsolete. It is also
generally a good idea to note its removal in the release
notes.</para>
</listitem>
</orderedlist>
</sect2>
</sect1>
<sect1 id="archs">
<title>Support for Multiple Architectures</title>
<para>&os; is a highly portable operating system intended to
function on many different types of hardware architectures.
Maintaining clean separation of Machine Dependent (MD) and
Machine Independent (MI) code, as well as minimizing MD code, is
an important part of our strategy to remain agile with regards
to current hardware trends. Each new hardware architecture
supported by &os; adds substantially to the cost of code
maintenance, toolchain support, and release engineering. It
also dramatically increases the cost of effective testing of
kernel changes. As such, there is strong motivation to
differentiate between classes of support for various
architectures while remaining strong in a few key architectures
that are seen as the &os; <quote>target audience</quote>.</para>
<sect2>
<title>Statement of General Intent</title>
<para>The &os; Project targets "production quality commercial
off-the-shelf (COTS) workstation, server, and high-end
embedded systems". By retaining a focus on a narrow set of
architectures of interest in these environments, the &os;
Project is able to maintain high levels of quality, stability,
and performance, as well as minimize the load on various
support teams on the project, such as the ports team,
documentation team, security officer, and release engineering
teams. Diversity in hardware support broadens the options for
&os; consumers by offering new features and usage
opportunities (such as support for 64-bit CPUs, use in
embedded environments, etc.), but these benefits must always
be carefully considered in terms of the real-world maintenance
cost associated with additional platform support.</para>
<para>The &os; Project differentiates platform targets into
four tiers. Each tier includes a specification of the
requirements for an architecture to be in that tier,
as well as specifying the obligations of developers with
regards to the platform. In addition, a policy is defined
regarding the circumstances required to change the tier
of an architecture.</para>
</sect2>
<sect2>
<title>Tier 1: Fully Supported Architectures</title>
<para>Tier 1 platforms are fully supported by the security
officer, release engineering, and toolchain maintenance staff.
New features added to the operating system must be fully
functional across all Tier 1 architectures for every release
(features which are inherently architecture-specific, such as
support for hardware device drivers, may be exempt from this
requirement). In general, all Tier 1 platforms must have
build and Tinderbox support either in the FreeBSD.org cluster,
or be easily available for all developers. Embedded platforms
may substitute an emulator available in the &os; cluster
for actual hardware.</para>
<para>Tier 1 architectures are expected to be Production Quality
with respects to all aspects of the &os; operating system,
including installation and development environments.</para>
<para>Tier 1 architectures are expected to be completely
integrated into the source tree and have all features
necessary to produce an entire system relevant for that target
architecture. Tier 1 architectures generally have at least 6
active developers.</para>
<para>Tier 1 architectures are expected to be fully supported by
the ports system. All the ports should build on a Tier 1
platform, or have the appropriate filters to prevent the
inappropriate ones from building there. The packaging system
must support all Tier 1 architectures. To ensure an
architecture's Tier 1 status, proponents of that architecture
must show that all relevant packages can be built on that
platform.</para>
<para>Tier 1 embedded architectures must be able to cross-build
packages on at least one other Tier 1 architecture. The
packages must be the most relevant for the platform, but may
be a non-empty subset of those that build natively.</para>
<para>Tier 1 architectures must be fully documented. All basic
operations need to be covered by the handbook or other
documents. All relevant integration documentation must also
be integrated into the tree, or readily available.</para>
<para>Current Tier 1 platforms are &arch.i386; and
&arch.amd64;.</para>
</sect2>
<sect2>
<title>Tier 2: Developmental Architectures</title>
<para>Tier 2 platforms are not supported by the security officer
and release engineering teams. Platform maintainers are
responsible for toolchain support in the tree. The toolchain
maintainer is expected to work with the platform maintainers
to refine these changes. Major new toolchain components are
allowed to break support for Tier 2 architectures if the
&os;-local changes have not been incorporated upstream.
The toolchain maintainers are expected to provide prompt
review of any proposed changes and cannot block, through their
inaction, changes going into the tree. New features added to
&os; should be feasible to implement on these platforms,
but an implementation is not required before the feature may
be added to the &os; source tree. New features that may be
difficult to implement on Tier 2 architectures should provide
a means of disabling them on those architectures. The
implementation of a Tier 2 architecture may be committed to
the main &os; tree as long as it does not interfere with
production work on Tier 1 platforms, or substantially with
other Tier 2 platforms. Before a Tier 2 platform can be added
to the &os; base source tree, the platform must be able to
boot multi-user on actual hardware. Generally, there must be
at least three active developers working on the
platform.</para>
<para>Tier 2 architectures are usually systems targeted at Tier
1 support, but that are still under development.
Architectures reaching end of life may also be moved from Tier
1 status to Tier 2 status as the availability of resources to
continue to maintain the system in a Production Quality state
diminishes. Well supported niche architectures may also be
Tier 2.</para>
<para>Tier 2 architectures may have some support for them
integrated into the ports infrastructure. They may have cross
compilation support added, at the discretion of portmgr. Some
ports must built natively into packages if the package system
supports that architecture. If not integrated into the base
system, some external patches for the architecture for ports
must be available.</para>
<para>Tier 2 architectures can be integrated into the &os;
handbook. The basics for how to get a system running must be
documented, although not necessarily for every single board or
system a Tier 2 architecture supports. The supported hardware
list must exist and should be no more than a couple of months
old. It should be integrated into the &os;
documentation.</para>
<para>Current Tier 2 platforms are &arch.arm;, &arch.ia64;,
&arch.pc98;, &arch.powerpc;, and &arch.sparc64;.</para>
</sect2>
<sect2>
<title>Tier 3: Experimental Architectures</title>
<para>Tier 3 platforms are not supported by the security officer
and release engineering teams. At the discretion of the
toolchain maintainer, they may be supported in the toolchain.
Tier 3 platforms are architectures in the early stages of
development, for non-mainstream hardware platforms, or which
are considered legacy systems unlikely to see broad future
use. New Tier 3 systems will not be committed to the base
source tree. Support for Tier 3 systems may be worked on in
the &os; Perforce Repository, providing source control and
easier change integration from the main &os; tree.
Platforms that transition to Tier 3 status may be removed from
the tree if they are no longer actively supported by the
&os; developer community at the discretion of the release
engineer.</para>
<para>Tier 3 platforms may have ports support, either integrated
or external, but do not require it.</para>
<para>Tier 3 platforms must have the basics documented for how
to build a kernel and how to boot it on at least one target
hardware or emulation environment. This documentation need
not be integrated into the &os; tree.</para>
<para>Current Tier 3 platforms are &arch.mips; and
&s390;.</para>
</sect2>
<sect2>
<title>Tier 4: Unsupported Architectures</title>
<para>Tier 4 systems are not supported in any form by the
project.</para>
<para>All systems not otherwise classified into a support tier
are Tier 4 systems.</para>
</sect2>
<sect2>
<title>Policy on Changing the Tier of an Architecture</title>
<para>Systems may only be moved from one tier to another by
approval of the &os; Core Team, which shall make that
decision in collaboration with the Security Officer, Release
Engineering, and toolchain maintenance teams.</para>
</sect2>
</sect1>
<sect1 id="ports">
<title>Ports Specific FAQ</title>
<qandaset>
<qandadiv>
<title>Adding a New Port</title>
<qandaentry>
<question>
<para>How do I add a new port?</para>
</question>
<answer>
<para>First, please read the section about repository
copies.</para>
<para>The easiest way to add a new port is to use the
<command>addport</command> script from your machine
(located in the <filename>ports/Tools/scripts</filename>
directory). It will add a port from the directory you
specify, determining the category automatically from the
port <filename>Makefile</filename>. It will also add an
entry to the port's category
<filename>Makefile</filename>. It was written by
&a.mharo.email;, &a.will.email;, and &a.garga.email;.
When sending
questions about this script to the &a.ports;, please
also CC &a.crees.email;, the current maintainer.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>Any other things I need to know when I add a new
port?</para>
</question>
<answer>
<para>Check the port, preferably to make sure it compiles
and packages correctly. This is the recommended
sequence:</para>
<screen>&prompt.root; <userinput>make install</userinput>
&prompt.root; <userinput>make package</userinput>
&prompt.root; <userinput>make deinstall</userinput>
&prompt.root; <userinput>pkg_add <replaceable>package you built above</replaceable></userinput>
&prompt.root; <userinput>make deinstall</userinput>
&prompt.root; <userinput>make reinstall</userinput>
&prompt.root; <userinput>make package</userinput></screen>
<para>The <ulink
url="&url.books.porters-handbook;/index.html">Porters
Handbook</ulink> contains more detailed
instructions.</para>
<para>Use &man.portlint.1; to check the syntax of the
port. You do not necessarily have to eliminate all
warnings but make sure you have fixed the simple
ones.</para>
<para>If the port came from a submitter who has not
contributed to the Project before, add that person's
name to the <ulink
url="&url.articles.contributors;/contrib-additional.html">Additional
Contributors</ulink> section of the &os;
Contributors List.</para>
<para>Close the PR if the port came in as a PR. To close
a PR, just do <userinput>edit-pr
<replaceable>PR#</replaceable></userinput> on
<hostid>freefall</hostid> and change the
<varname>state</varname> from <constant>open</constant>
to <constant>closed</constant>. You will be asked to
enter a log message and then you are done.</para>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Removing an Existing Port</title>
<qandaentry>
<question>
<para>How do I remove an existing port?</para>
</question>
<answer>
<para>First, please read the section about repository
copies. Before you remove the port, you have to verify
there are no other ports depending on it.</para>
<itemizedlist>
<listitem>
<para>Make sure there is no dependency on the port
in the ports collection:</para>
<itemizedlist>
<listitem>
<para>The port's PKGNAME should appear in exactly
one line in a recent INDEX file.</para>
</listitem>
<listitem>
<para>No other ports should contain any reference
to the port's directory or PKGNAME in their
Makefiles</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>Then, remove the port:</para>
<procedure>
<step>
<para>Remove the port's files and directory with
<command>svn remove</command>.</para>
</step>
<step>
<para>Remove the <makevar>SUBDIR</makevar> listing
of the port in the parent directory
<filename>Makefile</filename>.</para>
</step>
<step>
<para>Add an entry to
<filename>ports/MOVED</filename>.</para>
</step>
<step>
<para>Remove the port from
<filename>ports/LEGAL</filename> if it is
there.</para>
</step>
</procedure>
</listitem>
</itemizedlist>
<para>Alternatively, you can use the
<command>rmport</command> script, from <filename
class="directory">ports/Tools/scripts</filename>.
This script was written by &a.vd.email;. When sending
questions about this script to the &a.ports;, please
also CC &a.crees.email;, the current maintainer.</para>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Re-adding a Deleted Port</title>
<qandaentry>
<question>
<para>How do I re-add a deleted port?</para>
</question>
<answer>
<para>This is essentially the reverse of deleting a
port.</para>
<procedure>
<step>
<para>Figure out when the port was removed. Use this
<ulink
url="http://people.freebsd.org/~crees/removed_ports/index.xml">list</ulink>
and then copy the last living revision of the port:
<screen>&prompt.user; <userinput>cd /usr/ports/<replaceable>category
</replaceable></userinput>
&prompt.user; <userinput>svn cp 'svn+ssh://svn.freebsd.org/ports/<replaceable>category</replaceable>/<replaceable>portname</replaceable>/@{<replaceable>YYYY-MM-DD</replaceable>}' <replaceable>portname</replaceable>
</userinput></screen>
Pick a date that is before the removal but after the
last true commit.</para>
</step>
<step>
<para>Perform whatever changes are necessary to make
the port work again. If it was deleted because the
distfiles are no longer available you will need to
volunteer to host them yourself, or find someone
else to do so.</para>
</step>
<step>
<para><command>svn add</command> or
<command>svn remove</command> any appropriate
files.</para>
</step>
<step>
<para>Restore the <makevar>SUBDIR</makevar> listing of
the port in the parent directory
<filename>Makefile</filename>, and delete the entry
from <filename>ports/MOVED</filename>.</para>
</step>
<step>
<para>If the port had an entry in
<filename>ports/LEGAL</filename>, restore it.</para>
</step>
<step>
<para><command>svn commit</command> these changes,
preferably in one step.</para>
</step>
</procedure>
<tip>
<para><command>addport</command> now detects when the
port to add has previously existed, and should handle
all except the <filename>ports/LEGAL</filename> step
automatically.</para>
</tip>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Repository Copies</title>
<qandaentry>
<question>
<para>When do we need a repository copy?</para>
</question>
<answer>
<para>When you want to add a port that is related to
any port that is already in the tree in a separate
directory, you have to do a repository copy.
Here <wordasword>related</wordasword> means
it is a different version or a slightly modified
version. Examples are
<filename>print/ghostscript*</filename> (different
versions) and <filename>x11-wm/windowmaker*</filename>
(English-only and internationalized version).</para>
<para>Another example is when a port is moved from one
subdirectory to another, or when you want to change the
name of a directory because the author(s) renamed their
software even though it is a
descendant of a port already in a tree.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What do I need to do?</para>
</question>
<answer>
<para>With Subversion, a repo copy can be done by any
committer:</para>
<itemizedlist>
<listitem>
<para>Doing a repo copy:</para>
<procedure>
<step>
<para>First make sure that you were using an up to
date ports tree and the target directory does
not exist.</para>
</step>
<step>
<para>Use <command>svn move</command> or
<command>svn copy</command> to do the repo
copy.</para>
</step>
<step>
<para>Upgrade the copied port to the new version.
Remember to change the
<makevar>LATEST_LINK</makevar> so there are no
duplicate ports with the same name. In some
rare cases it may be necessary to change the
<makevar>PORTNAME</makevar> instead of
<makevar>LATEST_LINK</makevar>, but this should
only be done when it is really needed &mdash;
e.g., using an existing port as the base for a
very similar program with a different name, or
upgrading a port to a new upstream version which
actually changes the distribution name, like the
transition from
<filename>textproc/libxml</filename> to
<filename>textproc/libxml2</filename>. In most
cases, changing <makevar>LATEST_LINK</makevar>
should suffice.</para>
</step>
<step>
<para>Add the new subdirectory to the
<makevar>SUBDIR</makevar> listing in the parent
directory <filename>Makefile</filename>. You
can run <command>make checksubdirs</command> in
the parent directory to check this.</para>
</step>
<step>
<para>If the port changed categories, modify the
<makevar>CATEGORIES</makevar> line of the port's
<filename>Makefile</filename> accordingly</para>
</step>
<step>
<para>Add an entry to
<filename>ports/MOVED</filename>, if you remove
the original port.</para>
</step>
<step>
<para>Commit all changes on one commit. A forced
commit is no longer needed with
Subversion.</para>
</step>
</procedure>
</listitem>
<listitem>
<para>When removing a port:</para>
<procedure>
<step>
<para>Perform a thorough check of the ports
collection for any dependencies on the old port
location/name, and update them. Running
<command>grep</command> on
<filename>INDEX</filename> is not enough because
some ports have dependencies enabled by
compile-time options. A full
<command>grep -r</command> of the ports
collection is recommended.</para>
</step>
<step>
<para>Remove the old port and the
old <makevar>SUBDIR</makevar> entry.</para>
</step>
<step>
<para>Add an entry to
<filename>ports/MOVED</filename>.</para>
</step>
</procedure>
</listitem>
<listitem>
<para>After repo moves (<quote>rename</quote>
operations where a port is copied and the old
location is removed):</para>
<procedure>
<step>
<para>Follow the same steps that are outlined in
the previous two entries, to activate the new
location of the port and remove the old
one.</para>
</step>
</procedure>
</listitem>
</itemizedlist>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Ports Freeze</title>
<qandaentry>
<question>
<para>What is a <quote>ports freeze</quote>?</para>
</question>
<answer>
<para>Before a release, it is necessary to restrict
commits to the ports tree for a short period of time
while the packages and the release itself are being
built. This is to ensure consistency among the various
parts of the release, and is called the
<quote>ports freeze</quote>.</para>
<para>For more information on the background and
policies surrounding a ports freeze, see the
<ulink url="&url.base;/portmgr/qa.html">Portmgr
Quality Assurance page</ulink>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What is a <quote>ports slush</quote> or
<quote>feature freeze</quote>?</para>
</question>
<answer>
<para>During a release cycle the ports tree may be in a
<quote>slush</quote> state instead of in a hard freeze.
The goal during a slush is to reach a stable ports tree
to avoid rebuilding large sets of packages for the
release and to tag the tree. During this time
<quote>sweeping changes</quote> are prohibited unless
specifically permitted by portmgr. Complete details
about what qualifies as a sweeping change can be found
on the <ulink
url="&url.base;/portmgr/implementation.html">Portmgr
Implementation page</ulink>.</para>
<para>The benefit of a slush as opposed to a complete
freeze is that it allows maintainers to continue adding
new ports, making routine version updates, and bug fixes
to most existing ports, as long as the number of
affected ports is minimal. For example, updating the
shared library version on a port that many other ports
depend on.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How long is a ports freeze or slush?</para>
</question>
<answer>
<para>A freeze only lasts long enough to tag the tree.
A slush usually lasts a week or two, but may last
longer.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What does it mean to me?</para>
</question>
<answer>
<para>During a ports freeze, you are not allowed to
commit anything to the tree without explicit approval
from the Ports Management Team. <quote>Explicit
approval</quote> here means that you send a patch to
the Ports Management Team for review and get a reply
saying, <quote>Go ahead and commit it.</quote></para>
<para>Not everything is allowed to be committed during
a freeze. Please see the <ulink
url="&url.base;/portmgr/qa.html">Portmgr Quality
Assurance page</ulink> for more information.</para>
<para>Note that you do not have implicit permission to fix
a port during the freeze just because it is
broken.</para>
<para>During a ports slush, you are still allowed to
commit but you must exercise more caution in what you
commit. Furthermore a special note (typically
<quote>Feature Safe: yes</quote>) must be added to the
commit message.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I know when the ports slush starts?</para>
</question>
<answer>
<para>The Ports Management Team will send out warning
messages to the &a.ports; and &a.committers;
announcing the start of the impending release, usually
two or three weeks in advance. The exact starting time
will not be determined until a few days before the
actual release. This is because the ports slush has to
be synchronized with the release, and it is usually not
known until then when exactly the release will be
rolled.</para>
<para>When the slush starts, there will be another
announcement to the &a.ports; and &a.committers;, of
course.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I know when the freeze or slush ends?</para>
</question>
<answer>
<para>A few hours after the release, the Ports Management
Team will send out a mail to the &a.ports; and
&a.committers; announcing the end of the ports freeze or
slush. Note that the release being cut does not
automatically indicate the end of the freeze. We have
to make sure there will be no last minute snafus that
result in an immediate re-rolling of the release.</para>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Creating a New Category</title>
<qandaentry>
<question>
<para>What is the procedure for creating a new
category?</para>
</question>
<answer>
<para>Please see <ulink
url="&url.books.porters-handbook;/makefile-categories.html#PROPOSING-CATEGORIES">
Proposing a New Category</ulink> in the Porter's
Handbook. Once that procedure has been followed and the
PR has been assigned to &a.portmgr;, it is their
decision whether or not to approve it. If they do, it
is their responsibility to do the following:</para>
<procedure>
<step>
<para>Perform any needed moves. (This only applies
to physical categories.)</para>
</step>
<step>
<para>Update the <makevar>VALID_CATEGORIES</makevar>
definition in
<filename>ports/Mk/bsd.port.mk</filename>.</para>
</step>
<step>
<para>Assign the PR back to you.</para>
</step>
</procedure>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What do I need to do to implement a new physical
category?</para>
</question>
<answer>
<procedure>
<step>
<para>Upgrade each moved port's
<filename>Makefile</filename>. Do not connect the
new category to the build yet.</para>
<para>To do this, you will need to:</para>
<procedure>
<step>
<para>Change the port's
<makevar>CATEGORIES</makevar> (this was the
point of the exercise, remember?) The new
category should be listed
<emphasis>first</emphasis>. This will help to
ensure that the <makevar>PKGORIGIN</makevar> is
correct.</para>
</step>
<step>
<para>Run a <command>make describe</command>.
Since the top-level
<command>make index</command> that you will be
running in a few steps is an iteration of
<command>make describe</command> over the entire
ports hierarchy, catching any errors here will
save you having to re-run that step later
on.</para>
</step>
<step>
<para>If you want to be really thorough, now
might be a good time to run
&man.portlint.1;.</para>
</step>
</procedure>
</step>
<step>
<para>Check that the <makevar>PKGORIGIN</makevar>s are
correct. The ports system uses each port's
<makevar>CATEGORIES</makevar> entry to create its
<makevar>PKGORIGIN</makevar>, which is used to
connect installed packages to the port directory
they were built from. If this entry is wrong,
common port tools like &man.pkg.version.1; and
&man.portupgrade.1; fail.</para>
<para>To do this, use the
<filename>chkorigin.sh</filename> tool, as follows:
<command>env
PORTSDIR=<replaceable>/path/to/ports</replaceable>
sh -e
<replaceable>/path/to/ports</replaceable>/Tools/scripts/chkorigin.sh</command>.
This will check <emphasis>every</emphasis> port in
the ports tree, even those not connected to the
build, so you can run it directly after the move
operation. Hint: do not forget to look at the
<makevar>PKGORIGIN</makevar>s of any slave ports of
the ports you just moved!</para>
</step>
<step>
<para>On your own local system, test the proposed
changes: first, comment out the
<makevar>SUBDIR</makevar> entries in the old ports'
categories' <filename>Makefile</filename>s; then
enable building the new category in
<filename>ports/Makefile</filename>. Run
<command>make checksubdirs</command> in the affected
category directories to check the
<makevar>SUBDIR</makevar> entries. Next, in the
<filename class="directory">ports/</filename>
directory, run <command>make index</command>. This
can take over 40 minutes on even modern systems;
however, it is a necessary step to prevent problems
for other people.</para>
</step>
<step>
<para>Once this is done, you can commit the updated
<filename>ports/Makefile</filename> to connect the
new category to the build and also commit the
<filename>Makefile</filename> changes for the old
category or categories.</para>
</step>
<step>
<para>Add appropriate entries to
<filename>ports/MOVED</filename>.</para>
</step>
<step>
<para>Update the documentation by modifying the
following:</para>
<itemizedlist>
<listitem>
<para>the <ulink
url="&url.books.porters-handbook;/makefile-categories.html#PORTING-CATEGORIES">list
of categories</ulink> in the Porter's
Handbook</para>
</listitem>
<listitem>
<para>
<filename>www/en/ports/categories</filename>.
Note that these are now displayed by sub-groups,
as specified in
<filename>www/en/ports/categories.descriptions</filename>.</para>
</listitem>
</itemizedlist>
<para>(Note: these are in the docs, not the ports,
repository). If you are not a docs committer, you
will need to submit a PR for this.</para>
</step>
<step>
<para>Only once all the above have been done, and no
one is any longer reporting problems with the new
ports, should the old ports be deleted from their
previous locations in the repository.</para>
</step>
</procedure>
<para>It is not necessary to manually update the
<ulink url="&url.base;/ports/index.html">ports web
pages</ulink> to reflect the new category. This is
now done automatically via your change to
<filename>www/en/ports/categories</filename> and the
daily automated rebuild of
<filename>INDEX</filename>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What do I need to do to implement a new virtual
category?</para>
</question>
<answer>
<para>This is much simpler than a physical category. You
only need to modify the following:</para>
<itemizedlist>
<listitem>
<para>the <ulink
url="&url.books.porters-handbook;/makefile-categories.html#PORTING-CATEGORIES">list
of categories</ulink> in the Porter's
Handbook</para>
</listitem>
<listitem>
<para><filename>www/en/ports/categories</filename></para>
</listitem>
</itemizedlist>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Miscellaneous Questions</title>
<qandaentry>
<question>
<para>How do I know if my port is building correctly or
not?</para>
</question>
<answer>
<para>First, go check <ulink
url="http://pointyhat.FreeBSD.org/errorlogs/"></ulink>.
There you will find error logs from the latest package
building runs on all supported platforms for the most
recent branches.</para>
<para>However, just because the port does not show up
there does not mean it is building correctly. (One of
the dependencies may have failed, for instance.) The
relevant directories are available on
<hostid>pointyhat</hostid> under <filename
class="directory">/a/portbuild/&lt;arch&gt;/&lt;major_version&gt;</filename>
so feel free to dig around. Each architecture and
version has the following subdirectories:</para>
<programlisting>errors error logs from latest &lt;major_version&gt; run on &lt;arch&gt;
logs all logs from latest &lt;major_version&gt; run on &lt;arch&gt;
packages packages from latest &lt;major_version&gt; run on &lt;arch&gt;
bak/errors error logs from last complete &lt;major_version&gt; run on &lt;arch&gt;
bak/logs all logs from last complete &lt;major_version&gt; run on &lt;arch&gt;
bak/packages packages from last complete &lt;major_version&gt; run on &lt;arch&gt;</programlisting>
<para>Basically, if the port shows up in
<filename>packages</filename>, or it is in
<filename>logs</filename> but not in
<filename>errors</filename>, it built fine. (The
<filename>errors</filename> directories are what you get
from the web page.)</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>I added a new port. Do I need to add it to the
<filename>INDEX</filename>?</para>
</question>
<answer>
<para>No, <filename>INDEX</filename> is no longer stored
in the SVN repository. The file can either be generated
by running <command>make index</command>, or a
pre-generated version can be downloaded with
<command>make fetchindex</command>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>Are there any other files I am not allowed to
touch?</para>
</question>
<answer>
<para>Any file directly under <filename>ports/</filename>,
or any file under a subdirectory that starts with an
uppercase letter (<filename>Mk/</filename>,
<filename>Tools/</filename>, etc.). In particular, the
Ports Management Team is very protective of
<filename>ports/Mk/bsd.port*.mk</filename> so do not
commit changes to those files unless you want to face
his wra(i)th.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What is the proper procedure for updating the
checksum for a port's distfile when the file changes
without a version change?</para>
</question>
<answer>
<para>When the checksum for a port's distfile is updated
due to the author updating the file without changing the
port's revision, the commit message should include a
summary of the relevant diffs between the original and
new distfile to ensure that the distfile has not been
corrupted or maliciously altered. If the current
version of the port has been in the ports tree for a
while, a copy of the old distfile will usually be
available on the ftp servers; otherwise the author or
maintainer should be contacted to find out why the
distfile has changed.</para>
</answer>
</qandaentry>
</qandadiv>
</qandaset>
</sect1>
<sect1 id="non-committers">
<title>Issues Specific to Developers Who Are Not
Committers</title>
<para>A few people who have access to the &os; machines do not
have commit bits. For instance, the project is willing to give
access to the GNATS database to contributors who have shown
interest and dedication in working on Problem Reports.</para>
<para>Almost all of this document will apply to these developers
as well (except things specific to commits and the mailing list
memberships that go with them). In particular, we recommend
that you read:</para>
<itemizedlist>
<listitem>
<para><link linkend="admin">Administrative
Details</link></para>
</listitem>
<listitem>
<para><link
linkend="conventions-everyone">Conventions</link></para>
<note>
<para>You should get your mentor to add you to the
<quote>Additional Contributors</quote>
(<filename>doc/en_US.ISO8859-1/articles/contributors/contrib.additional.xml</filename>),
if you are not already listed there.</para>
</note>
</listitem>
<listitem>
<para><link
linkend="developer.relations">Developer
Relations</link></para>
</listitem>
<listitem>
<para><link linkend="ssh.guide">SSH Quick-Start
Guide</link></para>
</listitem>
<listitem>
<para><link linkend="rules">The &os; Committers' Big List
of Rules</link></para>
</listitem>
</itemizedlist>
</sect1>
<sect1 id="google-analytics">
<title>Information About &ga;</title>
<para>As of December 12, 2012, &ga; was enabled on the
&os;&nbsp;Project website to collect anonymized usage statistics
regarding usage of the site. The information collected is
valuable to the &os;&nbsp;Documentation Project, in order to
identify various problems on the &os; website.</para>
<sect2 id="google-analytics-policy">
<title>&ga; General Policy</title>
<para>The &os;&nbsp;Project takes visitor privacy very
seriously. As such, the &os;&nbsp;Project website honors the
<quote>Do Not Track</quote> header <emphasis>before</emphasis>
fetching the tracking code from Google. For more information,
please see the
<ulink url="http://www.FreeBSD.org/privacy.html">&os; Privacy
Policy</ulink>.</para>
<para>&ga; access is <emphasis>not</emphasis> arbitrarily
allowed &mdash; access must be requested, voted on by the
&a.doceng;, and explicitly granted.</para>
<para>Requests for &ga; data must include a specific purpose.
For example, a valid reason for requesting access would be
<quote>to see the most frequently used web browsers when
viewing &os; web pages to ensure page rendering speeds are
acceptable.</quote></para>
<para>Conversely, <quote>to see what web browsers are most
frequently used</quote> (without stating
<emphasis>why</emphasis>) would be rejected.</para>
<para>All requests must include the timeframe for which the data
would be required. For example, it must be explicitly stated
if the requested data would be needed for a timeframe covering
a span of 3 weeks, or if the request would be one-time
only.</para>
<para>Any request for &ga; data without a clear, reasonable
reason beneficial to the &os;&nbsp;Project will be
rejected.</para>
</sect2>
<sect2 id="google-analytics-data">
<title>Data Available Through &ga;</title>
<para>A few examples of the types of &ga; data available
include:</para>
<itemizedlist>
<listitem>
<para>Commonly used web browsers</para>
</listitem>
<listitem>
<para>Page load times</para>
</listitem>
<listitem>
<para>Site access by language</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="perks">
<title>Perks of the Job</title>
<para>Unfortunately, there are not many perks involved with being
a committer. Recognition as a competent software engineer is
probably the only thing that will be of benefit in the long run.
However, there are at least some perks:</para>
<variablelist>
<varlistentry>
<term>Free 4-CD and DVD Sets</term>
<listitem>
<para>&os; committers can get a free 4-CD or DVD set at
conferences from <ulink url="http://www.freebsdmall.com">
&os; Mall, Inc.</ulink>. The sets are no longer
available as a subscription due to the high shipment costs
to countries outside the USA.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>Freenode IRC Cloaks</term>
<listitem>
<para>&os; developers may request a cloaked hostmask for
their account on the Freenode IRC network in the form of
<literal>freebsd/developer/</literal><replaceable>freefall
name</replaceable> or
<literal>freebsd/developer/</literal><replaceable>NickServ
name</replaceable>. To request a cloak, send an email to
&a.eadler.email; with your requested hostmask and NickServ
account name.</para>
</listitem>
</varlistentry>
</variablelist>
</sect1>
<sect1 id="misc">
<title>Miscellaneous Questions</title>
<qandaset>
<qandaentry>
<question>
<para>Why are trivial or cosmetic changes to files on a
vendor branch a bad idea?</para>
</question>
<answer>
<itemizedlist>
<listitem>
<para>From now on, every new vendor release of that file
will need to have patches merged in by hand.</para>
</listitem>
<listitem>
<para>From now on, every new vendor release of that file
will need to have patches
<emphasis>verified</emphasis> by hand.</para>
</listitem>
</itemizedlist>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I add a new file to a branch?</para>
</question>
<answer>
<para>To add a file onto a branch, simply checkout or update
to the branch you want to add to and then add the file
using the add operation as you normally would. This works
fine for the <literal>doc</literal> and
<literal>ports</literal> trees. The
<literal>src</literal> tree uses SVN and requires more
care because of the <literal>mergeinfo</literal>
properties. See section 1.4.6 of the <ulink
url="http://wiki.freebsd.org/SubversionPrimer">Subversion
Primer</ulink> for details. Refer to <ulink
url="http://wiki.freebsd.org/SubversionPrimer/Merging">SubversionPrimer/Merging</ulink>
for details on how to perform an MFC.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What <quote>meta</quote> information should I include
in a commit message?</para>
</question>
<answer>
<para>As well as including an informative message with each
commit you may need to include some additional information
as well.</para>
<para>This information consists of one or more lines
containing the key word or phrase, a colon, tabs for
formatting, and then the additional information.</para>
<para>The key words or phrases are:</para>
<informaltable frame="none" pgwide="1">
<tgroup cols="2">
<tbody>
<row>
<entry><literal>PR:</literal></entry>
<entry>The problem report (if any) which is affected
(typically, by being closed) by this
commit.</entry>
</row>
<row>
<entry><literal>Submitted by:</literal></entry>
<entry>The name and e-mail address of the person
that submitted the fix; for committers, just the
username on the &os; cluster.</entry>
</row>
<row>
<entry><literal>Reviewed by:</literal></entry>
<entry>The name and e-mail address of the person or
people that reviewed the change; for committers,
just the username on the &os; cluster. If a
patch was submitted to a mailing list for review,
and the review was favorable, then just include
the list name.</entry>
</row>
<row>
<entry><literal>Approved by:</literal></entry>
<entry>The name and e-mail address of the person or
people that approved the change; for committers,
just the username on the &os; cluster. It is
customary to get prior approval for a commit if it
is to an area of the tree to which you do not
usually commit. In addition, during the run up to
a new release all commits
<emphasis>must</emphasis> be approved by the
release engineering team. If these are your first
commits then you should have passed them past your
mentor first, and you should list your mentor, as
in ``<replaceable>username-of-mentor</replaceable>
<literal>(mentor)</literal>''.</entry>
</row>
<row>
<entry><literal>Obtained from:</literal></entry>
<entry>The name of the project (if any) from which
the code was obtained.</entry>
</row>
<row>
<entry><literal>MFC after:</literal></entry>
<entry>If you wish to receive an e-mail reminder to
<acronym>MFC</acronym> at a later date, specify
the number of days, weeks, or months after which
an <acronym>MFC</acronym> is planned.</entry>
</row>
<row>
<entry><literal>Security:</literal></entry>
<entry>If the change is related to a security
vulnerability or security exposure, include one or
more references or a description of the
issue.</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<example>
<title>Commit Log for a Commit Based on a PR</title>
<para>You want to commit a change based on a PR submitted
by John Smith containing a patch. The end of the commit
message should look something like this.</para>
<programlisting>...
PR: foo/12345
Submitted by: John Smith &lt;John.Smith@example.com></programlisting>
</example>
<example>
<title>Commit Log for a Commit Needing Review</title>
<para>You want to change the virtual memory system. You
have posted patches to the appropriate mailing list (in
this case, <literal>freebsd-arch</literal>) and the
changes have been approved.</para>
<programlisting>...
Reviewed by: -arch</programlisting>
</example>
<example>
<title>Commit Log for a Commit Needing Approval</title>
<para>You want to commit a change to a section of the tree
with a MAINTAINER assigned. You have collaborated with
the listed MAINTAINER, who has told you to go ahead and
commit.</para>
<programlisting>...
Approved by: <replaceable>abc</replaceable></programlisting>
<para>Where <replaceable>abc</replaceable> is the account
name of the person who approved.</para>
</example>
<example>
<title>Commit Log for a Commit Bringing in Code from
OpenBSD</title>
<para>You want to commit some code based on work done in
the OpenBSD project.</para>
<programlisting>...
Obtained from: OpenBSD</programlisting>
</example>
<example>
<title>Commit Log for a Change to &os.current; with a
Planned Commit to &os.stable; to Follow at a Later
Date.</title>
<para>You want to commit some code which will be merged
from &os.current; into the &os.stable; branch after two
weeks.</para>
<programlisting>...
MFC after: <replaceable>2 weeks</replaceable></programlisting>
<para>Where <replaceable>2</replaceable> is the number of
days, weeks, or months after which an
<acronym>MFC</acronym> is planned. The
<replaceable>weeks</replaceable> option may be
<literal>day</literal>, <literal>days</literal>,
<literal>week</literal>, <literal>weeks</literal>,
<literal>month</literal>, <literal>months</literal>, or
may be left off (in which case, days will be
assumed).</para>
</example>
<para>In some cases you may need to combine some of
these.</para>
<para>Consider the situation where a user has submitted a PR
containing code from the NetBSD project. You are looking
at the PR, but it is not an area of the tree you normally
work in, so you have decided to get the change reviewed by
the <literal>arch</literal> mailing list. Since the
change is complex, you opt to <acronym>MFC</acronym> after
one month to allow adequate testing.</para>
<para>The extra information to include in the commit would
look something like</para>
<programlisting>PR: foo/54321
Submitted by: John Smith &lt;John.Smith@example.com>
Reviewed by: -arch
Obtained from: NetBSD
MFC after: 1 month</programlisting>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I access
<hostid role="fqdn">people.FreeBSD.org</hostid> to put up
personal or project information?</para>
</question>
<answer>
<para><hostid role="fqdn">people.FreeBSD.org</hostid> is the
same as
<hostid role="fqdn">freefall.FreeBSD.org</hostid>. Just
create a <filename>public_html</filename> directory.
Anything you place in that directory will automatically be
visible under <ulink
url="http://people.FreeBSD.org/"></ulink>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>Where are the mailing list archives stored?</para>
</question>
<answer>
<para>The mailing lists are archived under
<filename>/g/mail</filename> which will show up as
<filename>/hub/g/mail</filename> with &man.pwd.1;. This
location is accessible from any machine on the &os;
cluster.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>I would like to mentor a new committer. What process
do I need to follow?</para>
</question>
<answer>
<para>See the <ulink
url="http://www.freebsd.org/internal/new-account.html">New
Account Creation Procedure</ulink> document on the
internal pages.</para>
</answer>
</qandaentry>
</qandaset>
</sect1>
</article>