forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject cvs commit: xml-forrest/src/documentation/skins/forrest-site/xslt/html document2html.xsl
Date Wed, 22 May 2002 12:08:03 GMT
stevenn     02/05/22 05:08:03

  Modified:    src/documentation/content/xdocs book.xml index.xml
  Added:       src/documentation/content/xdocs primer.xml
               src/documentation/resources/images page-areas.png
  initial version of the Forrest primer and some more doc2html patches
  Revision  Changes    Path
  1.8       +1 -0      xml-forrest/src/documentation/content/xdocs/book.xml
  Index: book.xml
  RCS file: /home/cvs/xml-forrest/src/documentation/content/xdocs/book.xml,v
  retrieving revision 1.7
  retrieving revision 1.8
  diff -u -r1.7 -r1.8
  --- book.xml	19 May 2002 18:56:41 -0000	1.7
  +++ book.xml	22 May 2002 12:08:02 -0000	1.8
  @@ -21,6 +21,7 @@
     <menu label="Get Involved">
  +    <menu-item label="Forrest Primer" href="primer.html"/>
       <menu-item label="Contributing" href="contrib.html"/>
       <menu-item label="CVS" href=""/>
       <menu-item label="Mail Lists" href="mail-lists.html"/>
  1.6       +11 -8     xml-forrest/src/documentation/content/xdocs/index.xml
  Index: index.xml
  RCS file: /home/cvs/xml-forrest/src/documentation/content/xdocs/index.xml,v
  retrieving revision 1.5
  retrieving revision 1.6
  diff -u -r1.5 -r1.6
  --- index.xml	21 May 2002 21:16:16 -0000	1.5
  +++ index.xml	22 May 2002 12:08:02 -0000	1.6
  @@ -22,16 +22,19 @@
         <p>The Forrest developers -
  -      href=""></link></p>

  +       href=""></link></p>

       <section title="Project Status"> 
  -      <p>The Forrest project is new and needs your help to get started. Please
  -        see the <link href="faq.html">FAQ</link> and join the discussion on
  -      <link
  -      href=""></link>.</p>
  -      <p>There is also a fledgling <link href="dreams.html">Dream List</link>
  -      to help provide some direction.</p> 
  +      <p>The Forrest project is new and needs your help to get started.</p>
  +      <ul>
  +        <li>Read our <link href="primer.html">Forrest Primer</link> to
  +      understand what Forrest is and how you can use / contribute to it.</li>
  +      <li>Please see the <link href="faq.html">FAQ</link> and join the
  +      discussion on <link
  +      href=""></link>.</li>
  +      <li>There is also a fledgling <link href="dreams.html">Dream List</link>
  +      to help provide some direction.</li>
  +      </ul> 
  1.1                  xml-forrest/src/documentation/content/xdocs/primer.xml
  Index: primer.xml
  <?xml version="1.0"?>
  <!DOCTYPE document PUBLIC "-//APACHE//DTD Documentation V1.1//EN" "document-v11.dtd">
      <title>The Forrest Primer</title> 
      <subtitle>Don't panic!</subtitle><authors><person
        name="Steven Noels" email=""/></authors> 
      <abstract>Forrest is a so-called
        project that will have a broad impact on <link
        href=""></link> projects. This document
        helps you to better understand the vision and scope of Forrest, so that you
        learn what to expect (or not) from it, and eventually will help you discovering
        places where your contribution could be valuable to all of us.</abstract> 
      <note>This document is work-in-progress and will be updated to reflect the
        actual progress of the Forrest project. As such, the primer will be enhanced,
        restructured and edited to provide an accessible entry point for new
        Forresteers. Please send all comments, patches and suggestions to
        <link href="">the Forrest
      <section title="History"> 
        <p>Forrest has come into existence because of the abysmal state of the
          <link href=""></link> website in
        with other open source community sites such as Sourceforge. The old site had no
        consistent visual look and feel, which was largely due to each and every
        sub-project managing its own site. Furthermore, much information which could
        potentially support community-based open source development was hidden inside
        CVS repositories, mailing lists or word of mouth. Once we experienced the
        usefullness of cross-project collaboration supported by the Jakarta
        <link href="">Gump</link> project, we reckoned
        having a single application responsible for the management of the site could be of benefit to our visitors. And if we added
        aggregated access to other available resources such as download stats or
        mailing list archives, the new website could be a true
        information clearinghouse for interested parties, both users and contributors
        <p>The Forrest vision was articulated by Stefano Mazzocchi and Sam Ruby,
          both long-time contributors to Apache projects, in the beginning of 2002, and
          was rapidly picked up by a bunch of other <link
        href="who.html">contributors</link> as well, after a headstart by Nicola
        Barozzi. So here we are, plenty of work-in-progress to erect what eventually
        will become a true community website infrastructure for Apache open source
      <section title="What is Forrest"> 
        <p>Forrest is a framework that supports the cross-project generation and
          management of development project websites using Cocoon as its XML publishing
          framework. It not only provides access to project documentation, but also to
          other types of information open source developers daily depend upon: source
          code repositories, mailing lists, contact info and the like. It aggregates all
          these resources and publishes them on a regular basis to a website, ensuring a
          consistent look and feel using skins implemented with XSLT stylesheets. While
          Forrest's primary focus is XML Apache project websites, it can be adapted to
          other community development projects as well, as long as they are willing to
          commit to proven best practices such as Ant for build automation, CVS for
          source code control and XML as a documentation source format.</p> 
        <p>Forrest is currently based on an
          <link href="">Ant</link>-based project
        system called <link href="">Centipede</link>
        that drives a <link href="">Cocoon</link>-based
        document publication system. It contains a set of standard XML document type
        declarations (DTD's) for project documentation, and different 'skins'
        consisting of XSLT stylesheets that produce HTML renditions of XML documents
        using these DTD's.</p> 
        <p>The primary mode of operations for Forrest will be as follows:</p>

        <note>This process isn't quite ready for prime time yet, but it gives you
          an idea where we are heading at. Website generation with skins currently works,
          try using the <code>docs</code> target when invoking the <code>build</code>
          script. Add a <code></code> property when invoking the build
script to
          experience Forrest skins: <code>build{.bat|.sh}
;thenameoftheskintouse&gt; docs</code>. Read our <jump
          href="cvs">CVS crash course</jump> to get hold of the current codebase
          start playing with it.</note> 
          <li>Forrest will harvest documentation and related source files from
            the different projects that use Forrest for their website, presumably using
            CVS, configured by a project descriptor file. It will do so on an automated
            basis, several times a day.</li> 
          <li>Forrest uses Cocoon to generate an HTML rendition of each project's
            website, configured by a generic sitemap. This process will be static, the
            result of a Forrest run is a collection of HTML documents and related images or
            stylesheets comprising the project's website. The impact Forrest has on the
            participating projects should be minimal, i.e. one should simply author XML
            documents, put them in a well-specified filesystem hierarchy, and Forrest will
            do its work.</li> 
          <li>Forrest will enrich the provided documentation source files with
            common information: a cross-project navigation structure (and rendition, of
            course), and useful 'community indicators' such as download statistics, number
            of contributors with commit access, ...</li> 
          <li>If the individual project build runs are successful, the project's
            website is automagically (re-)published to the (Apache) website, also several
            times day.</li> 
        <p>The Forrest website and the overall website are
          maintained and published using the same mechanism.</p> 
      <section title="Forrest roles"> 
        <p>Depending on your interests, your involvement with Forrest may vary,
          hence your <em>role</em>. We currently envision three different roles:</p>

          <li><strong>User</strong> you want or need to use Forrest for
            project because it uses Forrest to manage its documentation.</li> 
          <li><strong>Adaptor</strong> you want to adapt Forrest to support
            individual project needs, presumably outside the XML Apache context, building
            your own skins or DTD's and the like.</li> 
          <li><strong>Contributor</strong> you are a fledgling Forresteer
            want to contribute to the further development of it. If your contributions are
            valuable and in true community spirit, you can possibly gain commit access to
            the Forrest CVS repository and become an Apache committer.</li> 
        <p>Depending on your role, your potential area of interest in Forrest
          will vary:</p> 
            <td>Forrest DTD's and documentation filesystem hierarchy (Cocoon
            <td>+ skin system and build environment</td> 
            <td>+ the Forrest codebase and runtime environment</td> 
      <section title="Getting hold of Forrest through CVS" id="cvs"> 
        <section title="System requirements"> 
          <p>Forrest requires the following systems to be already installed on
            your system:</p> 
            <li><em>Java Virtual Machine</em> A Java virtual machine must
              present. Forrest has been tested against the latest Sun 1.3 JDK.</li>

        <section title="Getting Forrest"> 
          <p>You have to retrieve Forrest from its CVS repository - there are no
            binary, nor source, nor snapshot distributions yet. Some help, courtesy our
            friends of the Cocoon project:</p> 
        <section title="Step-by-step cvs instructions for Windows"> 
            <li>Download a recent release of WinCVS (homepage is
              <link href=""></link>);
          <li>Install it;</li> 
          <li>Start it;</li> 
          <li>Click on Admin-&gt;Preferences;</li> 
          <li> In "Enter the CVSROOT:" enter
            "<code></code>" (without
          <li>In "Authentication:" choose "passwd file on the cvs server";</li>

          <li>Click "Ok";</li> 
          <li>Click Admin-&gt;Login;</li> 
          <li> When asked for the password: answer "<code>anoncvs</code>"
            (without quotes);</li> 
          <li> Click "Create-&gt;Checkout module";</li> 
          <li>Module name and path on the server is "<code>xml-forrest</code>"
            (no quotes);</li> 
          <li>Choose a dir to put the source code in;</li> 
          <li>Click "Ok";</li> 
          <li>If everything goes well, messages will start to appear in the log
          <li>Wait until you see "<code>*****CVS exited normally with code
            0*****</code>" in the log window;</li> 
          <li>The Forrest source is now on your harddrive.</li> 
        <section title="Step-by-step cvs instructions for Unix"> 
            <li>Make sure you have a CVS client package installed on your Unix
            <li>Start the shell of your choice.</li> 
            <li>Enter "<code>cvs -d

            <li>When asked for the password: answer "<code>anoncvs</code>".</li>

            <li>Enter "<code>cvs -d
     -z3 checkout
              xml-forrest</code>". This will create a directory called
              "<code>xml-forrest</code>" where the Forrest source will be stored.</li>

            <li>Wait until cvs has finished.</li> 
            <li>The Forrest source is now on your harddrive.</li> 
          <p>In case you want to update your Forrest source tree to the current
            version, change to the "<code>xml-forrest</code>" directory and invoke
            "<code>cvs -z3 update -d -P</code>".</p> 
      <section title="Forrest distribution"> 
        <p>Once you retrieved Forrest from its CVS repository, you'll end up with
          a filesystem hierarchy similar to this inside the <code>xml-forrest</code>
        <warning>This is highly volatile information!</warning> 
        <source><![CDATA[+---legal                             various licenses for
projects included with Forrest
  +---lib                               jar library
  |   +---contrib                       unused
  |   +---documentation                 main Forrest source tree
  |   |   +---content                   content of the Forrest website
  |   |   |   +---xdocs                 Forrest website XML documents
  |   |   +---library                   common components (not skin-specific)
  |   |   |   +---xslt                  document format transformers (faq & howto ->
  |   |   +---resources                 project documentation specific resources
  |   |   |   +---images
  |   |   +---skins                     Forrest skins
  |   |       +---avalon-site
  |   |       +---forrest-site          the future skin
  |   |       |   +---css               Cascading Stylesheets
  |   |       |   +---images            skin-specific images
  |   |       |   +---xslt              the skin stylesheets (per medium)
  |   |       |       +---fo
  |   |       |       +---html          html rendering skins
  |   |       +---jakarta-site
  |   |       +---scarab-site
  |   |       +---xml-apache-site
  |   +---examples                      unused
  |   +---java                          unused
  |   +---resources                     project resources
  |   |   +---charts                    charting trials
  |   |   +---layout                    HTML page mock-ups
  |   |   |   +---resources
  |   |   |
  |   |   |       +---images
  |   |   +---schema                    Forrest DTD's
  |   |       +---dtd
  |   |       +---entity
  |   +---scratchpad                    unused
  |   +---targets                       unused
  |   +---testcases                     unused
  +---tools                             Forrest build environment (Centipede)
        <p>The <code>xml-forrest</code> home directory consists of the main
          build script (<code>build.xml</code>) and platform-specific batch files/shell
          scripts to invoke it. Forrest comes with Ant included, so you don't need to
          install Ant separately.</p> 
        <p>Running Forrest is a batch operation you can start using the provided
          <code>build.{sh|bat} &lt;targetname&gt;</code>. The current
main targets
          <li><strong><code>docs</code></strong> - generates
an HTML rendition of
            the Forrest website using the default <code>forrest-site</code> skin</li>

          <li><strong><code>clean</code></strong> - cleans out
            <code>build</code> directory</li> 
          <li><strong><code>webapp</code></strong> - for those
who can't resist
            running Forrest live instead of its commandline invocation, this target builds
            a WAR file you can deploy in your servlet container (currently only tested for
            Tomcat 4.0.1). Mount-point of the web application will be
        <p>After a build run, Forrest creates a <code>build</code> directory.
          can find the generated website in <code>build/xml-forrest/docs/</code>.
          also creates a <code>tools/tmp/anttasks/</code> upon its first invocation.
          These are Centipede-specific compiled Ant tasks.</p> 
      <section title="The Forrest DTD's"> 
        <p>Forrest is the reference repository for the XML Apache documentation
          DTD's. Special care is taken to provide a set of modular, extensible and
          well-maintained DTD's for project documentation purposes. This modularity is
          ensured using the <link
        href="">OASIS catalog</link>
        mechanism, extensive use of external parameter entities and an entity resolver
        capable of resolving entities through the aforementioned catalog mechanism. For
        the docheads amongst us, this means we adhere to the strict use of
        <code>PUBLIC</code> entity identifiers both in document instances and
        <p>We have currently identified the following document types:</p> 
          <li>General documents (<code>document-v11.dtd</code>),</li>

          <li>How-To's (<code>howto-v10.dtd</code>),</li> 
          <li>Collections of FAQs (<code>faq-v11.dtd)</code>.</li>

        <p>Some work is also under its way for other document types, in close
          collaboration with the Cocoon project. You will also find some older document
          types such as <code>changes</code>, <code>javadoc</code>,
          <code>specification</code> and <code>todo</code>, which
are currently under
          consideration for automatic generation and maintenance using Gump or Centipede
          descriptors and the like. DTD's will be subject of serious version management
          as soon as Forrest has a 1.0 release: they are made to depend upon.</p> 
        <p>The DTD's are located in <code>src/resources/schema/dtd</code>,
          also refer to some character entity collections stored in
          <code>src/resources/schema/entity</code>. You should read these together
          the declarations as found in the <code>src/resources/schema/catalog</code>
          file. Take special care using the correct <code>PUBLIC</code> identifiers
          the DTD declaration of your instances:</p> 
        <source><![CDATA[<?xml version="1.0"?>
  <!DOCTYPE document PUBLIC "-//APACHE//DTD Documentation V1.1//EN" "document-v11.dtd">
        <p>The exact location of the DTD for validation purposes is obtained by
          the entity resolver evaluating the mapping scheme as defined in the
          <code>catalog</code> file. This makes sure that you can move and re-arrange
          your document instances apart from your DTD files. Later on, the DTD's will be
          web-accessible from the Forrest website for your perusal.</p> 
      <section title="Forrest site generation using Cocoon"> 
        <p>The <code>docs</code> target of the Forrest build environment
          Cocoon as a command-line application to generate an HTML rendition of the
          project's documentation. It is not within the scope of this document to explain
          the Cocoon internals, please read the its own
          <link href="">documentation</link> to fully
        understand the power of Cocoon.</p> 
        <p>Cocoon's site rendition behaviour is configured in a so-called
          <em>sitemap</em>, a switchboard that binds URL's to an XML processing
          This pipeline typically consists of a Generator, one or more Transformers and a
          Serializer. Forrest also makes use of Cocoon's aggregation capabilities that
          merge multiple pipelines into one resulting output document.</p> 
        <p>A typical page generated using Forrest looks like this:</p>
        <figure src="../../resources/images/page-areas.png" height="291" width="336"
         alt="Pages areas"/> 
        <p>As you might see, this page is currently composed of two XML sources
          which are transformed by a different XSLT stylesheet, aggregated by Cocoon with
          a post-aggregation stylesheet adding the overall page grid and look&amp;feel.
          This simple example is handled by the following <em>sitemap</em> snippets
        <source><![CDATA[<map:match pattern="*.html">
    <map:aggregate element="site">
      <map:part src="cocoon:/book-{1}.xml"/>
      <map:part src="cocoon:/body-{1}.xml" label="content"/>
    <map:call resource="skinit">
      <map:parameter name="type" value="site2xhtml"/>
        <source><![CDATA[<map:match pattern="**book-**.xml">
    <map:generate src="content/xdocs/{1}book.xml"/>
    <map:call resource="skinit">
      <map:parameter name="type" value="book2menu"/>
        <source><![CDATA[<map:match pattern="body-**.xml">
    <map:generate src="content/xdocs/{1}.xml"/>
    <map:call resource="skinit">
      <map:parameter name="type" value="document2html"/>
        <source><![CDATA[<map:resource name="skinit">
    <map:transform src="skins/@skin@/xslt/html/{type}.xsl">
      <map:parameter name="isfaq" value="{isfaq}"/>
        <p>The pipeline flow basically is:</p>
          <li>A URL matched by the <code>*.html</code> pattern is fetched.</li>
          <li>The response will be the result of aggregating two
          <li>A URL matching the <code>**book-**.xml</code> pattern, which
is the
            result of reading the <code>book.xml</code> document and running that
            the <code>book2menu</code> stylesheet (which produces an HTML fragment
            comprising the site navigation, the red area in the image.</li>
          <li>A URL matching the <code>body-**.xml</code> pattern, corresponding
            to <code>*.xml</code> transformed using the <code>document2html</code>
        <li>The aggregation result is then transformed using the
          <code>site2xhtml</code> stylesheet which adds the cherries to the cake.</li>
        <p>These <em>skin-specific</em> stylesheets are located in
so if you
          want to add your own skin, this is the place to be. Apart from these, there
          exist a number of other stylesheets located in
          <code>src\documentation\library\xslt</code>, more importantly:</p>
          <li><code>faq2document</code>: transforms documents following
            <code>faq-v11</code> DTD to <code>document-v11</code>
          <li><code>howto2document</code>: transforms documents following
            <code>howto-v10</code> DTD to <code>document-v11</code>
          <li>and some others.</li>
        <p>As you see, all documents, regardless of their original DTD, are
          transformed to the <code>document</code> DTD prior to rendition. This
          alleviates the burden of adding new skins to implementing 3 simple stylesheets:
          <code>book2menu</code>, <code>document2html</code> and
      <section title="Where we are heading at"> 
        <p>We have been explaining so far where we are now and what already
          works. The purpose of this document however is to attract newcomers and entice
          them to start contributing to Forrest. We have a decent generation system for
          static project documentation, a nice set of skins and some simple but effective
          DTD's. Our goals however are much more ambitious: we have compiled a
          <link href="dreams.html">dream list</link> that lists most of them.</p>
          <li>Our first ambition is to support the project site generation and
            maintenance of other Apache projects in an automated manner, starting with our
            own website as a showcase. We are in the process of setting up the shell
            scripts and Ant tasks for this and will assist projects transitioning to
          <li>As it is often the case with collaborative open source development,
            there is no formal planning nor task assignments, and we will stick to that
            practice. We have however compiled a number of functional work areas:</li>
            <th>URI Namespace Management</th>
            <td>Forrest will offer access to a broad set of information resources
              using durable URI's: please review
              <link href="">Tim Berners-Lee</link>'s
        and <link href="">Jakob
        Nielsen</link>'s opinion on this. We need a unified URI Namespace management
        approach, bearing in mind mirroring and 'hackable' URI's.</td>
          <td>We currently have a nice set of skins which should be solidified.
            Furthermore, we need some serious finetuning of the <code>forrest-site</code>
            skin that will become the new look&amp;feel.</td>
          <th>Aggregation<br/>and Syndication</th>
          <td>We plan to aggregate on a per-project basis a number of relevant
            developer resources, such as project-related news, download statistics,
            committer bio pages (with photos!), navigable source code listings and the
            like. Some of these resources need to be made available across content
            syndication methods such as <link
          <th>Build Management</th>
          <td>Fool-proof automation of Forrest runs and site publication using
            secure transfer methods and <code>cron</code> jobs.</td>
          <th>Document Types</th>
          <td>Expanding the collection of DTD's, documenting them using formal
            How-To's and example documents.</td>
          <td>Formation of an editorial team for the main website,
            working in close collaboration with the
            <link href="">PMC</link> and the
        sub-project leads.</td>
          <td> Forrest needs to coexist with existing cross-project collaboration
            tools such as <link href="">Gump</link>,
        <link href="">Scarab</link> and
        <link href="">Eyebrowse</link> and provide
        integrated access to them.</td>
          <th>Authoring support</th>
          <td>Supporting document authors with preconfigured XML editing
          <th>Content Management</th>
          <td>Establish an efficient content management practice, supporting
            versioning, remote access and work flow, presumably supported by a CMS such as
            <link href="">Slide</link>.</td>
          <th>Information Accessibility</th>
          <td>We need to be accessible using a wide range of browsing devices
            operating on different platforms. Special care should be taken to support the
            <link href="">WAI</link> guidelines.</td>
      <section title="Where you can help">
        <p>By now, you should have a better understanding of Forrest (if that is
          not the case, consider contributing to this document<img
          src="../../resources/images/wink_smile.png" height="19" width="19"
          alt=";-) (c) MSN Messenger"/> to start with). We need more people to get the
          job done. Forrest is a fun project to work on, and there's something in it for
          all of us:</p> 
          <li>XML docheads with skills for document analysis and DTD's
          <li>Cocoon developers creating custom Cocoon components connecting
            Forrest with external resources</li> 
          <li>Graphical whizzkids for true cross-browser HTML/CSS
          <li>People who believe XSLT will bring peace to earth (it will, but
            keep that quiet)</li> 
          <li>Ant wizards able to compete with Nicola and Stefan</li> 
          <li>Unix shell scripting / CVS / cron gurus, preferably bearded</li>

        <p>Just drop us a line at <link
      <p>That's all, folks.</p> 
          <th>Revision history</th>
          <td>Initial version, Steven Noels,</td>
  1.1                  xml-forrest/src/documentation/resources/images/page-areas.png
  	<<Binary file>>
  1.1                  xml-forrest/src/documentation/resources/images/wink_smile.png
  	<<Binary file>>
  1.9       +10 -8     xml-forrest/src/documentation/skins/forrest-site/xslt/html/document2html.xsl
  Index: document2html.xsl
  RCS file: /home/cvs/xml-forrest/src/documentation/skins/forrest-site/xslt/html/document2html.xsl,v
  retrieving revision 1.8
  retrieving revision 1.9
  diff -u -r1.8 -r1.9
  --- document2html.xsl	21 May 2002 12:23:30 -0000	1.8
  +++ document2html.xsl	22 May 2002 12:08:02 -0000	1.9
  @@ -26,14 +26,16 @@
  -      <xsl:choose>
  -        <xsl:when test="normalize-space(header/title)!=''">
  -          <h2>
  -            <xsl:value-of select="header/title"/>
  -          </h2>
  -        </xsl:when>
  -        <xsl:otherwise>&#160;</xsl:otherwise>
  -      </xsl:choose>
  +      <xsl:if test="normalize-space(header/title)!=''">
  +        <h2>
  +          <xsl:value-of select="header/title"/>
  +        </h2>
  +      </xsl:if>
  +      <xsl:if test="normalize-space(header/subtitle)!=''">
  +        <h3>
  +          <xsl:value-of select="header/subtitle"/>
  +        </h3>
  +      </xsl:if>
         <xsl:if test="header/authors">
           <font size="-2">

View raw message