<!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook V3.1//EN"[
<!entity % dummy "IGNORE">
<!entity supported SYSTEM "supported.sgml">
<!entity newfeatures SYSTEM "newfeatures.sgml">
<!entity p-intro SYSTEM "privoxy.sgml">
<!entity history SYSTEM "history.sgml">
<!entity seealso SYSTEM "seealso.sgml">
<!entity contacting SYSTEM "contacting.sgml">
<!entity copyright SYSTEM "copyright.sgml">
<!entity license SYSTEM "license.sgml">
<!entity p-version "3.0.11">
<!entity p-status "stable">
<!entity % p-not-stable "IGNORE">
<!entity % p-stable "INCLUDE">
<!entity % p-text "IGNORE"> <!-- define we are not a text only doc -->
<!entity % p-doc "INCLUDE"> <!-- and we are a formal doc -->
<!entity % seealso-extra "INCLUDE"> <!-- extra stuff from seealso.sgml -->
<!entity my-copy "©"> <!-- kludge for docbook2man -->
]>
<!--
File : $Source: /cvsroot/ijbswa/current/doc/source/developer-manual.sgml,v $
Purpose : developer manual
This file belongs into
ijbswa.sourceforge.net:/home/groups/i/ij/ijbswa/htdocs/
$Id: developer-manual.sgml,v 2.27 2009/02/19 02:20:22 hal9 Exp $
Copyright (C) 2001-2009 Privoxy Developers http://www.privoxy.org/
See LICENSE.
========================================================================
NOTE: Please read developer-manual/documentation.html before touching
anything in this, or other Privoxy documentation. You have been warned!
Failure to abide by this rule will result in the revocation of your license
to live a peaceful existence!
========================================================================
-->
<article id="index">
<artheader>
<title>Privoxy Developer Manual</title>
<pubdate>
<subscript>
<!-- Completely the wrong markup, but very little is allowed -->
<!-- in this part of an article. FIXME -->
<link linkend="copyright">Copyright</link> &my-copy; 2001-2009 by
<ulink url="http://www.privoxy.org/">Privoxy Developers</ulink>
</subscript>
</pubdate>
<pubdate>$Id: developer-manual.sgml,v 2.27 2009/02/19 02:20:22 hal9 Exp $</pubdate>
<!--
Note: this should generate a separate page, and a live link to it.
But it doesn't for some mysterious reason. Please leave commented
unless it can be fixed proper. For the time being, the copyright
statement will be in copyright.smgl.
Hal.
<legalnotice id="legalnotice">
<para>
text goes here ........
</para>
</legalnotice>
-->
<abstract>
<![%dummy;[
<para>
<comment>
This is here to keep vim syntax file from breaking :/
If I knew enough to fix it, I would.
PLEASE DO NOT REMOVE! HB: hal@foobox.net
</comment>
</para>
]]>
<para>
The developer manual provides guidance on coding, testing, packaging, documentation
and other issues of importance to those involved with
<application>Privoxy</application> development. It is mandatory (and helpful!) reading
for anyone who wants to join the team. Note that it's currently out of date
and may not be entirely correct. As always, patches are welcome.
</para>
<!-- Include privoxy.sgml boilerplate text: -->
<!-- &p-intro; Someone interested enough in the project to contribute
will already know at this point what Privoxy is. -->
<!-- end boilerplate -->
<para>
Please note that this document is constantly evolving. This copy represents
the state at the release of version &p-version;.
You can find the latest version of the this manual at <ulink
url="http://www.privoxy.org/developer-manual/">http://www.privoxy.org/developer-manual/</ulink>.
Please see <link linkend="contact">the Contact section</link>
on how to contact the developers.
</para>
<!-- <para> -->
<!-- Feel free to send a note to the developers at <email>ijbswa-developers@lists.sourceforge.net</email>. -->
<!-- </para> -->
</abstract>
</artheader>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="introduction"><title>Introduction</title>
<!--
I don't like seeing blank space :) So added *something* here.
-->
<para>
<application>Privoxy</application>, as an heir to
<application>Junkbuster</application>, is a Free Software project
and the code is licensed under the GNU General Public License version 2.
As such, <application>Privoxy</application> development is potentially open
to anyone who has the time, knowledge, and desire to contribute
in any capacity. Our goals are simply to continue the mission,
to improve <application>Privoxy</application>, and
to make it available to as wide an audience as possible.
</para>
<para>
One does not have to be a programmer to contribute. Packaging, testing,
documenting and porting, are all important jobs as well.
</para>
<!-- ~~~~~ New section ~~~~~ -->
<sect2 id="quickstart"><title>Quickstart to Privoxy Development</title>
<para>
The first step is to join the <ulink
url="mailto:ijbswa-developers@lists.sourceforge.net">developer's mailing list</ulink>.
You can submit your ideas, or even better patches. Patches are best
submitted to the Sourceforge tracker set up for this purpose, but
can be sent to the list for review too.
</para>
<para>
You will also need to have a cvs package installed, which will
entail having ssh installed as well (which seems to be a requirement of
SourceForge), in order to access the cvs repository. Having the GNU build
tools is also going to be important (particularly, autoconf and gmake).
</para>
<para>
For the time being (read, this section is under construction), you can
also refer to the extensive comments in the source code. In fact,
reading the code is recommended in any case.
</para>
</sect2>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="cvs"><title>The CVS Repository</title>
<para>
If you become part of the active development team, you will eventually
need write access to our holy grail, the CVS repository. One of the
team members will need to set this up for you. Please read
this chapter completely before accessing via CVS.
</para>
<sect2 id="cvsaccess"><title>Access to CVS</title>
<para>
The project's CVS repository is hosted on
<ulink url="http://sourceforge.net/">SourceForge.</ulink>
Please refer to the chapters 6 and 7 in
<ulink url="http://sourceforge.net/docman/?group_id=1">SF's site
documentation</ulink> for the technical access details for your
operating system. For historical reasons, the CVS server is
called <literal>ijbswa.cvs.sourceforge.net</literal>, the repository is
called <literal>ijbswa</literal>, and the source tree module is called
<literal>current</literal>.
</para>
</sect2>
<sect2 id="cvsbranches">
<title>Branches</title>
<para>
Within the CVS repository, there are modules and branches. As
mentioned, the sources are in the <literal>current</literal>
<quote>module</quote>. Other modules are present for platform specific
issues. There is a webview of the CVS hierarchy at <ulink
url="http://ijbswa.cvs.sourceforge.net/ijbswa/">http://ijbswa.cvs.sourceforge.net/ijbswa/</ulink>,
which might help with visualizing how these pieces fit together.
</para>
<para>
Branches are used to fork a sub-development path from the main trunk.
Within the <literal>current</literal> module where the sources are, there
is always at least one <quote>branch</quote> from the main trunk
devoted to a stable release series. The main trunk is where active
development takes place for the next stable series (e.g. 3.2.x).
So just prior to each stable series (e.g. 3.0.x), a branch is created
just for stable series releases (e.g. 3.0.0 -> 3.0.1 -> 3.0.2, etc).
Once the initial stable release of any stable branch has taken place,
this branch is <emphasis>only used for bugfixes</emphasis>, which have
had prior testing before being committed to CVS. (See <link
linkend="versionnumbers">Version Numbers</link> below for details on
versioning.)
</para>
<para>
At one time there were two distinct branches: stable and unstable. The
more drastic changes were to be in the unstable branch. These branches
have now been merged to minimize time and effort of maintaining two
branches.
</para>
<!--
<para>
This will result in at least two active branches, which means there may
be occasions that require the same (or similar) item to be
checked into to two different places (assuming its a bugfix and needs
fixing in both the stable and unstable trees). This also means that in
order to have access to both trees, both will have to be checked out
separately. Use the <literal>cvs -r</literal> flag to check out a
branch, e.g: <literal>cvs co -r v_3_0_branch current</literal>.
</para>
-->
</sect2>
<sect2 id="cvscommit"><title>CVS Commit Guidelines</title>
<para>
The source tree is the heart of every software project. Every effort must
be made to ensure that it is readable, compilable and consistent at all
times. There are differing guidelines for the stable branch and the
main development trunk, and we ask anyone with CVS access to strictly
adhere to the following guidelines:
</para>
<para>
Basic Guidelines, for all branches:
</para>
<para>
<itemizedlist>
<listitem><para>
Please don't commit even
a small change without testing it thoroughly first. When we're
close to a public release, ask a fellow developer to review your
changes.
</para></listitem>
<listitem><para>
Your commit message should give a concise overview of <emphasis>what you
changed</emphasis> (no big details) and <emphasis>why you changed it</emphasis>
Just check previous messages for good examples.
</para></listitem>
<listitem><para>
Don't use the same message on multiple files, unless it equally applies to
all those files.
</para></listitem>
<listitem><para>
If your changes span multiple files, and the code won't recompile unless
all changes are committed (e.g. when changing the signature of a function),
then commit all files one after another, without long delays in between.
If necessary, prepare the commit messages in advance.
</para></listitem>
<listitem><para>
Before changing things on CVS, make sure that your changes are in line
with the team's general consensus on what should be done.
</para></listitem>
<listitem>
<para>
Note that near a major public release, we get more cautious.
There is always the possibility to submit a patch to the <ulink
url="http://sourceforge.net/tracker/?atid=311118&group_id=11118&func=browse">patch
tracker</ulink> instead.
</para>
</listitem>
</itemizedlist>
</para>
<!--
<para>
Stable branches are handled with more care, especially after the
initial *.*.0 release, and we are just in bugfix mode. In addition to
the above, the below applies only to the stable branch (currently the
<literal>v_3_0_branch</literal> branch):
</para>
<para>
<itemizedlist>
<listitem>
<para>
Do not commit <emphasis>anything</emphasis> unless your proposed
changes have been well tested first, preferably by other members of the
project, or have prior approval of the project leaders or consensus
of the devel list.
</para>
</listitem>
<listitem>
<para>
Where possible, bugfixes and changes should be tested in the main
development trunk first. There may be occasions where this is not
feasible, though.
</para>
</listitem>
<listitem>
<para>
Alternately, proposed changes can be submitted as patches to the patch tracker on
Sourceforge first: <ulink
url="http://sourceforge.net/tracker/?group_id=11118&atid=311118">http://sourceforge.net/tracker/?group_id=11118&atid=311118</ulink>.
Then ask for peer review.
</para>
</listitem>
<listitem>
<para>
Do not even think about anything except bugfixes. No new features!
</para>
</listitem>
</itemizedlist>
</para>
-->
</sect2>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="documentation"><title>Documentation Guidelines</title>
<para>
All formal documents are maintained in Docbook SGML and located in the
<computeroutput>doc/source/*</computeroutput> directory. You will need
<ulink url="http://www.docbook.org">Docbook</ulink>, the Docbook
DTD's and the Docbook modular stylesheets (or comparable alternatives),
and either <application>jade</application> or
<application>openjade</application> (recommended) installed in order to
build docs from source. Currently there is <ulink
url="../user-manual/index.html"><citetitle>user-manual</citetitle></ulink>,
<ulink url="../faq/index.html"><citetitle>FAQ</citetitle></ulink>, and, of
course this, the <citetitle>developer-manual</citetitle> in this format.
The <citetitle>README</citetitle>, <citetitle>AUTHORS</citetitle>,
<citetitle>INSTALL</citetitle>,
<citetitle>privoxy.1</citetitle> (man page), and
<citetitle>config</citetitle> files are also now maintained as Docbook
SGML. These files, when built, in the top-level source directory are
generated files! Also, the <application>Privoxy</application> <filename>index.html</filename> (and a
variation on this file, <filename>privoxy-index.html</filename>,
meant for inclusion with doc packages), are maintained as SGML as well.
<emphasis>DO NOT edit these directly</emphasis>. Edit the SGML source, or
contact someone involved in the documentation.
</para>
<para>
<filename>config</filename> requires some special handling. The reason it
is maintained this way is so that the extensive comments in the file
mirror those in <citetitle>user-manual</citetitle>. But the conversion
process requires going from SGML to HTML to text to special formatting
required for the embedded comments. Some of this does not survive so
well. Especially some of the examples that are longer than 80 characters.
The build process for this file outputs to <filename>config.new</filename>,
which should be reviewed for errors and mis-formatting. Once satisfied
that it is correct, then it should be hand copied to
<filename>config</filename>.
</para>
<para>
Other, less formal documents (e.g. <filename>LICENSE</filename>) are
maintained as plain text files in the top-level source directory.
</para>
<para>
Packagers are encouraged to include this documentation. For those without
the ability to build the docs locally, text versions of each are kept in
CVS. HTML versions are also being kept in CVS under
<filename>doc/webserver/*</filename>. And PDF version are kept in
<filename>doc/pdf/*</filename>.
</para>
<para>
Formal documents are built with the Makefile targets of
<computeroutput>make dok</computeroutput>, or alternately
<computeroutput>make redhat-dok</computeroutput>. If you have problems,
try both. The build process uses the document SGML sources in
<computeroutput>doc/source/*/*</computeroutput> to update all text files in
<computeroutput>doc/text/</computeroutput> and to update all HTML
documents in <computeroutput>doc/webserver/</computeroutput>.
</para>
<para>
Documentation writers should please make sure documents build
successfully before committing to CVS, if possible.
</para>
<para>
How do you update the webserver (i.e. the pages on privoxy.org)?
<orderedlist numeration="arabic">
<listitem><para>
First, build the docs by running <computeroutput>make
dok</computeroutput> (or alternately <computeroutput>make
redhat-dok</computeroutput>). For PDF docs, do <computeroutput>make
dok-pdf</computeroutput>.
</para></listitem>
<listitem><para>
Run <computeroutput>make webserver</computeroutput> which copies all
files from <computeroutput>doc/webserver</computeroutput> to the
sourceforge webserver via scp.
</para></listitem>
</orderedlist>
</para>
<para>
Finished docs should be occasionally submitted to CVS
(<filename>doc/webserver/*/*.html</filename>) so that those without
the ability to build them locally, have access to them if needed.
This is especially important just prior to a new release! Please
do this <emphasis>after</emphasis> the <literal>$VERSION</literal> and
other release specific data in <filename>configure.in</filename> has been
updated (this is done just prior to a new release).
</para>
<!-- ~~~~~ New section ~~~~~ -->
<sect2 id="sgml">
<title>Quickstart to Docbook and SGML</title>
<para>
If you are not familiar with SGML, it is a markup language similar to HTML.
Actually, not a mark up language per se, but a language used to define
markup languages. In fact, HTML is an SGML application. Both will use
<quote>tags</quote> to format text and other content. SGML tags can be much
more varied, and flexible, but do much of the same kinds of things. The tags,
or <quote>elements</quote>, are definable in SGML. There is no set
<quote>standards</quote>. Since we are using
<application>Docbook</application>, our tags are those that are defined by
<application>Docbook</application>. Much of how the finish document is
rendered is determined by the <quote>stylesheets</quote>.
The stylesheets determine how each tag gets translated to HTML, or other
formats.
</para>
<para>
Tags in Docbook SGML need to be always <quote>closed</quote>. If not, you
will likely generate errors. Example: <literal><title>My
Title</title></literal>. They are also case-insensitive, but we
strongly suggest using all lower case. This keeps compatibility with
[Docbook] <application>XML</application>.
</para>
<para>
Our documents use <quote>sections</quote> for the most part. Sections
will be processed into HTML headers (e.g. <literal>h1</literal> for
<literal>sect1</literal>). The <application>Docbook</application> stylesheets
will use these to also generate the Table of Contents for each doc. Our
TOC's are set to a depth of three. Meaning <literal>sect1</literal>,
<literal>sect2</literal>, and <literal>sect3</literal> will have TOC
entries, but <literal>sect4</literal> will not. Each section requires
a <literal><title></literal> element, and at least one
<literal><para></literal>. There is a limit of five section
levels in Docbook, but generally three should be sufficient for our
purposes.
</para>
<para>
Some common elements that you likely will use:
</para>
<para>
<simplelist>
<member>
<emphasis><para></para></emphasis>, paragraph delimiter. Most
text needs to be within paragraph elements (there are some exceptions).
</member>
<member>
<emphasis><emphasis></emphasis></emphasis>, the stylesheets
make this italics.
</member>
<member>
<emphasis><filename></filename></emphasis>, files and directories.
</member>
<member>
<emphasis><command></command></emphasis>, command examples.
</member>
<member>
<emphasis><literallayout></literallayout></emphasis>, like
<literal><pre></literal>, more or less.
</member>
<member>
<emphasis><itemizedlist></itemizedlist></emphasis>, list with bullets.
</member>
<member>
<emphasis><listitem></listitem></emphasis>, member of the above.
</member>
<member>
<emphasis><screen></screen></emphasis>, screen output, implies
<literal><literallayout></literal>.
</member>
<member>
<emphasis><ulink url="example.com"></ulink></emphasis>, like
HTML <literal><a></literal> tag.
</member>
<member>
<emphasis><quote></quote></emphasis>, for, doh, quoting text.
</member>
</simplelist>
</para>
<para>
Look at any of the existing docs for examples of all these and more.
</para>
<para>
You might also find <quote><ulink
url="http://opensource.bureau-cornavin.com/crash-course/index.html">Writing Documentation
Using DocBook - A Crash Course</ulink></quote> useful.
</para>
</sect2>
<!-- ~~~~~ New section ~~~~~ -->
<sect2 id="docstyle">
<title><application>Privoxy</application> Documentation Style</title>
<para>
It will be easier if everyone follows a similar writing style. This
just makes it easier to read what someone else has written if it
is all done in a similar fashion.
</para>
<para>
Here it is:
</para>
<para>
<itemizedlist>
<listitem>
<para>
All tags should be lower case.
</para>
</listitem>
<listitem>
<para>
Tags delimiting a <emphasis>block</emphasis> of text (even small
blocks) should be on their own line. Like:
<literallayout>
<para>
Some text goes here.
</para>
</literallayout>
Tags marking individual words, or few words, should be in-line:
<literallayout>
Just to <emphasis>emphasize</emphasis>, some text goes here.
</literallayout>
</para>
</listitem>
<listitem>
<para>
Tags should be nested and step indented for block text like: (except
in-line tags)
<literallayout>
<para>
<itemizedlist>
<para>
<listitem>
Some text goes here in our list example.
</listitem>
</para>
</itemizedlist>
</para>
</literallayout>
This makes it easier to find the text amongst the tags ;-)
</para>
</listitem>
<listitem>
<para>
Use white space to separate logical divisions within a document,
like between sections. Running everything together consistently
makes it harder to read and work on.
</para>
</listitem>
<listitem>
<para>
Do not hesitate to make comments. Comments can either use the
<comment> element, or the <!-- --> style comment
familiar from HTML. (Note in Docbook v4.x <comment> is
replaced by <remark>.)
</para>
</listitem>
<listitem>
<para>
We have an international audience. Refrain from slang, or English
idiosyncrasies (too many to list :). Humor also does not translate
well sometimes.
</para>
</listitem>
<listitem>
<para>
Try to keep overall line lengths in source files to 80 characters or less
for obvious reasons. This is not always possible, with lengthy URLs for
instance.
</para>
</listitem>
<listitem>
<para>
Our documents are available in differing formats. Right now, they
are just plain text, HTML, and PDF, but others are always a
future possibility. Be careful with URLs (<ulink>), and avoid
this mistake:
</para>
<para>
My favorite site is <ulink url="http://example.com">here</ulink>.
</para>
<para>
This will render as <quote>My favorite site is here</quote>, which is
not real helpful in a text doc. Better like this:
</para>
<para>
My favorite site is <ulink url="http://example.com">example.com</ulink>.
</para>
</listitem>
<listitem>
<para>
All documents should be spell checked occasionally.
<application>aspell</application> can check SGML with the
<literal>-H</literal> option. (<application>ispell</application> I think
too.)
</para>
</listitem>
</itemizedlist>
</para>
</sect2>
<!-- ~~~~~ New section ~~~~~ -->
<sect2><title>Privoxy Custom Entities</title>
<para>
<application>Privoxy</application> documentation is using
a number of customized <quote>entities</quote> to facilitate
documentation maintenance.
</para>
<para>
We are using a set of <quote>boilerplate</quote> files with generic text,
that is used by multiple docs. This way we can write something once, and use
it repeatedly without having to re-write the same content over and over again.
If editing such a file, keep in mind that it should be
<emphasis>generic</emphasis>. That is the purpose; so it can be used in varying
contexts without additional modifications.
</para>
<para>
We are also using what <application>Docbook</application> calls
<quote>internal entities</quote>. These are like variables in
programming. Well, sort of. For instance, we have the
<literal>p-version</literal> entity that contains the current
<application>Privoxy</application> version string. You are strongly
encouraged to use these where possible. Some of these obviously
require re-setting with each release (done by the Makefile). A sampling of
custom entities are listed below. See any of the main docs for examples.
</para>
<para>
<itemizedlist>
<listitem>
<para>
Re- <quote>boilerplate</quote> text entities are defined like:
</para>
<para>
<literal><!entity supported SYSTEM "supported.sgml"></literal>
</para>
<para>
In this example, the contents of the file,
<filename>supported.sgml</filename> is available for inclusion anywhere
in the doc. To make this happen, just reference the now defined
entity: <literal>&supported;</literal> (starts with an ampersand
and ends with a semi-colon), and the contents will be dumped into
the finished doc at that point.
</para>
</listitem>
<listitem>
<para>
Commonly used <quote>internal entities</quote>:
</para>
<simplelist>
<member>
<emphasis>p-version</emphasis>: the <application>Privoxy</application>
version string, e.g. <quote>&p-version;</quote>.
</member>
<member>
<emphasis>p-status</emphasis>: the project status, either
<quote>alpha</quote>, <quote>beta</quote>, or <quote>stable</quote>.
</member>
<member>
<emphasis>p-not-stable</emphasis>: use to conditionally include
text in <quote>not stable</quote> releases (e.g. <quote>beta</quote>).
</member>
<member>
<emphasis>p-stable</emphasis>: just the opposite.
</member>
<member>
<emphasis>p-text</emphasis>: this doc is only generated as text.
</member>
</simplelist>
</listitem>
</itemizedlist>
</para>
<para>
There are others in various places that are defined for a specific
purpose. Read the source!
</para>
</sect2>
</sect1>
<!-- <listitem><para>be consistent with the redirect script (i.e. the <application>Privoxy</application> program -->
<!-- points via the redirect URL at sf to valid end-points in the document)</para></listitem> -->
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="coding"><title>Coding Guidelines</title>
<sect2 id="s1"><title>Introduction</title>
<para>This set of standards is designed to make our lives easier. It is
developed with the simple goal of helping us keep the "new and improved
<application>Privoxy</application>" consistent and reliable. Thus making
maintenance easier and increasing chances of success of the
project.</para>
<para>And that of course comes back to us as individuals. If we can
increase our development and product efficiencies then we can solve more
of the request for changes/improvements and in general feel good about
ourselves. ;-></para>
</sect2>
<sect2 id="s2"><title>Using Comments</title>
<sect3 id="s3"><title>Comment, Comment, Comment</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Comment as much as possible without commenting the obvious.
For example do not comment "variable_a is equal to variable_b".
Instead explain why variable_a should be equal to the variable_b.
Just because a person can read code does not mean they will
understand why or what is being done. A reader may spend a lot
more time figuring out what is going on when a simple comment
or explanation would have prevented the extra research. Please
help your brother IJB'ers out!</para>
<para>The comments will also help justify the intent of the code.
If the comment describes something different than what the code
is doing then maybe a programming error is occurring.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
/* if page size greater than 1k ... */
if ( page_length() > 1024 )
{
... "block" the page up ...
}
/* if page size is small, send it in blocks */
if ( page_length() > 1024 )
{
... "block" the page up ...
}
This demonstrates 2 cases of "what not to do". The first is a
"syntax comment". The second is a comment that does not fit what
is actually being done.
</programlisting>
</sect3>
<sect3 id="s4"><title>Use blocks for comments</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Comments can help or they can clutter. They help when they
are differentiated from the code they describe. One line
comments do not offer effective separation between the comment
and the code. Block identifiers do, by surrounding the code
with a clear, definable pattern.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
/*********************************************************************
* This will stand out clearly in your code!
*********************************************************************/
if ( this_variable == that_variable )
{
do_something_very_important();
}
/* unfortunately, this may not */
if ( this_variable == that_variable )
{
do_something_very_important();
}
if ( this_variable == that_variable ) /* this may not either */
{
do_something_very_important();
}</programlisting>
<para><emphasis>Exception:</emphasis></para>
<para>If you are trying to add a small logic comment and do not
wish to "disrupt" the flow of the code, feel free to use a 1
line comment which is NOT on the same line as the code.</para>
</sect3>
<sect3 id="s5"><title>Keep Comments on their own line</title>
<para><emphasis>Explanation:</emphasis></para>
<para>It goes back to the question of readability. If the comment
is on the same line as the code it will be harder to read than
the comment that is on its own line.</para>
<para>There are three exceptions to this rule, which should be
violated freely and often: during the definition of variables,
at the end of closing braces, when used to comment
parameters.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
/*********************************************************************
* This will stand out clearly in your code,
* But the second example won't.
*********************************************************************/
if ( this_variable == this_variable )
{
do_something_very_important();
}
if ( this_variable == this_variable ) /*can you see me?*/
{
do_something_very_important(); /*not easily*/
}
/*********************************************************************
* But, the encouraged exceptions:
*********************************************************************/
int urls_read = 0; /* # of urls read + rejected */
int urls_rejected = 0; /* # of urls rejected */
if ( 1 == X )
{
do_something_very_important();
}
short do_something_very_important(
short firstparam, /* represents something */
short nextparam /* represents something else */ )
{
...code here...
} /* -END- do_something_very_important */
</programlisting>
</sect3>
<sect3 id="s6"><title>Comment each logical step</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Logical steps should be commented to help others follow the
intent of the written code and comments will make the code more
readable.</para>
<para>If you have 25 lines of code without a comment, you should
probably go back into it to see where you forgot to put
one.</para>
<para>Most "for", "while", "do", etc... loops _probably_ need a
comment. After all, these are usually major logic
containers.</para>
</sect3>
<sect3 id="s7"><title>Comment All Functions Thoroughly</title>
<para><emphasis>Explanation:</emphasis></para>
<para>A reader of the code should be able to look at the comments
just prior to the beginning of a function and discern the
reason for its existence and the consequences of using it. The
reader should not have to read through the code to determine if
a given function is safe for a desired use. The proper
information thoroughly presented at the introduction of a
function not only saves time for subsequent maintenance or
debugging, it more importantly aids in code reuse by allowing a
user to determine the safety and applicability of any function
for the problem at hand. As a result of such benefits, all
functions should contain the information presented in the
addendum section of this document.</para>
</sect3>
<sect3 id="s8"><title>Comment at the end of braces if the
content is more than one screen length</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Each closing brace should be followed on the same line by a
comment that describes the origination of the brace if the
original brace is off of the screen, or otherwise far away from
the closing brace. This will simplify the debugging,
maintenance, and readability of the code.</para>
<para>As a suggestion , use the following flags to make the
comment and its brace more readable:</para>
<para>use following a closing brace: } /* -END- if() or while ()
or etc... */</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
if ( 1 == X )
{
do_something_very_important();
...some long list of commands...
} /* -END- if x is 1 */
or:
if ( 1 == X )
{
do_something_very_important();
...some long list of commands...
} /* -END- if ( 1 == X ) */
</programlisting>
</sect3>
</sect2>
<sect2 id="s9"><title>Naming Conventions</title>
<sect3 id="s10"><title>Variable Names</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Use all lowercase, and separate words via an underscore
('_'). Do not start an identifier with an underscore. (ANSI C
reserves these for use by the compiler and system headers.) Do
not use identifiers which are reserved in ANSI C++. (E.g.
template, class, true, false, ...). This is in case we ever
decide to port Privoxy to C++.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
int ms_iis5_hack = 0;</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>
<programlisting>
int msiis5hack = 0; int msIis5Hack = 0;
</programlisting>
</para>
</sect3>
<sect3 id="s11"><title>Function Names</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Use all lowercase, and separate words via an underscore
('_'). Do not start an identifier with an underscore. (ANSI C
reserves these for use by the compiler and system headers.) Do
not use identifiers which are reserved in ANSI C++. (E.g.
template, class, true, false, ...). This is in case we ever
decide to port Privoxy to C++.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
int load_some_file( struct client_state *csp )</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>
<programlisting>
int loadsomefile( struct client_state *csp )
int loadSomeFile( struct client_state *csp )
</programlisting>
</para>
</sect3>
<sect3 id="s12"><title>Header file prototypes</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Use a descriptive parameter name in the function prototype
in header files. Use the same parameter name in the header file
that you use in the c file.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
(.h) extern int load_aclfile( struct client_state *csp );
(.c) int load_aclfile( struct client_state *csp )</programlisting>
<para><emphasis>Instead of:</emphasis>
<programlisting>
(.h) extern int load_aclfile( struct client_state * ); or
(.h) extern int load_aclfile();
(.c) int load_aclfile( struct client_state *csp )
</programlisting>
</para>
</sect3>
<sect3 id="s13"><title>Enumerations, and #defines</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Use all capital letters, with underscores between words. Do
not start an identifier with an underscore. (ANSI C reserves
these for use by the compiler and system headers.)</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
(enumeration) : enum Boolean { FALSE, TRUE };
(#define) : #define DEFAULT_SIZE 100;</programlisting>
<para><emphasis>Note:</emphasis> We have a standard naming scheme for #defines
that toggle a feature in the preprocessor: FEATURE_>, where
> is a short (preferably 1 or 2 word) description.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
#define FEATURE_FORCE 1
#ifdef FEATURE_FORCE
#define FORCE_PREFIX blah
#endif /* def FEATURE_FORCE */
</programlisting>
</sect3>
<sect3 id="s14"><title>Constants</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Spell common words out entirely (do not remove vowels).</para>
<para>Use only widely-known domain acronyms and abbreviations.
Capitalize all letters of an acronym.</para>
<para>Use underscore (_) to separate adjacent acronyms and
abbreviations. Never terminate a name with an underscore.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
#define USE_IMAGE_LIST 1</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>
<programlisting>
#define USE_IMG_LST 1 or
#define _USE_IMAGE_LIST 1 or
#define USE_IMAGE_LIST_ 1 or
#define use_image_list 1 or
#define UseImageList 1
</programlisting>
</para>
</sect3>
</sect2>
<sect2 id="s15"><title>Using Space</title>
<sect3 id="s16"><title>Put braces on a line by themselves.</title>
<para><emphasis>Explanation:</emphasis></para>
<para>The brace needs to be on a line all by itself, not at the
end of the statement. Curly braces should line up with the
construct that they're associated with. This practice makes it
easier to identify the opening and closing braces for a
block.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
if ( this == that )
{
...
}</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>if ( this == that ) { ... }</para>
<para>or</para>
<para>if ( this == that ) { ... }</para>
<para><emphasis>Note:</emphasis> In the special case that the if-statement is
inside a loop, and it is trivial, i.e. it tests for a
condition that is obvious from the purpose of the block,
one-liners as above may optically preserve the loop structure
and make it easier to read.</para>
<para><emphasis>Status:</emphasis> developer-discretion.</para>
<para><emphasis>Example exception:</emphasis></para>
<programlisting>
while ( more lines are read )
{
/* Please document what is/is not a comment line here */
if ( it's a comment ) continue;
do_something( line );
}
</programlisting>
</sect3>
<sect3 id="s17"><title>ALL control statements should have a
block</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Using braces to make a block will make your code more
readable and less prone to error. All control statements should
have a block defined.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
if ( this == that )
{
do_something();
do_something_else();
}</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>if ( this == that ) do_something(); do_something_else();</para>
<para>or</para>
<para>if ( this == that ) do_something();</para>
<para><emphasis>Note:</emphasis> The first example in "Instead of" will execute
in a manner other than that which the developer desired (per
indentation). Using code braces would have prevented this
"feature". The "explanation" and "exception" from the point
above also applies.</para>
</sect3>
<sect3 id="s18"><title>Do not belabor/blow-up boolean
expressions</title>
<para><emphasis>Example:</emphasis></para>
<programlisting>
structure->flag = ( condition );</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>if ( condition ) { structure->flag = 1; } else {
structure->flag = 0; }</para>
<para><emphasis>Note:</emphasis> The former is readable and concise. The later
is wordy and inefficient. Please assume that any developer new
to the project has at least a "good" knowledge of C/C++. (Hope
I do not offend by that last comment ... 8-)</para>
</sect3>
<sect3 id="s19"><title>Use white space freely because it is
free</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Make it readable. The notable exception to using white space
freely is listed in the next guideline.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
int first_value = 0;
int some_value = 0;
int another_value = 0;
int this_variable = 0;
if ( this_variable == this_variable )
first_value = old_value + ( ( some_value - another_value ) - whatever )
</programlisting>
</sect3>
<sect3 id="s20"><title>Don't use white space around structure
operators</title>
<para><emphasis>Explanation:</emphasis></para>
<para>- structure pointer operator ( "->" ) - member operator (
"." ) - functions and parentheses</para>
<para>It is a general coding practice to put pointers, references,
and function parentheses next to names. With spaces, the
connection between the object and variable/function name is not
as clear.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
a_struct->a_member;
a_struct.a_member;
function_name();</programlisting>
<para><emphasis>Instead of:</emphasis> a_struct -> a_member; a_struct . a_member;
function_name ();</para>
</sect3>
<sect3 id="s21"><title>Make the last brace of a function stand
out</title>
<para><emphasis>Example:</emphasis></para>
<programlisting>
int function1( ... )
{
...code...
return( ret_code );
} /* -END- function1 */
int function2( ... )
{
} /* -END- function2 */
</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>int function1( ... ) { ...code... return( ret_code ); } int
function2( ... ) { }</para>
<para><emphasis>Note:</emphasis> Use 1 blank line before the closing brace and 2
lines afterward. This makes the end of function standout to
the most casual viewer. Although function comments help
separate functions, this is still a good coding practice. In
fact, I follow these rules when using blocks in "for", "while",
"do" loops, and long if {} statements too. After all whitespace
is free!</para>
<para><emphasis>Status:</emphasis> developer-discretion on the number of blank
lines. Enforced is the end of function comments.</para>
</sect3>
<sect3 id="s22"><title>Use 3 character indentions</title>
<para><emphasis>Explanation:</emphasis></para>
<para>If some use 8 character TABs and some use 3 character TABs,
the code can look *very* ragged. So use 3 character indentions
only. If you like to use TABs, pass your code through a filter
such as "expand -t3" before checking in your code.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
static const char * const url_code_map[256] =
{
NULL, ...
};
int function1( ... )
{
if ( 1 )
{
return( ALWAYS_TRUE );
}
else
{
return( HOW_DID_YOU_GET_HERE );
}
return( NEVER_GETS_HERE );
}
</programlisting>
</sect3>
</sect2>
<sect2 id="s23"><title>Initializing</title>
<sect3 id="s24"><title>Initialize all variables</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Do not assume that the variables declared will not be used
until after they have been assigned a value somewhere else in
the code. Remove the chance of accidentally using an unassigned
variable.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
short a_short = 0;
float a_float = 0;
struct *ptr = NULL;</programlisting>
<para><emphasis>Note:</emphasis> It is much easier to debug a SIGSEGV if the
message says you are trying to access memory address 00000000
and not 129FA012; or array_ptr[20] causes a SIGSEV vs.
array_ptr[0].</para>
<para><emphasis>Status:</emphasis> developer-discretion if and only if the
variable is assigned a value "shortly after" declaration.</para>
</sect3>
</sect2>
<sect2 id="s25"><title>Functions</title>
<sect3 id="s26"><title>Name functions that return a boolean as a
question.</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Value should be phrased as a question that would logically
be answered as a true or false statement</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
should_we_block_this();
contains_an_image();
is_web_page_blank();
</programlisting>
</sect3>
<sect3 id="s27"><title>Always specify a return type for a
function.</title>
<para><emphasis>Explanation:</emphasis></para>
<para>The default return for a function is an int. To avoid
ambiguity, create a return for a function when the return has a
purpose, and create a void return type if the function does not
need to return anything.</para>
</sect3>
<sect3 id="s28"><title>Minimize function calls when iterating by
using variables</title>
<para><emphasis>Explanation:</emphasis></para>
<para>It is easy to write the following code, and a clear argument
can be made that the code is easy to understand:</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
for ( size_t cnt = 0; cnt < block_list_length(); cnt++ )
{
....
}</programlisting>
<para><emphasis>Note:</emphasis> Unfortunately, this makes a function call for
each and every iteration. This increases the overhead in the
program, because the compiler has to look up the function each
time, call it, and return a value. Depending on what occurs in
the block_list_length() call, it might even be creating and
destroying structures with each iteration, even though in each
case it is comparing "cnt" to the same value, over and over.
Remember too - even a call to block_list_length() is a function
call, with the same overhead.</para>
<para>Instead of using a function call during the iterations,
assign the value to a variable, and evaluate using the
variable.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
size_t len = block_list_length();
for ( size_t cnt = 0; cnt < len; cnt++ )
{
....
}</programlisting>
<para><emphasis>Exceptions:</emphasis> if the value of block_list_length()
*may* change or could *potentially* change, then you must code the
function call in the for/while loop.</para>
</sect3>
<sect3 id="s29"><title>Pass and Return by Const Reference</title>
<para><emphasis>Explanation:</emphasis></para>
<para>This allows a developer to define a const pointer and call
your function. If your function does not have the const
keyword, we may not be able to use your function. Consider
strcmp, if it were defined as: extern int strcmp( char *s1,
char *s2 );</para>
<para>I could then not use it to compare argv's in main: int main(
int argc, const char *argv[] ) { strcmp( argv[0], "privoxy"
); }</para>
<para>Both these pointers are *const*! If the c runtime library
maintainers do it, we should too.</para>
</sect3>
<sect3 id="s30"><title>Pass and Return by Value</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Most structures cannot fit onto a normal stack entry (i.e.
they are not 4 bytes or less). Aka, a function declaration
like: int load_aclfile( struct client_state csp )</para>
<para>would not work. So, to be consistent, we should declare all
prototypes with "pass by value": int load_aclfile( struct
client_state *csp )</para>
</sect3>
<sect3 id="s31"><title>Names of include files</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Your include statements should contain the file name without
a path. The path should be listed in the Makefile, using -I as
processor directive to search the indicated paths. An exception
to this would be for some proprietary software that utilizes a
partial path to distinguish their header files from system or
other header files.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
#include <iostream.h> /* This is not a local include */
#include "config.h" /* This IS a local include */
</programlisting>
<para><emphasis>Exception:</emphasis></para>
<para>
<programlisting>
/* This is not a local include, but requires a path element. */
#include <sys/fileName.h>
</programlisting>
</para>
<para><emphasis>Note:</emphasis> Please! do not add "-I." to the Makefile
without a _very_ good reason. This duplicates the #include
"file.h" behavior.</para>
</sect3>
<sect3 id="s32"><title>Provide multiple inclusion
protection</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Prevents compiler and linker errors resulting from
redefinition of items.</para>
<para>Wrap each header file with the following syntax to prevent
multiple inclusions of the file. Of course, replace PROJECT_H
with your file name, with "." Changed to "_", and make it
uppercase.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
#ifndef PROJECT_H_INCLUDED
#define PROJECT_H_INCLUDED
...
#endif /* ndef PROJECT_H_INCLUDED */
</programlisting>
</sect3>
<sect3 id="s33"><title>Use `extern "C"` when appropriate</title>
<para><emphasis>Explanation:</emphasis></para>
<para>If our headers are included from C++, they must declare our
functions as `extern "C"`. This has no cost in C, but increases
the potential re-usability of our code.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
#ifdef __cplusplus
extern "C"
{
#endif /* def __cplusplus */
... function definitions here ...
#ifdef __cplusplus
}
#endif /* def __cplusplus */
</programlisting>
</sect3>
<sect3 id="s34"><title>Where Possible, Use Forward Struct
Declaration Instead of Includes</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Useful in headers that include pointers to other struct's.
Modifications to excess header files may cause needless
compiles.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
/*********************************************************************
* We're avoiding an include statement here!
*********************************************************************/
struct file_list;
extern file_list *xyz;</programlisting>
<para><emphasis>Note:</emphasis> If you declare "file_list xyz;" (without the
pointer), then including the proper header file is necessary.
If you only want to prototype a pointer, however, the header
file is unnecessary.</para>
<para><emphasis>Status:</emphasis> Use with discretion.</para>
</sect3>
</sect2>
<sect2 id="s35"><title>General Coding Practices</title>
<sect3 id="s36"><title>Turn on warnings</title>
<para><emphasis>Explanation</emphasis></para>
<para>Compiler warnings are meant to help you find bugs. You
should turn on as many as possible. With GCC, the switch is
"-Wall". Try and fix as many warnings as possible.</para>
</sect3>
<sect3 id="s37"><title>Provide a default case for all switch
statements</title>
<para><emphasis>Explanation:</emphasis></para>
<para>What you think is guaranteed is never really guaranteed. The
value that you don't think you need to check is the one that
someday will be passed. So, to protect yourself from the
unknown, always have a default step in a switch statement.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
switch( hash_string( cmd ) )
{
case hash_actions_file :
... code ...
break;
case hash_confdir :
... code ...
break;
default :
log_error( ... );
... anomaly code goes here ...
continue; / break; / exit( 1 ); / etc ...
} /* end switch( hash_string( cmd ) ) */</programlisting>
<para><emphasis>Note:</emphasis> If you already have a default condition, you
are obviously exempt from this point. Of note, most of the
WIN32 code calls `DefWindowProc' after the switch statement.
This API call *should* be included in a default statement.</para>
<para><emphasis>Another Note:</emphasis> This is not so much a readability issue
as a robust programming issue. The "anomaly code goes here" may
be no more than a print to the STDERR stream (as in
load_config). Or it may really be an abort condition.</para>
<para><emphasis>Status:</emphasis> Programmer discretion is advised.</para>
</sect3>
<sect3 id="s38"><title>Try to avoid falling through cases in a
switch statement.</title>
<para><emphasis>Explanation:</emphasis></para>
<para>In general, you will want to have a 'break' statement within
each 'case' of a switch statement. This allows for the code to
be more readable and understandable, and furthermore can
prevent unwanted surprises if someone else later gets creative
and moves the code around.</para>
<para>The language allows you to plan the fall through from one
case statement to another simply by omitting the break
statement within the case statement. This feature does have
benefits, but should only be used in rare cases. In general,
use a break statement for each case statement.</para>
<para>If you choose to allow fall through, you should comment both
the fact of the fall through and reason why you felt it was
necessary.</para>
</sect3>
<sect3 id="s39"><title>Use 'long' or 'short' Instead of
'int'</title>
<para><emphasis>Explanation:</emphasis></para>
<para>On 32-bit platforms, int usually has the range of long. On
16-bit platforms, int has the range of short.</para>
<para><emphasis>Status:</emphasis> open-to-debate. In the case of most FSF
projects (including X/GNU-Emacs), there are typedefs to int4,
int8, int16, (or equivalence ... I forget the exact typedefs
now). Should we add these to IJB now that we have a "configure"
script?</para>
</sect3>
<sect3 id="s40"><title>Don't mix size_t and other types</title>
<para><emphasis>Explanation:</emphasis></para>
<para>The type of size_t varies across platforms. Do not make
assumptions about whether it is signed or unsigned, or about
how long it is. Do not compare a size_t against another
variable of a different type (or even against a constant)
without casting one of the values.</para>
</sect3>
<sect3 id="s41"><title>Declare each variable and struct on its
own line.</title>
<para><emphasis>Explanation:</emphasis></para>
<para>It can be tempting to declare a series of variables all on
one line. Don't.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
long a = 0;
long b = 0;
long c = 0;</programlisting>
<para><emphasis>Instead of:</emphasis></para>
<para>long a, b, c;</para>
<para><emphasis>Explanation:</emphasis> - there is more room for comments on the
individual variables - easier to add new variables without
messing up the original ones - when searching on a variable to
find its type, there is less clutter to "visually"
eliminate</para>
<para><emphasis>Exceptions:</emphasis> when you want to declare a bunch of loop
variables or other trivial variables; feel free to declare them
on one line. You should, although, provide a good comment on
their functions.</para>
<para><emphasis>Status:</emphasis> developer-discretion.</para>
</sect3>
<sect3 id="s42"><title>Use malloc/zalloc sparingly</title>
<para><emphasis>Explanation:</emphasis></para>
<para>Create a local struct (on the stack) if the variable will
live and die within the context of one function call.</para>
<para>Only "malloc" a struct (on the heap) if the variable's life
will extend beyond the context of one function call.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
If a function creates a struct and stores a pointer to it in a
list, then it should definitely be allocated via `malloc'.
</programlisting>
</sect3>
<sect3 id="s43"><title>The Programmer Who Uses 'malloc' is
Responsible for Ensuring 'free'</title>
<para><emphasis>Explanation:</emphasis></para>
<para>If you have to "malloc" an instance, you are responsible for
insuring that the instance is `free'd, even if the deallocation
event falls within some other programmer's code. You are also
responsible for ensuring that deletion is timely (i.e. not too
soon, not too late). This is known as "low-coupling" and is a
"good thing (tm)". You may need to offer a
free/unload/destructor type function to accommodate this.</para>
<para><emphasis>Example:</emphasis></para>
<programlisting>
int load_re_filterfile( struct client_state *csp ) { ... }
static void unload_re_filterfile( void *f ) { ... }</programlisting>
<para><emphasis>Exceptions:</emphasis></para>
<para>The developer cannot be expected to provide `free'ing
functions for C run-time library functions ... such as
`strdup'.</para>
<para><emphasis>Status:</emphasis> developer-discretion. The "main" use of this
standard is for allocating and freeing data structures (complex
or nested).</para>
</sect3>
<sect3 id="s44"><title>Add loaders to the `file_list' structure
and in order</title>
<para><emphasis>Explanation:</emphasis></para>
<para>I have ordered all of the "blocker" file code to be in alpha
order. It is easier to add/read new blockers when you expect a
certain order.</para>
<para><emphasis>Note:</emphasis> It may appear that the alpha order is broken in
places by POPUP tests coming before PCRS tests. But since
POPUPs can also be referred to as KILLPOPUPs, it is clear that
it should come first.</para>
</sect3>
<sect3 id="s45"><title>"Uncertain" new code and/or changes to
existing code, use FIXME or XXX</title>
<para><emphasis>Explanation:</emphasis></para>
<para>If you have enough confidence in new code or confidence in
your changes, but are not *quite* sure of the repercussions,
add this:</para>
<para>/* FIXME: this code has a logic error on platform XYZ, *
attempting to fix */ #ifdef PLATFORM ...changed code here...
#endif</para>
<para>or:</para>
<para>/* FIXME: I think the original author really meant this...
*/ ...changed code here...</para>
<para>or:</para>
<para>/* FIXME: new code that *may* break something else... */
...new code here...</para>
<para><emphasis>Note:</emphasis> If you make it clear that this may or may not
be a "good thing (tm)", it will be easier to identify and
include in the project (or conversely exclude from the
project).</para>
</sect3>
</sect2>
<sect2 id="s46"><title>Addendum: Template for files and function
comment blocks:</title>
<para><emphasis>Example for file comments:</emphasis></para>
<programlisting>
const char FILENAME_rcs[] = "$I<!-- Break CVS Substitution -->d$";
/*********************************************************************
*
* File : $S<!-- Break CVS Substitution -->ource$
*
* Purpose : (Fill me in with a good description!)
*
* Copyright : Written by and Copyright (C) 2001-2009
* the Privoxy team. http://www.privoxy.org/
*
* This program is free software; you can redistribute it
* and/or modify it under the terms of the GNU General
* Public License as published by the Free Software
* Foundation; either version 2 of the License, or (at
* your option) any later version.
*
* This program is distributed in the hope that it will
* be useful, but WITHOUT ANY WARRANTY; without even the
* implied warranty of MERCHANTABILITY or FITNESS FOR A
* PARTICULAR PURPOSE. See the GNU General Public
* License for more details.
*
* The GNU General Public License should be included with
* this file. If not, you can view it at
* http://www.gnu.org/licenses/old-licenses/gpl-2.0.html
* or write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 ,
* USA
*
* Revisions :
* $L<!-- Break CVS Substitution -->og$
*
*********************************************************************/
#include "config.h"
...necessary include files for us to do our work...
const char FILENAME_h_rcs[] = FILENAME_H_VERSION;
</programlisting>
<para><emphasis>Note:</emphasis> This declares the rcs variables that should be
added to the "show-proxy-args" page. If this is a brand new
creation by you, you are free to change the "Copyright" section
to represent the rights you wish to maintain.</para>
<para><emphasis>Note:</emphasis> The formfeed character that is present right
after the comment flower box is handy for (X|GNU)Emacs users to
skip the verbiage and get to the heart of the code (via
`forward-page' and `backward-page'). Please include it if you
can.</para>
<para><emphasis>Example for file header comments:</emphasis></para>
<programlisting>
#ifndef _FILENAME_H
#define _FILENAME_H
#define FILENAME_H_VERSION "$I<!-- Break CVS Substitution -->d$"
/*********************************************************************
*
* File : $S<!-- Break CVS Substitution -->ource$
*
* Purpose : (Fill me in with a good description!)
*
* Copyright : Written by and Copyright (C) 2001-2009
* the Privoxy team. http://www.privoxy.org/
*
* This program is free software; you can redistribute it
* and/or modify it under the terms of the GNU General
* Public License as published by the Free Software
* Foundation; either version 2 of the License, or (at
* your option) any later version.
*
* This program is distributed in the hope that it will
* be useful, but WITHOUT ANY WARRANTY; without even the
* implied warranty of MERCHANTABILITY or FITNESS FOR A
* PARTICULAR PURPOSE. See the GNU General Public
* License for more details.
*
* The GNU General Public License should be included with
* this file. If not, you can view it at
* http://www.gnu.org/licenses/old-licenses/gpl-2.0.html
* or write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 ,
* USA
*
* Revisions :
* $L<!-- Break CVS Substitution -->og$
*
*********************************************************************/
#include "project.h"
#ifdef __cplusplus
extern "C" {
#endif
... function headers here ...
/* Revision control strings from this header and associated .c file */
extern const char FILENAME_rcs[];
extern const char FILENAME_h_rcs[];
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* ndef _FILENAME_H */
/*
Local Variables:
tab-width: 3
end:
*/
</programlisting>
<para><emphasis>Example for function comments:</emphasis></para>
<programlisting>
/*********************************************************************
*
* Function : FUNCTION_NAME
*
* Description : (Fill me in with a good description!)
*
* parameters :
* 1 : param1 = pointer to an important thing
* 2 : x = pointer to something else
*
* Returns : 0 => Ok, everything else is an error.
*
*********************************************************************/
int FUNCTION_NAME( void *param1, const char *x )
{
...
return( 0 );
}
</programlisting>
<para><emphasis>Note:</emphasis> If we all follow this practice, we should be
able to parse our code to create a "self-documenting" web
page.</para>
</sect2>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="testing"><title>Testing Guidelines</title>
<para>To be filled.
</para>
<!-- ~~~~~ New section ~~~~~ -->
<sect2 id="testing-plan"><title>Testplan for releases</title>
<para>
Explain release numbers. major, minor. developer releases. etc.
<orderedlist numeration="arabic">
<listitem><para>
Remove any existing rpm with rpm -e
</para></listitem>
<listitem><para>
Remove any file that was left over. This includes (but is not limited to)
<itemizedlist>
<listitem><para>/var/log/privoxy</para></listitem>
<listitem><para>/etc/privoxy</para></listitem>
<listitem><para>/usr/sbin/privoxy</para></listitem>
<listitem><para>/etc/init.d/privoxy</para></listitem>
<listitem><para>/usr/doc/privoxy*</para></listitem>
</itemizedlist>
</para></listitem>
<listitem><para>
Install the rpm. Any error messages?
</para></listitem>
<listitem><para>start,stop,status <application>Privoxy</application> with the specific script
(e.g. /etc/rc.d/init/privoxy stop). Reboot your machine. Does
autostart work?</para></listitem>
<listitem><para>Start browsing. Does <application>Privoxy</application> work? Logfile written?</para></listitem>
<listitem><para>Remove the rpm. Any error messages? All files removed?</para></listitem>
</orderedlist>
</para>
</sect2>
<!-- ~~~~~ New section ~~~~~ -->
<sect2 id="testing-report"><title>Test reports</title>
<para>
Please submit test reports only with the <ulink url="http://sourceforge.net/tracker/?func=add&group_id=11118&atid=395005">test form</ulink>
at sourceforge. Three simple steps:
<itemizedlist>
<listitem><para>Select category: the distribution you test on.</para></listitem>
<listitem><para>Select group: the version of <application>Privoxy</application> that we are about to release.</para></listitem>
<listitem><para>Fill the Summary and Detailed Description with something
intelligent (keep it short and precise).</para>
</listitem>
</itemizedlist>
Do not mail to the mailing list (we cannot keep track on issues there).
</para>
</sect2>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="newrelease"><title>Releasing a New Version</title>
<para>
When we release versions of <application>Privoxy</application>,
our work leaves our cozy secret lab and has to work in the cold
RealWorld[tm]. Once it is released, there is no way to call it
back, so it is very important that great care is taken to ensure
that everything runs fine, and not to introduce problems in the
very last minute.
</para>
<para>
So when releasing a new version, please adhere exactly to the
procedure outlined in this chapter.
</para>
<para>
The following programs are required to follow this process:
<filename>ncftpput</filename> (ncftp), <filename>scp, ssh</filename> (ssh),
<filename>gmake</filename> (GNU's version of make), autoconf, cvs.
</para>
<sect2 id="versionnumbers">
<title>Version numbers</title>
<para>
First you need to determine which version number the release will have.
<application>Privoxy</application> version numbers consist of three numbers,
separated by dots, like in X.Y.Z (e.g. 3.0.0), where:
<itemizedlist>
<listitem>
<para>
X, the version major, is rarely ever changed. It is increased by one if
turning a development branch into stable substantially changes the functionality,
user interface or configuration syntax. Majors 1 and 2 were
<application>Junkbuster</application>, and 3 will be the first stable
<application>Privoxy</application> release.
</para>
</listitem>
<listitem>
<para>
Y, the version minor, represents the branch within the major version.
At any point in time, there are two branches being maintained:
The stable branch, with an even minor, say, 2N, in which no functionality is
being added and only bug-fixes are made, and 2N+1, the development branch, in
which the further development of <application>Privoxy</application> takes
place.
This enables us to turn the code upside down and inside out, while at the same time
providing and maintaining a stable version.
The minor is reset to zero (and one) when the major is incremented. When a development
branch has matured to the point where it can be turned into stable, the old stable branch
2N is given up (i.e. no longer maintained), the former development branch 2N+1 becomes the
new stable branch 2N+2, and a new development branch 2N+3 is opened.
</para>
</listitem>
<listitem>
<para>
Z, the point or sub version, represents a release of the software within a branch.
It is therefore incremented immediately before each code freeze.
In development branches, only the even point versions correspond to actual releases,
while the odd ones denote the evolving state of the sources on CVS in between.
It follows that Z is odd on CVS in development branches most of the time. There, it gets
increased to an even number immediately before a code freeze, and is increased to an odd
number again immediately thereafter.
This ensures that builds from CVS snapshots are easily distinguished from released versions.
The point version is reset to zero when the minor changes.
</para>
<para>
Stable branches work a little differently, since there should be
little to no development happening in such branches. Remember,
only bugfixes, which presumably should have had some testing
before being committed. Stable branches will then have their
version reported as <literal>0.0.0</literal>, during that period
between releases when changes are being added. This is to denote
that this code is <emphasis>not for release</emphasis>. Then
as the release nears, the version is bumped according: e.g.
<literal>3.0.1 -> 0.0.0 -> 3.0.2</literal>.
</para>
</listitem>
</itemizedlist>
</para>
<para>
In summary, the main CVS trunk is the development branch where new
features are being worked on for the next stable series. This should
almost always be where the most activity takes place. There is always at
least one stable branch from the trunk, e.g now it is
<literal>3.0</literal>, which is only used to release stable versions.
Once the initial *.0 release of the stable branch has been done, then as a
rule, only bugfixes that have had prior testing should be committed to
the stable branch. Once there are enough bugfixes to justify a new
release, the version of this branch is again incremented Example: 3.0.0
-> 3.0.1 -> 3.0.2, etc are all stable releases from within the stable
branch. 3.1.x is currently the main trunk, and where work on 3.2.x is
taking place. If any questions, please post to the devel list
<emphasis>before</emphasis> committing to a stable branch!
</para>
<para>
Developers should remember too that if they commit a bugfix to the stable
branch, this will more than likely require a separate submission to the
main trunk, since these are separate development trees within CVS. If you
are working on both, then this would require at least two separate check
outs (i.e main trunk, <emphasis>and</emphasis> the stable release branch,
which is <literal>v_3_0_branch</literal> at the moment).
</para>
</sect2>
<sect2 id="beforerelease">
<title>Before the Release: Freeze</title>
<para>
The following <emphasis>must be done by one of the
developers</emphasis> prior to each new release.
</para>
<para>
<itemizedlist>
<listitem>
<para>
Make sure that everybody who has worked on the code in the last
couple of days has had a chance to yell <quote>no!</quote> in case
they have pending changes/fixes in their pipelines. Announce the
freeze so that nobody will interfere with last minute changes.
</para>
</listitem>
<listitem>
<para>
Increment the version number (point from odd to even in development
branches!) in <filename>configure.in</filename>. (RPM spec files
will need to be incremented as well.)
</para>
</listitem>
<listitem>
<para>
If <filename>default.action</filename> has changed since last
release (i.e. software release or standalone actions file release),
bump up its version info to A.B in this line:
</para>
<para>
<programlisting>
{+add-header{X-Actions-File-Version: A.B} -filter -no-popups}
</programlisting>
</para>
<para>
Then change the version info in doc/webserver/actions/index.php,
line: '$required_actions_file_version = "A.B";'
</para>
</listitem>
<listitem>
<para>
All documentation should be rebuild after the version bump.
Finished docs should be then be committed to CVS (for those
without the ability to build these). Some docs may require
rather obscure processing tools. <filename>config</filename>,
the man page (and the html version of the man page), and the PDF docs
fall in this category. REAMDE, the man page, AUTHORS, and config
should all also be committed to CVS for other packagers. The
formal docs should be uploaded to the webserver. See the
Section "Updating the webserver" in this manual for details.
</para>
</listitem>
<listitem>
<para>
The <citetitle>User Manual</citetitle> is also used for context
sensitive help for the CGI editor. This is version sensitive, so that
the user will get appropriate help for his/her release. So with
each release a fresh version should be uploaded to the webserver
(this is in addition to the main <citetitle>User Manual</citetitle>
link from the main page since we need to keep manuals for various
versions available). The CGI pages will link to something like
<literal>http://privoxy.org/$(VERSION)/user-manual/</literal>. This
will need to be updated for each new release. There is no Makefile
target for this at this time!!! It needs to be done manually.
</para>
</listitem>
<listitem>
<para>
All developers should look at the <filename>ChangeLog</filename> and
make sure noteworthy changes are referenced.
</para>
</listitem>
<listitem>
<para>
<emphasis>Commit all files that were changed in the above steps!</emphasis>
</para>
</listitem>
<listitem>
<para>
Tag all files in CVS with the version number with
<quote><command>cvs tag v_X_Y_Z</command></quote>.
Don't use vX_Y_Z, ver_X_Y_Z, v_X.Y.Z (won't work) etc.
</para>
</listitem>
<listitem>
<para>
If the release was in a development branch, increase the point version
from even to odd (X.Y.(Z+1)) again in <filename>configure.in</filename> and
commit your change.
</para>
</listitem>
<listitem>
<para>
On the webserver, copy the user manual to a new top-level directory
called <filename>X.Y.Z</filename>. This ensures that help links from the CGI
pages, which have the version as a prefix, will go into the right version of the manual.
If this is a development branch release, also symlink <filename>X.Y.(Z-1)</filename>
to <filename>X.Y.Z</filename> and <filename>X.Y.(Z+1)</filename> to
<filename>.</filename> (i.e. dot).
</para>
</listitem>
</itemizedlist>
</para>
</sect2>
<sect2 id="therelease">
<title>Building and Releasing the Packages</title>
<para>
Now the individual packages can be built and released. Note that for
GPL reasons the first package to be released is always the source tarball.
</para>
<para>
For <emphasis>all</emphasis> types of packages, including the source tarball,
<emphasis>you must make sure that you build from clean sources by exporting
the right version from CVS into an empty directory</emphasis> (just press return when
asked for a password):
</para>
<para>
<programlisting>
mkdir dist # delete or choose different name if it already exists
cd dist
cvs -d:pserver:anonymous@ijbswa.cvs.sourceforge.net:/cvsroot/ijbswa login
cvs -z3 -d:pserver:anonymous@ijbswa.cvs.sourceforge.net:/cvsroot/ijbswa export -r v_X_Y_Z current
</programlisting>
</para>
<para>
<emphasis>Do NOT change</emphasis> a single bit, including, but not limited to
version information after export from CVS. This is to make sure that
all release packages, and with them, all future bug reports, are based
on exactly the same code.
</para>
<warning>
<para>
Every significant release of Privoxy has included at least one
package that either had incorrect versions of files, missing files,
or incidental leftovers from a previous build process that gave
unknown numbers of users headaches to try to figure out what was
wrong. PLEASE, make sure you are using pristene sources, and are
following the prescribed process!
</para>
</warning>
<para>
Please find additional instructions for the source tarball and the
individual platform dependent binary packages below. And details
on the Sourceforge release process below that.
</para>
<sect3 id="pack-guidelines">
<title>Note on Privoxy Packaging</title>
<para>
Please keep these general guidelines in mind when putting together
your package. These apply to <emphasis>all</emphasis> platforms!
</para>
<para>
<itemizedlist>
<listitem>
<para>
<application>Privoxy</application> <emphasis>requires</emphasis>
write access to: all <filename>*.action</filename> files, all
logfiles, and the <filename>trust</filename> file. You will
need to determine the best way to do this for your platform.
</para>
</listitem>
<listitem>
<para>
Please include up to date documentation. At a bare minimum:
</para>
<simplelist>
<member>
<filename>LICENSE</filename> (top-level directory)
</member>
</simplelist>
<simplelist>
<member>
<filename>README</filename> (top-level directory)
</member>
</simplelist>
<simplelist>
<member>
<filename>AUTHORS</filename> (top-level directory)
</member>
</simplelist>
<simplelist>
<member>
<filename>man page</filename> (top-level directory, Unix-like
platforms only)
</member>
</simplelist>
<simplelist>
<member>
<filename>The User Manual</filename> (doc/webserver/user-manual/)
</member>
</simplelist>
<simplelist>
<member>
<filename>FAQ</filename> (doc/webserver/faq/)
</member>
</simplelist>
<para>
Also suggested: <filename>Developer Manual</filename>
(doc/webserver/developer-manual) and <filename>ChangeLog</filename>
(top-level directory). <filename>FAQ</filename> and the manuals are
HTML docs. There are also text versions in
<filename>doc/text/</filename> which could conceivably also be
included.
</para>
<para>
The documentation has been designed such that the manuals are linked
to each other from parallel directories, and should be packaged
that way. <filename>privoxy-index.html</filename> can also be
included and can serve as a focal point for docs and other links of
interest (and possibly renamed to <filename>index.html</filename>).
This should be one level up from the manuals. There is a link also
on this page to an HTMLized version of the man page. To avoid 404 for
this, it is in CVS as
<filename>doc/webserver/man-page/privoxy-man-page.html</filename>,
and should be included along with the manuals. There is also a
css stylesheets that can be included for better presentation:
<filename>p_doc.css</filename>. This should be in the same directory
with <filename>privoxy-index.html</filename>, (i.e. one level up from
the manual directories).
</para>
</listitem>
<listitem>
<para>
<filename>user.action</filename> and <filename>user.filter</filename>
are designed for local preferences. Make sure these do not get overwritten!
<filename>config</filename> should not be overwritten either. This
has especially important configuration data in it.
<filename>trust</filename> should be left in tact as well.
</para>
</listitem>
<listitem>
<para>
Other configuration files (<filename>default.action</filename> and
<filename>default.filter</filename>) should be installed as the new
defaults, but all previously installed configuration files should be
preserved as backups. This is just good manners :-) These files are
likely to change between releases and contain important new features
and bug fixes.
</para>
</listitem>
<listitem>
<para>
Please check platform specific notes in this doc, if you haven't
done <quote>Privoxy</quote> packaging before for other platform
specific issues. Conversely, please add any notes that you know
are important for your platform (or contact one of the doc
maintainers to do this if you can't).
</para>
</listitem>
<listitem>
<para>
Packagers should do a <quote>clean</quote> install of their
package after building it. So any previous installs should be
removed first to ensure the integrity of the newly built package.
Then run the package for a while to make sure there are no
obvious problems, before uploading.
</para>
</listitem>
</itemizedlist>
</para>
</sect3>
<sect3 id="newrelease-tarball"><title>Source Tarball</title>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then do:
</para>
<para>
<programlisting>
make tarball-dist
</programlisting>
</para>
<para>
To upload the package to Sourceforge, simply issue
</para>
<para>
<programlisting>
make tarball-upload
</programlisting>
</para>
<para>
Go to the displayed URL and release the file publicly on Sourceforge.
For the change log field, use the relevant section of the
<filename>ChangeLog</filename> file.
</para>
</sect3>
<sect3 id="newrelease-rpm"><title>SuSE, Conectiva or Red Hat RPM</title>
<para>
In following text, replace <replaceable class="parameter">dist</replaceable>
with either <quote>rh</quote> for Red Hat or <quote>suse</quote> for SuSE.
</para>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above).
</para>
<para>
As the only exception to not changing anything after export from CVS,
now examine the file <filename>privoxy-</filename><replaceable class="parameter">dist</replaceable><filename>.spec</filename>
and make sure that the version information and the RPM release number are
correct. The RPM release numbers for each version start at one. Hence it must
be reset to one if this is the first RPM for
<replaceable class="parameter">dist</replaceable> which is built from version
X.Y.Z. Check the
<ulink url="http://sourceforge.net/project/showfiles.php?group_id=11118">file
list</ulink> if unsure. Else, it must be set to the highest already available RPM
release number for that version plus one.
</para>
<para>
Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then do
</para>
<para>
<programlisting>
make <replaceable class="parameter">dist</replaceable>-dist
</programlisting>
</para>
<para>
To upload the package to Sourceforge, simply issue
</para>
<para>
<programlisting>
make <replaceable class="parameter">dist</replaceable>-upload <replaceable class="parameter">rpm_packagerev</replaceable>
</programlisting>
</para>
<para>
where <replaceable class="parameter">rpm_packagerev</replaceable> is the
RPM release number as determined above.
Go to the displayed URL and release the file publicly on Sourceforge.
Use the release notes and change log from the source tarball package.
</para>
</sect3>
<sect3 id="newrelease-os2"><title>OS/2</title>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then get the OS/2 Setup module:
</para>
<para>
<programlisting>
cvs -z3 -d:pserver:anonymous@ijbswa.cvs.sourceforge.net:/cvsroot/ijbswa co os2setup
</programlisting>
</para>
<para>
You will need a mix of development tools.
The main compilation takes place with IBM Visual Age C++.
Some ancillary work takes place with GNU tools, available from
various sources like hobbes.nmsu.edu.
Specificially, you will need <filename>autoheader</filename>,
<filename>autoconf</filename> and <filename>sh</filename> tools.
The packaging takes place with WarpIN, available from various sources, including
its home page: <ulink url="http://www.xworkplace.org/">xworkplace</ulink>.
</para>
<para>
Change directory to the <filename>os2setup</filename> directory.
Edit the os2build.cmd file to set the final executable filename.
For example,
</para>
<para>
<programlisting>
installExeName='privoxyos2_setup_X.Y.Z.exe'
</programlisting>
</para>
<para>
Next, edit the <filename>IJB.wis</filename> file so the release number matches
in the <filename>PACKAGEID</filename> section:
</para>
<para>
<programlisting>
PACKAGEID="Privoxy Team\Privoxy\Privoxy Package\X\Y\Z"
</programlisting>
</para>
<para>
You're now ready to build. Run:
</para>
<para>
<programlisting>
os2build
</programlisting>
</para>
<para>
You will find the WarpIN-installable executable in the
<filename>./files</filename> directory. Upload this anonymously to
<filename>uploads.sourceforge.net/incoming</filename>, create a release
for it, and you're done. Use the release notes and Change Log from the
source tarball package.
</para>
</sect3>
<sect3 id="newrelease-solaris"><title>Solaris</title>
<para>
Login to Sourceforge's compilefarm via ssh:
</para>
<para>
<programlisting>
ssh cf.sourceforge.net
</programlisting>
</para>
<para>
Choose the right operating system (not the Debian one).
When logged in, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then run
</para>
<para>
<programlisting>
gmake solaris-dist
</programlisting>
</para>
<para>
which creates a gzip'ed tar archive. Sadly, you cannot use <command>make
solaris-upload</command> on the Sourceforge machine (no ncftpput). You now have
to manually upload the archive to Sourceforge's ftp server and release
the file publicly. Use the release notes and Change Log from the
source tarball package.
</para>
</sect3>
<sect3 id="newrelease-windows"><title>Windows</title>
<para>
You should ensure you have the latest version of Cygwin (from
<ulink url="http://www.cygwin.com/">http://www.cygwin.com/</ulink>).
Run the following commands from within a Cygwin bash shell.
</para>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then get the Windows setup module:
</para>
<para>
<programlisting>
cvs -z3 -d:pserver:anonymous@ijbswa.cvs.sourceforge.net:/cvsroot/ijbswa co winsetup
</programlisting>
</para>
<para>
Then you can build the package. This is fully automated, and is
controlled by <filename>winsetup/GNUmakefile</filename>.
All you need to do is:
</para>
<para>
<programlisting>
cd winsetup
make
</programlisting>
</para>
<para>
Now you can manually rename <filename>privoxy_setup.exe</filename> to
<filename>privoxy_setup_X_Y_Z.exe</filename>, and upload it to
SourceForge. When releasing the package on SourceForge, use the release notes
and Change Log from the source tarball package.
</para>
</sect3>
<sect3 id="newrelease-debian"><title>Debian</title>
<para>
First, <emphasis>make sure that you have freshly exported the
right version into an empty directory</emphasis>. (See
"Building and releasing packages" above). Then add a log
entry to <filename>debian/changelog</filename>, if it is not
already there, for example by running:
</para>
<para>
<programlisting>
debchange -v &p-version;-&p-status;-1 "New upstream version"
</programlisting>
</para>
<para>
Then, run:
</para>
<para>
<programlisting>
dpkg-buildpackage -rfakeroot -us -uc -b
</programlisting>
</para>
<para>
This will create
<filename>../privoxy_&p-version;-&p-status;-1_i386.deb</filename>
which can be uploaded. To upload the package to Sourceforge, simply
issue
</para>
<para>
<programlisting>
make debian-upload
</programlisting>
</para>
</sect3>
<sect3 id="newrelease-macosx"><title>Mac OS X</title>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then get the Mac OS X setup module:
</para>
<para>
<programlisting>
cvs -z3 -d:pserver:anonymous@ijbswa.cvs.sourceforge.net:/cvsroot/ijbswa co osxsetup
</programlisting>
</para>
<para>
Then run:
</para>
<para>
<programlisting>
cd osxsetup
build
</programlisting>
</para>
<para>
This will run <filename>autoheader</filename>, <filename>autoconf</filename> and
<filename>configure</filename> as well as <filename>make</filename>.
Finally, it will copy over the necessary files to the ./osxsetup/files directory
for further processing by <filename>PackageMaker</filename>.
</para>
<para>
Bring up PackageMaker with the PrivoxyPackage.pmsp definition file, modify the package
name to match the release, and hit the "Create package" button.
If you specify ./Privoxy.pkg as the output package name, you can then create
the distributable zip file with the command:
</para>
<para>
<programlisting>
zip -r privoxyosx_setup_x.y.z.zip Privoxy.pkg
</programlisting>
</para>
<para>
You can then upload <filename>privoxyosx_setup_x.y.z.zip</filename> anonymously to
<filename>uploads.sourceforge.net/incoming</filename>,
create a release for it, and you're done. Use the release notes
and Change Log from the source tarball package.
</para>
</sect3>
<sect3 id="newrelease-freebsd"><title>FreeBSD</title>
<para>
Login to Sourceforge's compile-farm via ssh:
</para>
<para>
<programlisting>
ssh cf.sourceforge.net
</programlisting>
</para>
<para>
Choose the right operating system.
When logged in, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then run:
</para>
<para>
<programlisting>
gmake freebsd-dist
</programlisting>
</para>
<para>
which creates a gzip'ed tar archive. Sadly, you cannot use <command>make
freebsd-upload</command> on the Sourceforge machine (no ncftpput). You now have
to manually upload the archive to Sourceforge's ftp server and release
the file publicly. Use the release notes and Change Log from the
source tarball package.
</para>
</sect3>
<sect3 id="newrelease-hpux"><title>HP-UX 11</title>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then do FIXME.
</para>
</sect3>
<sect3 id="newrelease-amiga"><title>Amiga OS</title>
<para>
First, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then do FIXME.
</para>
</sect3>
<sect3 id="newrelease-aix"><title>AIX</title>
<para>
Login to Sourceforge's compilefarm via ssh:
</para>
<para>
<programlisting>
ssh cf.sourceforge.net
</programlisting>
</para>
<para>
Choose the right operating system.
When logged in, <emphasis>make sure that you have freshly exported the right
version into an empty directory</emphasis>. (See "Building and releasing
packages" above). Then run:
</para>
<para>
<programlisting>
cd current
autoheader && autoconf && ./configure
</programlisting>
</para>
<para>
Then run:
</para>
<para>
<programlisting>
make aix-dist
</programlisting>
</para>
<para>
which creates a gzip'ed tar archive. Sadly, you cannot use <command>make
aix-upload</command> on the Sourceforge machine (no ncftpput). You now have
to manually upload the archive to Sourceforge's ftp server and release
the file publicly. Use the release notes and Change Log from the
source tarball package.
</para>
</sect3>
</sect2>
<sect2 id="releasing">
<title>Uploading and Releasing Your Package</title>
<para>
After the package is ready, it is time to upload it
to SourceForge, and go through the release steps. The upload
is done via FTP:
</para>
<para>
<itemizedlist>
<listitem>
<para>
Upload to: <ulink url="ftp://upload.sourceforge.net/incoming">ftp://upload.sourceforge.net/incoming</ulink>
</para>
</listitem>
<listitem>
<para>
user: <literal>anonymous</literal>
</para>
</listitem>
<listitem>
<para>
password: <literal>ijbswa-developers@lists.sourceforge.net</literal>
</para>
</listitem>
</itemizedlist>
</para>
<para>
Or use the <command>make</command> targets as described above.
</para>
<para>
Once this done go to <ulink
url="https://sourceforge.net/project/admin/editpackages.php?group_id=11118"
>https://sourceforge.net/project/admin/editpackages.php?group_id=11118</ulink>,
making sure you are logged in. Find your target platform in the
second column, and click <literal>Add Release</literal>. You will
then need to create a new release for your package, using the format
of <literal>$VERSION ($CODE_STATUS)</literal>, e.g. <emphasis>&p-version;
(beta)</emphasis>.
</para>
<para>
Now just follow the prompts. Be sure to add any appropriate Release
notes. You should see your freshly uploaded packages in
<quote>Step 2. Add Files To This Release</quote>. Check the
appropriate box(es). Remember at each step to hit the
<quote>Refresh/Submit</quote> buttons! You should now see your
file(s) listed in Step 3. Fill out the forms with the appropriate
information for your platform, being sure to hit <quote>Update</quote>
for each file. If anyone is monitoring your platform, check the
<quote>email</quote> box at the very bottom to notify them of
the new package. This should do it!
</para>
<para>
If you have made errors, or need to make changes, you can go through
essentially the same steps, but select <literal>Edit Release</literal>,
instead of <literal>Add Release</literal>.
</para>
</sect2>
<sect2 id="afterrelease">
<title>After the Release</title>
<para>
When all (or: most of the) packages have been uploaded and made available,
send an email to the <ulink url="mailto:ijbswa-announce@lists.sourceforge.net">announce
mailing list</ulink>, Subject: "Version X.Y.Z available for download". Be sure to
include the
<ulink url="http://sourceforge.net/project/showfiles.php?group_id=11118">download
location</ulink>, the release notes and the Changelog. Also, post an
updated News item on the project page Sourceforge, and update the Home
page and docs linked from the Home page (see below). Other news sites
and release oriented sites, such as Freshmeat, should also be notified.
</para>
</sect2>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="webserver-update"><title>Update the Webserver</title>
<para>
The webserver should be updated at least with each stable release. When
updating, please follow these steps to make sure that no broken links,
inconsistent contents or permission problems will occur (as it has many
times in the past!):
</para>
<para>
If you have changed anything in the stable-branch documentation source
SGML files, do:
</para>
<para>
<programlisting>
make dok dok-pdf # (or 'make redhat-dok dok-pdf' if 'make dok' doesn't work for you)
</programlisting>
</para>
<para>
That will generate <filename>doc/webserver/user-manual</filename>,
<filename>doc/webserver/developer-manual</filename>,
<filename>doc/webserver/faq</filename>,
<filename>doc/pdf/*.pdf</filename> and
<filename>doc/webserver/index.html</filename> automatically.
</para>
<para>
If you changed the manual page sources, generate
<filename>doc/webserver/man-page/privoxy-man-page.html</filename>
by running <quote><command>make man</command></quote>. (This is
a separate target due to dependencies on some obscure perl scripts
[now in CVS, but not well tested]. See comments in <filename>GNUmakefile</filename>.)
</para>
<para>
If you want to add new files to the webserver, create them locally in
the <filename>doc/webserver/*</filename> directory (or
create new directories under <filename>doc/webserver</filename>).
</para>
<para>
Next, commit any changes from the above steps to CVS. All set?
If these are docs in the stable branch, then do:
</para>
<para>
<programlisting>
make webserver
</programlisting>
</para>
<para>
This will do the upload to <ulink url="http://www.privoxy.org/">the
webserver</ulink> (www.privoxy.org) and ensure all files and directories
there are group writable.
</para>
<para>
Please do <emphasis>NOT</emphasis> use any other means of transferring
files to the webserver to avoid permission problems. Also, please do not
upload docs from development branches or versions. The publicly posted
docs should be in sync with the last official release.
</para>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="contact"><title>Contacting the developers, Bug Reporting and Feature Requests</title>
<!-- Include contacting.sgml -->
&contacting;
<!-- end contacting -->
</sect1>
<!-- ~~~~~~~~ New section Header ~~~~~~~~~ -->
<sect1 id="copyright"><title>Privoxy Copyright, License and History</title>
<!-- Include copyright.sgml -->
©right;
<!-- end -->
<!-- ~~~~~ New section ~~~~~ -->
<sect2><title>License</title>
<!-- Include copyright.sgml: -->
&license;
<!-- end copyright -->
</sect2>
<!-- ~ End section ~ -->
<!-- ~~~~~ New section ~~~~~ -->
<sect2><title>History</title>
<!-- Include history.sgml -->
&history;
<!-- end -->
</sect2>
</sect1>
<!-- ~~~~~ New section ~~~~~ -->
<sect1 id="seealso"><title>See also</title>
<!-- Include seealso.sgml -->
&seealso;
<!-- end -->
</sect1>
<!--
This program is free software; you can redistribute it
and/or modify it under the terms of the GNU General
Public License as published by the Free Software
Foundation; either version 2 of the License, or (at
your option) any later version.
This program is distributed in the hope that it will
be useful, but WITHOUT ANY WARRANTY; without even the
implied warranty of MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE. See the GNU General Public
License for more details.
The GNU General Public License should be included with
this file. If not, you can view it at
http://www.gnu.org/copyleft/gpl.html
or write to the Free Software Foundation, Inc., 59
Temple Place - Suite 330, Boston, MA 02111-1307, USA.
$Log: developer-manual.sgml,v $
Revision 2.27 2009/02/19 02:20:22 hal9
Make some links in seealso conditional. Man page is now privoxy only links.
Revision 2.26 2009/02/12 16:08:26 fabiankeil
Declare the code stable.
Revision 2.25 2009/02/12 15:37:05 fabiankeil
Update templates.
Revision 2.24 2009/01/13 16:50:35 fabiankeil
The standard.action file is gone.
Revision 2.23 2008/09/26 17:02:01 fabiankeil
- Break some more CVS substitutions in examples.
- Remove Junkbusters reference in example header
for new files.
Revision 2.22 2008/08/30 15:37:35 fabiankeil
Update entities.
Revision 2.21 2008/08/16 08:51:28 fabiankeil
Update version-related entities.
Revision 2.20 2008/06/14 13:21:24 fabiankeil
Prepare for the upcoming 3.0.9 beta release.
Revision 2.19 2008/05/12 11:13:33 fabiankeil
Clarify that Privoxy is licensed under GPL version 2.
Revision 2.18 2008/02/04 12:14:06 fabiankeil
Change "Edit Packages" URL to use https.
Revision 2.17 2008/02/03 21:37:41 hal9
Apply patch from Mark: s/OSX/OS X/
Revision 2.16 2008/01/19 17:52:38 hal9
Re-commit to fix various minor issues for new release.
Revision 2.15 2008/01/19 15:03:05 hal9
Doc sources tagged for 3.0.8 release.
Revision 2.14 2008/01/17 01:49:51 hal9
Change copyright notice for docs s/2007/2008/. All these will be rebuilt soon
enough.
Revision 2.13 2007/10/30 17:59:31 fabiankeil
- Bump p-version, p-status and copyright date.
- Mention that the manual is out of date.
- Don't use examples with HardToReadCamelCase after
explaining that we actually don't like that.
- Minor cosmetics.
Revision 2.12 2006/11/14 01:57:46 hal9
Dump all docs prior to 3.0.6 release. Various minor changes to faq and user
manual.
Revision 2.11 2006/09/26 02:36:29 hal9
Fix broken link per bug tracker.
Revision 2.10 2006/09/22 01:27:55 hal9
Final commit of probably various minor changes here and there. Unless
something changes this should be ready for pending release.
Revision 2.9 2006/09/14 02:30:07 hal9
Fix ijbswa cvs links. Update notes on release process, and which config files
should be overwritten and which not.
Revision 2.8 2006/08/22 23:35:01 hal9
Fix email address, cvs URI, address branching changes and various other
small updates.
Revision 2.7 2006/07/18 14:48:50 david__schmidt
Reorganizing the repository: swapping out what was HEAD (the old 3.1 branch)
with what was really the latest development (the v_3_0_branch branch)
Revision 1.46.2.11 2002/12/11 13:12:15 hal9
Rewrite cvs write access give-away section.
Revision 1.46.2.10 2002/09/26 21:53:45 hal9
Changes to reflect recent change in stable branch commit policy (hopefully
clearer now).
Revision 1.46.2.9 2002/09/26 01:21:40 hal9
Porting 3.1.1 changes: more on cvs and branches, more on versions and
releases.
Revision 1.46.2.8 2002/08/17 00:16:10 hal9
Add note on updating webserver for User-manual/CGI editor, which is version
dependent (and different from main UM link).
Revision 1.46.2.7 2002/08/14 17:29:25 hal9
Add small notes on post-release steps, and uploading docs to webserver.
Revision 1.46.2.6 2002/08/10 11:40:25 oes
Added disclaimer about probably being out-of-date and two small hints
Revision 1.46.2.5 2002/08/09 01:15:12 hal9
Added some notes on pre-release steps (test builds first, update ChangeLog).
Revision 1.46.2.4 2002/05/29 00:30:59 mal0rd
Fixed a little formatting. Clarified debian section.
Revision 1.46.2.3 2002/05/28 04:32:45 hal9
Change hints on bundling index.html to privoxy-index.html
Revision 1.46.2.2 2002/05/26 17:04:24 hal9
-Spellcheck, very minor edits, and sync across branches
Revision 1.48 2002/05/26 12:48:31 roro
Add releasing information about Debian.
Revision 1.47 2002/05/26 04:55:11 mal0rd
Added debian-dist and debian-upload targets. Also documented usage.
Revision 1.46 2002/05/22 17:15:00 oes
Updated intro
Revision 1.45 2002/05/19 23:01:54 hal9
Add small section on general packaging guidelines (e.g. actions files must
be writable).
Revision 1.44 2002/05/15 03:55:17 hal9
Fix ulink -> link, and minor modification to release process section for
clarification.
Revision 1.43 2002/05/10 01:48:19 hal9
This is mostly proposed copyright/licensing additions and changes. Docs
are still GPL, but licensing and copyright are more visible. Also, copyright
changed in doc header comments (eliminate references to JB except FAQ).
Revision 1.42 2002/05/05 20:26:02 hal9
Sorting out license vs copyright in these docs.
Revision 1.41 2002/05/04 08:44:44 swa
bumped version
Revision 1.40 2002/05/04 00:43:43 hal9
-Remove TOC/first page kludge with proper stylesheet fix.
-Combined the two very brief sections: Intro and Quickstart.
Revision 1.39 2002/05/02 15:08:25 oes
Added explanation about version numbers and RPM package revisions
Revision 1.38 2002/04/29 02:20:31 hal9
Add info on steps for uploading and the release process on SF.
Revision 1.37 2002/04/26 17:23:29 swa
bookmarks cleaned, changed structure of user manual, screen and programlisting cleanups, and numerous other changes that I forgot
Revision 1.36 2002/04/26 05:25:23 hal9
Mass commit to catch a few scattered fixes.
Revision 1.35 2002/04/17 15:16:15 oes
Added link to docbook crash course
Revision 1.34 2002/04/15 23:39:32 oes
- Extended & fixed the release section
- Added CVS guideline sections
- Separated webserver section from release section
- Commented out boilerplate inclusion (If you don't know yet what it is,
you shouldn't mess with its code ;-)
- Nits & fixes
Revision 1.33 2002/04/12 03:49:53 hal9
Spell checked. Clarification on where docs are kept.
Revision 1.32 2002/04/11 21:29:58 jongfoster
Documenting Win32 release procedure
Revision 1.31 2002/04/11 09:32:52 oes
more nits
Revision 1.30 2002/04/11 09:24:53 oes
nits
Revision 1.29 2002/04/10 18:45:14 swa
generated
Revision 1.28 2002/04/08 22:59:26 hal9
Version update. Spell chkconfig correctly :)
Revision 1.27 2002/04/08 15:31:18 hal9
Touch ups to documentation section.
Revision 1.26 2002/04/07 23:50:08 hal9
Documentation changes to reflect HTML docs now in CVS, and new generated files
list.
Revision 1.25 2002/04/06 05:07:28 hal9
-Add privoxy-man-page.sgml, for man page.
-Add authors.sgml for AUTHORS (and p-authors.sgml)
-Reworked various aspects of various docs.
-Added additional comments to sub-docs.
Revision 1.24 2002/04/04 21:33:37 hal9
More on documenting the documents.
Revision 1.23 2002/04/04 18:46:47 swa
consistent look. reuse of copyright, history et. al.
Revision 1.22 2002/04/04 17:27:56 swa
more single file to be included at multiple points. make maintaining easier
Revision 1.21 2002/04/04 06:48:37 hal9
Structural changes to allow for conditional inclusion/exclusion of content
based on entity toggles, e.g. 'entity % p-not-stable "INCLUDE"'. And
definition of internal entities, e.g. 'entity p-version "2.9.13"' that will
eventually be set by Makefile.
More boilerplate text for use across multiple docs.
Revision 1.20 2002/04/04 03:28:27 david__schmidt
Add Mac OS X section
Revision 1.19 2002/04/03 15:09:42 david__schmidt
Add OS/2 build section
Revision 1.18 2002/04/03 03:51:48 hal9
Touch ups.
Revision 1.17 2002/04/03 01:21:17 hal9
Implementing Andreas's suggestions for Release sections.
Revision 1.16 2002/03/31 23:04:40 hal9
Fleshed out the doc section, and added something for an intro so it was not
blank.
Revision 1.15 2002/03/30 22:29:47 swa
wrong make flavour
Revision 1.14 2002/03/30 19:04:08 swa
people release differently. no good.
I want to make parts of the docs only.
Revision 1.13 2002/03/27 01:16:41 hal9
ditto
Revision 1.12 2002/03/27 01:02:51 hal9
Touch up on name change...
Revision 1.11 2002/03/26 22:29:55 swa
we have a new homepage!
Revision 1.10 2002/03/24 12:33:01 swa
more additions.
Revision 1.9 2002/03/24 11:01:05 swa
name change
Revision 1.8 2002/03/23 15:13:11 swa
renamed every reference to the old name with foobar.
fixed "application foobar application" tag, fixed
"the foobar" with "foobar". left junkbustser in cvs
comments and remarks to history untouched.
Revision 1.7 2002/03/11 13:13:27 swa
correct feedback channels
Revision 1.6 2002/02/24 14:25:06 jongfoster
Formatting changes. Now changing the doctype to DocBook XML 4.1
will work - no other changes are needed.
Revision 1.5 2001/10/31 18:16:51 swa
documentation added: howto generate docs in text and html
format, howto move stuff to the webserver.
Revision 1.4 2001/09/23 10:13:48 swa
upload process established. run make webserver and
the documentation is moved to the webserver. documents
are now linked correctly.
Revision 1.3 2001/09/13 15:27:40 swa
cosmetics
Revision 1.2 2001/09/13 15:20:17 swa
merged standards into developer manual
Revision 1.1 2001/09/12 15:36:41 swa
source files for junkbuster documentation
Revision 1.3 2001/09/10 17:43:59 swa
first proposal of a structure.
Revision 1.2 2001/06/13 14:28:31 swa
docs should have an author.
Revision 1.1 2001/06/13 14:20:37 swa
first import of project's documentation for the webserver.
-->
</article>