Search the documentation
 Show GitHub edit links  Hide GitHub edit links
In OpenCms since: 8.5 Documented since: 9.0 Latest revision for: 9.0 Valid for OpenCms: 10.5.3

OpenCms has an inline editor that allows you to edit content inline - without opening the form-based content editor. The inline editor has to be configured in a content's formatter to make parts of the content inline editable. We summarize how to add inline editing facilities to your formatter.

What has to be configured to enable inline editing?

Inline editing has to be configured in a content's formatter. The value wrapper for each content field has the properties rdfa and rdfaAttr. The first returns the lazily initialized RDFA map for the nested sub-values, the second the RDFA annotation for the current value. These annotations are placed into HTML tags via the formatter. The page editor analysis this annotations and configures the inline editor accordingly.

Dependent on the type of the content field that should be directly editable, the place where to put the annotation differs.

  • For content fields of OpenCmsString or OpenCmsHtml the annotations must be placed at a tag directly surrounding only the unmanipulated content value.
  • For fields of other types, the annotations can be placed at arbitrary tags. For these types values cannot be edited directly inline, but, at the tag where the annotation is placed, an edit button will appear, that allows to edit the content value.
  • For content fields that are optional, or that can appear multiple times, also add and delete options are shown via RDFA annotations.
  • For whole nested contents, the annotations can also be placed at arbitrary HTML tags.

Example with all variants of placing RDFA annotations

We show you an example that makes extensive use of the inline editing facilities. It is designed to provide you an overview on what is possible.

Do not make everything editable inline. The many edit buttons will lead to confusion. So choose carefully what should be editable via the inline editor, especially if it is not directly editable text. For a production website, the example has definitely too many edit points exposed in the inline editor!
<%@page buffer="none" session="false" trimDirectiveWhitespaces="true" taglibs="c,cms"%>

<cms:formatter var="content" rdfa="rdfa" val="value">

<div class="margin-bottom-30">
<h2 class="headline" ${rdfa.Title}>${value.Title}</h2>
<c:forEach var="paragraph" items="${content.valueList.Paragraph}">
    <div ${paragraph.rdfaAttr}>
        <c:if test="${paragraph.value.Headline.isSet}">
            <div class="headline"><h3 ${paragraph.rdfa.Headline}>${paragraph.value.Headline}</h3></div>
        </c:if>
        <c:set var="rdfaFields">Link|Image${paragraph.value.Headline.isSet?'':'|Headline'}</c:set>
        <div ${paragraph.rdfa[rdfaFields]}>
            <div class="row">
                <div class='${paragraph.value.Image.exists?"col-md-8 col-sm-10 col-xs-12":"col-xs-12"}'>
                    <div ${paragraph.rdfa.Text}>${paragraph.value.Text}</div>		
                    <c:if test="${paragraph.value.Link.exists}">
                        <p><a class="btn-u btn-u-small" 
                              href="<cms:link>${paragraph.value.Link.value.URI}</cms:link>">
                             ${paragraph.value.Link.value.Text}
                        </a></p>
                    </c:if>		
                </div>
                <c:if test="${paragraph.value.Image.exists}">
                    <div class="col-md-4 col-sm-2 hidden-xs">
                        <div class="thumbnail-kenburn"><div class="overflow-hidden">
                            <cms:img src="${paragraph.value.Image}" 
                                     noDim="true" cssclass="img-responsive" />
                        </div></div>		
                    </div>
                </c:if>
            </div>			
        </div>
    </div>
</c:forEach> 
</div>

</cms:formatter>

The example formatter renders a content that has a field Title and a list of one or more Paragraphs as nested contents. Each Paragraph has an optional Headline, Image and Link (which is itself a nested content), and a mandatory Text field.

Here are the parts of the formatter relevant for inline editing:

  • Line 3: The attribute rdfa="rdfa" in the <cms:formatter> tag exposes a variable rdfa as abbreviation for content.rdfa. This is not mandatory, but a convenience feature, just as val="value" exposes value as abbreviation for content.value.
  • Line 6: The ${rdfa.Title} at the <h2> tag enables inline editing for the Title. It is mandatory that inside of the <h2> tag there is only the unmanipulated value of Title. Note that we use the rdfa variable when writing rdfa.Title exposed via the <cms:formatter> tag. We could have also written content.rdfa.Title.
  • Line 8: We use ${paragraph.rdfaAttr} to get the correct RDFA annotation for the paragraph. Here it is easier to use the property rdfaAttr, instead of content.rdfa.Paragraph[status.index], where status would have to be exposed via <c:forEach ... statusVar="status">.
  • Line 10: Here we make the paragraph's headline editable directly. It's similar to line 6, where the title is made directly editable.
  • Line 13: We set the RDFA annotations for Link, Image and, if the headline is not set and thus directly editable (see line 10), for Headline . At the annotated <div>, edit points for all two (or three) content fields will appear. Note that rdfaFields is set in line 12 to a pipe-separated list of content fields. Such a pipe separated list is the only way to add more than one RDFA annotation to an HTML tag. Writing <div ${paragraph.rdfa.Link} ${paragraph.rdfa.Image}> instead, does not work.
    A further interesting point here is, that the RDFA annotation for the headline is set differently, depending on the existence of a headline.
  • Line 16: Here the Text is printed in a <div> and made directly editable. Text is of type OpenCmsHtml. Here again, make sure that the value of Text is the only content of the annotated <div>.

What is not directly present in the example, is the possibility to navigate through the content using the attribute value and when you are at the right place, use rdfa. For example, content.value.NestedContent.rdfa.NestedTitle is a valid expression.

The following figure shows the resulting inline editor for (a slightly more complex variant of) the above formatter.


All texts rendered by the formatter can be edited inline. Moreover, a lot of edit points and plus symbols appear at the right-hand side of the editor. These are the shown because of RDFA annotations for content fields that are not directly editable. In particular, the edit point we hover over belongs to the first paragraph, the plus directly below to this paragraphs headline (that is not provided in the content). Furthermore, we get and edit point to edit or remove the image and a plus sign to add a link to the first paragraph.

For the second paragraph we again get an edit point for editing the paragraph and two pluses: one for adding an image and one for adding a link. Note that the headline of the second paragraph is directly editable and has no extra edit point - as it has in the first paragraph.


Fig. [inline_editor_result]: The inline editor as configured via (a variant of) the above formatter.

Limitations for inline editing

There are some practical limitations when using inline editing. Edit points (or plus symbols) do not directly show to which content field they belong. Therefore, if you add to much of them, inline editing will not be useful anymore. Moreover, when it comes to choices of different field types, inline editing is still supported, but will be even more complicated to use (we omitted such a use case in the example above).

Technically, inline editing is limited to content that is rendered with a formatter. That in particular means, that it is not supported for contents shown in lists.

You can improve this page

Please contribute your suggestions or comments regarding this topic on our wiki. For support questions, please use the OpenCms mailing list or go for professional support.