Page History

...

It is possible to add simple static pages to DSpace using Manakin. This is useful for adding "one-shot" pages like About and FAQs where creating a new aspect would be overkill. There are three ways to add a static page:

1. (Pre 1.8) XSL-based method. You can override the main body template and then add a check for the url of the page you want to add. The current page's url is always present in the DRI pageMeta section as request.URI. If the current page url is one you designated for your static page, then instead of rendering the body of the DRI page normally, you can have XSL insert different content. This content can either be placed in the code of the XSL template or imported from an external source using the document() function. This method is useful if the number of extra pages you want to include is relatively small and they are not edited frequently.

...

3. Tomcat/apache-based method. Outside of Manakin or DSpace you could reroute urls to static pages using Tomcat's web.xml or the Apache configuration. This method while easy doesn't allow you to have any of the dynamic content (such as who's loged in) available on the page.

Common Gotchas

Empty self-closed elements in XHTML

4. (Dspace 1.8.1 and later). Use XSL overrides on top of StaticPage.java. In DSpace 1.8.1. a StaticPage transformer has been added and is mapped to /xmlui/page/*, which provides an easy end point to override in XSL, and not get blocked by PageNotFound. Additionally, points in the XSL have been added to allow for the XSL customizer to match against a page/something url, and then add customized XHTML directly to the page. To add a new static page, all you need to edit is dri2xhtml/structural.xsl, or Mirage/lib/xsl/core/page-structure.xsl. Below is some snippets of XSL that you need to edit that is for structural.xsl.

To customize the Page Title:

<title>
    <xsl:choose>
        <xsl:when test="starts-with($request-uri, 'page/about')">
            <xsl:text>About This Repository</xsl:text>
        </xsl:when>
        ...
</title>

To customize the breadcrumb / trail:

<ul id="ds-trail">
    <xsl:choose>
        <xsl:when test="starts-with($request-uri, 'page/about')">
            <xsl:text>About This Repository</xsl:text>
        </xsl:when>
        ...
    </xsl:choose>
</ul>

To customize the body:

<xsl:template match="dri:body">
    <div id="ds-body">
    ...
    <xsl:choose>
        <xsl:when test="starts-with($request-uri, 'page/about')">
            <div>
                <h1>About This Repository</h1>
                <p>To add your own content to this page, edit webapps/xmlui/themes/dri2xhtml/structural.xsl and add your own content to the title, trail, and body.
                   ....
                </p>
           </div>
        </xsl:when>
        ...
    </div>
</xsl:template>

See: https://jira.duraspace.org/browse/DS-1076

Common Gotchas

Empty self-closed elements in XHTML

By convention, an empty XML element, that is an element with no text value, can be represented in two ways in By convention, an empty XML element, that is an element with no text value, can be represented in two ways in XML syntax:

1. open/close pair: <element></element>
2. self-closed element: <element/>
   Both are valid XML, although the latter is used more frequently for convenience. By extension this notation is also valid in XSL and XHTML, but problems do crop up in the latter. Some browsers will treat a self-closed div element as the div opening tag, for example, placing all the elements that follow it inside that div. This can cause a great deal of headache during CSS work as your page's body and footer might end up whole subsumed under the header. This can occur either accidentally (an element that usually holds a title, for example, can get collapsed into an empty tag if a title is missing) or intentionally (when using blank "spacer" divs to keep floated elements from collapsing their containing block). In either case, the problem can be solved by making sure that the element hold at least some text (like substituting "untitled" for an item with no title) or adding a non-breaking space to the element.

...

One of the more common entities used in HTML is the non-breaking white space, used to offset text and pad empty elements. While it is defined in XHTML as     (as are about 250 other entities like © ® → © ® → etc.), the XSL processor will throw an error if those entities are used in the XSL code. The reason for this is that only the following five entities are explicitly defined in XML, mostly because the symbols are part of the syntax structure:

In order to insert other symbols into the HTML, use the ISO 8859-1 codes for those entities, i.e. U+00A0 instead of     and U+00A9 instead of © ©. A full list of entities and their corresponding codes can be found in section 24 of the W3C Recommendation for HTML4:http://www.w3.org/TR/html401/sgml/entities.html

...