[privoxy.git] / doc / source / developer-manual.sgml

<!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 p-version "2.9.14">
<!entity p-status "beta">
<!entity % p-not-stable "INCLUDE">
<!entity % p-stable "IGNORE">
<!entity % p-text "IGNORE">        <!-- define we are not a text only doc -->
<!entity % p-doc "INCLUDE">        <!-- and we are a formal doc           -->
]>
<!--
 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 1.38 2002/04/29 02:20:31 hal9 Exp $

 Written by and Copyright (C) 2001 the SourceForge
 Privoxy team. http://www.privoxy.org/

 Based on the Internet Junkbuster originally written
 by and Copyright (C) 1997 Anonymous Coders and 
 Junkbusters Corporation.  http://www.junkbusters.com


 ========================================================================
 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>$Id: developer-manual.sgml,v 1.38 2002/04/29 02:20:31 hal9 Exp $</pubdate>

    <authorgroup>
      <author>
        <affiliation>
          <orgname>By: Privoxy Developers</orgname>
        </affiliation>
      </author>
    </authorgroup>

    <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 gives the users information on how to help the developer
 team. It provides guidance on coding, testing, documentation and other
 issues. 
 </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>
 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 <ulink url="contact.html">the Contact section</ulink> 
 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="intro" label=""><title></title>
<!-- dummy section to force TOC on page by itself -->
<!-- DO NOT REMOVE! please ;) -->
<para> </para>
</sect1>

<!--   ~~~~~       New section      ~~~~~     -->


<!--   ~~~~~       New section      ~~~~~     -->
  <sect1 label="1" 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 an Open Source project 
     and licensed under the GPL. 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,
     and porting, are all important jobs as well.
    </para>
  </sect1>

  <!--   ~~~~~       New section      ~~~~~     -->
  <sect1 id="quickstart"><title>Quickstart to Privoxy Development</title>
    <para>
      You'll need an account on <ulink
      url="http://sourceforge.net/">Sourceforge</ulink> to support our
      development.  Mail your ID to <ulink
      url="mailto:developers@privoxy.org">the list</ulink> and wait until a
      project manager has added you.
    </para>
    <para>
      For the time being (read, this section is under construction), please
      refer to the extensive comments in the source code.
    </para>
  </sect1>

  <!--   ~~~~~       New section      ~~~~~     -->
  <sect1 id="cvs"><title>The CVS Repository</title>
    <para>
      If you intend to help us with programming, documentation or packaging
      you will need write access to our holy grail, the CVS repository.
      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>cvs.ijbswa.sourceforge.net</literal>, the repository is
        called <literal>ijbswa</literal>, and the source tree module is called
        <literal>current</literal>.
      </para>
    </sect2>

    <sect2 id="cvscommit"><title>CVS Commit Guideline</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. We therefore ask anyone with CVS access to strictly adhere to the
        following guidelines:
        <itemizedlist>
          <listitem><para>
            Never (read: <emphasis>never, ever</emphasis>) be tempted to commit
            that 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 commited (e.g. when changing the signature of a function),
            then commit all files one after another, without long delays in beween.
            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 (see below).
          </para></listitem>
        </itemizedlist>
      </para>
    </sect2>

    <sect2 id="cvswhenask"><title>Discussing Changes First</title>
      <para>
        We don't have a too formal policy on this, just use common sense. Hints: If it is..
        <orderedlist numeration="arabic">
          <listitem><para>
            ..a bugfix / clean-up / cosmetic thing: shoot
          </para></listitem>
          <listitem><para>
            ..a new feature that can be turned off: shoot
          </para></listitem>
          <listitem><para>
            ..a clear improvement w/o side effects on other parts of the code: shoot
          </para></listitem>
          <listitem><para>
            ..a matter of taste: <ulink url="mailto:developers@privoxy.org">ask the list</ulink>
          </para></listitem>
          <listitem><para>
            ..a major redesign of some part of the code: <ulink url="mailto:developers@privoxy.org">ask
            the list</ulink>
          </para></listitem>
        </orderedlist>
      </para>
      <para>
        Note that near a major public release, we get a bit more cautious - if
        unsure, it doesn't hurt to ask first. There is always the possibility
        to submit a patch to the <ulink
        url="http://sourceforge.net/tracker/?atid=311118&amp;group_id=11118&amp;func=browse">patches
        tracker</ulink> instead.
      </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>privoxy.1</citetitle> (man page) files are also now maintained
    as Docbook SGML. The finished files are all in the top-level source
    directory are generated files! Also, <filename>index.html</filename>, the
    <application>Privoxy</application> home page, is maintained as SGML.
    <emphasis>DO NOT edit these directly</emphasis>. Edit the SGML source, or
    contact someone involved in the documentation (at present Stefan and
    Hal).
    </para> 
    <para>
     Other, less formal documents (e.g. <filename>LICENSE</filename>,
     <filename>INSTALL</filename>) are maintained as plain text files in the
     top-level source directory. At least for the time being.
    </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 now being kept in CVS under 
     <filename>doc/webserver/*</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>).                 
      </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>&lt;title&gt;My
 Title&lt;/title&gt;</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>&lt;title&gt;</literal> element, and at least one 
 <literal>&lt;para&gt;</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>&lt;para&gt;&lt;/para&gt;</emphasis>, paragraph delimiter. Most 
      text needs to be within paragraph elements (there are some exceptions).
    </member>
    <member>
      <emphasis>&lt;emphasis&gt;&lt;/emphasis&gt;</emphasis>, the stylesheets
      make this italics.
    </member>
    <member>
      <emphasis>&lt;filename&gt;&lt;/filename&gt;</emphasis>, files and directories.
    </member>
    <member>
      <emphasis>&lt;command&gt;&lt;/command&gt;</emphasis>, command examples.
    </member>
    <member>
      <emphasis>&lt;literallayout&gt;&lt;/literallayout&gt;</emphasis>, like 
      <literal>&lt;pre&gt;</literal>, more or less.
    </member>
    <member>
      <emphasis>&lt;itemizedlist&gt;&lt;/itemizedlist&gt;</emphasis>, list with bullets.
    </member>
    <member>
      <emphasis>&lt;listitem&gt;&lt;/listitem&gt;</emphasis>, member of the above.
    </member>
    <member>
      <emphasis>&lt;screen&gt;&lt;/screen&gt;</emphasis>, screen output, implies 
      <literal>&lt;literallayout&gt;</literal>.
    </member>
    <member>
      <emphasis>&lt;ulink url="example.com"&gt;&lt;/ulink&gt;</emphasis>, like 
      HTML <literal>&lt;a&gt;</literal> tag.
    </member>
    <member>
      <emphasis>&lt;quote&gt;&lt;/quote&gt;</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://www.bureau-cornavin.com/opensource/crash-course/">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>
 &lt;para&gt;
  Some text goes here.
 &lt;/para&gt;
       </literallayout>
       Tags marking individual words, or few words, should be in-line:
       <literallayout>
  Just to &lt;emphasis&gt;emphasize&lt;/emphasis&gt;, 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>
 &lt;para&gt;
  &lt;itemizedlist&gt;
   &lt;para&gt;
    &lt;listitem&gt;
      Some text goes here in our list example.
     &lt;/listitem&gt;
   &lt;/para&gt;
  &lt;/itemizedlist&gt;
 &lt;/para&gt;
       </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 
     &lt;comment&gt; element, or the &lt;!--  --&gt; style comment 
     familiar from HTML. (Note in Docbook v4.x &lt;comment&gt; is 
     replaced by &lt;remark&gt;.)
    </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, and HTML, but PDF, and others is always a 
    future possibility. Be careful with URLs (&lt;ulink&gt;), and avoid 
    this mistake:
   </para>
   <para>
     My favorite site is &lt;ulink url="http://example.com"&gt;here&lt;/ulink&gt;.
   </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 &lt;ulink url="http://example.com"&gt;example.com&lt;/ulink&gt;.
   </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>&lt;!entity supported SYSTEM "supported.sgml"&gt;</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>&#38;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 "aVariable is equal to bVariable".
    Instead explain why aVariable should be equal to the bVariable.
    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 ( PageLength() > 1024 )
{
    ... "block" the page up ...
}

/* if page size is small, send it in blocks */
if ( PageLength() > 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 ( thisVariable == thatVariable )
{
   DoSomethingVeryImportant();
}


/* unfortunately, this may not */
if ( thisVariable == thatVariable )
{
   DoSomethingVeryImportant();
}


if ( thisVariable == thatVariable ) /* this may not either */
{
   DoSomethingVeryImportant();
}</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 ( thisVariable == thatVariable )
{
   DoSomethingVeryImportant();
}

if ( thisVariable == thatVariable ) /*can you see me?*/
{
   DoSomethingVeryImportant(); /*not easily*/
}


/*********************************************************************
 * But, the encouraged exceptions:
 *********************************************************************/
int urls_read     = 0;     /* # of urls read + rejected */
int urls_rejected = 0;     /* # of urls rejected */

if ( 1 == X )
{
   DoSomethingVeryImportant();
}


short DoSomethingVeryImportant(
   short firstparam,   /* represents something */
   short nextparam     /* represents something else */ )
{
   ...code here...

}   /* -END- DoSomethingVeryImportant */
</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 )
{
   DoSomethingVeryImportant();
   ...some long list of commands...
} /* -END- if x is 1 */

or:

if ( 1 == X )
{
   DoSomethingVeryImportant();
   ...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 )
{
   DoSomething();
   DoSomethingElse();
}</programlisting>

    <para><emphasis>Instead of:</emphasis></para>

    <para>if ( this == that ) DoSomething(); DoSomethingElse();</para>

    <para>or</para>

    <para>if ( this == that ) DoSomething();</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 firstValue   = 0;
int someValue    = 0;
int anotherValue = 0;
int thisVariable = 0;

if ( thisVariable == thatVariable )

firstValue = oldValue + ( ( someValue - anotherValue ) - 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>
aStruct->aMember;
aStruct.aMember;
FunctionName();</programlisting>

    <para><emphasis>Instead of:</emphasis> aStruct -> aMember; aStruct . aMember;
    FunctionName ();</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( retCode );

}   /* -END- function1 */


int function2( ... )
{
}   /* -END- function2 */
</programlisting>

    <para><emphasis>Instead of:</emphasis></para>

    <para>int function1( ... ) { ...code... return( retCode ); } 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 anShort = 0;
float aFloat  = 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 arrayPtr[20] causes a SIGSEV vs.
    arrayPtr[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>
ShouldWeBlockThis();
ContainsAnImage();
IsWebPageBlank();
</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 &lt; blockListLength(); 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 blockListLength() 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 blockListLength() 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 = blockListLength();

for ( size_t cnt = 0; cnt &lt; len; cnt ++ )
{
   ....
}</programlisting>

    <para><emphasis>Exceptions:</emphasis> if the value of blockListLength() *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 &lt;iostream.h&gt;     /* 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 &lt;sys/fileName.h&gt;
</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 ABEND 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. Try to avoid using size_t if
    you can.</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 1 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/destuctor 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</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[] = "$Id: developer-manual.sgml,v 1.38 2002/04/29 02:20:31 hal9 Exp $";
/*********************************************************************
 *
 * File        :  $S<!-- Break CVS Substitution -->ource$
 *
 * Purpose     :  (Fill me in with a good description!)
 *
 * Copyright   :  Written by and Copyright (C) 2001 the SourceForge
 *                Privoxy team. http://www.privoxy.org/
 *
 *                Based on the Internet Junkbuster originally written
 *                by and Copyright (C) 1997 Anonymous Coders and
 *                Junkbusters Corporation.  http://www.junkbusters.com
 *
 *                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.
 *
 * 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 "$Id: developer-manual.sgml,v 1.38 2002/04/29 02:20:31 hal9 Exp $"
/*********************************************************************
 *
 * File        :  $S<!-- Break CVS Substitution -->ource$
 *
 * Purpose     :  (Fill me in with a good description!)
 *
 * Copyright   :  Written by and Copyright (C) 2001 the SourceForge
 *                Privoxy team. http://www.privoxy.org/
 *
 *                Based on the Internet Junkbuster originally written
 *                by and Copyright (C) 1997 Anonymous Coders and
 *                Junkbusters Corporation.  http://www.junkbusters.com
 *
 *                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.
 *
 * 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&amp;group_id=11118&amp;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 mailinglist (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, 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 bugfixes 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 inrcemented. 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>
          </listitem>
        </itemizedlist>
    </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>.
       </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>
        If the HTML documentation is not in sync with the SGML sources
        you need to regenerate and upload it to the webserver. (If in
        doubt, just do it.) See the Section "Updating the webserver" in
        this manual for details.
       </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>.
     </para>
      
     <para>
      <programlisting>
  mkdir dist # delete or choose different name if it already exists
  cd dist
  cvs -d:pserver:anonymous@cvs.ijbswa.sourceforge.net:/cvsroot/ijbswa login
  cvs -z3 -d:pserver:anonymous@cvs.ijbswa.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>
  
    <para>
     Please find additional instructions for the source tarball and the
     individual platform dependent binary packages below.
    </para>

    <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 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@cvs.ijbswa.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@cvs.ijbswa.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, run:
        </para>
        <para>
        <programlisting>
  cd current
  autoheader && autoconf && ./configure
</programlisting>
        </para>
        <para>
        Then do FIXME.
        </para>
    </sect3>

    <sect3 id="newrelease-macosx"><title>Mac OSX</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 OSX setup module:
        </para>
        <para>
        <programlisting>
  cvs -z3 -d:pserver:anonymous@cvs.ijbswa.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 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>
  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>
     Once this done go to <ulink url="http://sourceforge.net/project/admin/editpackages.php?group_id=11118">http://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 change log.
     </para>
   </sect2>

  </sect1>
  
  <!--   ~~~~~       New section      ~~~~~     -->
  <sect1 id="webserver-update"><title>Update the Webserver</title>
   <para>
    When updating the webserver, please follow these steps to make
    sure that no broken links, incosistent contents or permission
    problems will occur:
   </para>
   <para>
    If you have changed anything in the documentation source SGML files,
    do:
   </para>
   <para>
    <programlisting>
  make dok # (or make redkat-dok 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> and
    <filename>doc/webserver/index.html</filename> automatically.
   </para>
   <para>
    If you changed the manual page source, 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. 
    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? 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.
   </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      ~~~~~     -->
  <sect1 id="copyright"><title>Copyright and History</title>

<sect2><title>Copyright</title>
<!-- Include copyright.sgml -->
 &copyright;
<!-- end -->
</sect2>

<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 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 OSX 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>