XML Editor

Provides a special layout when entering in debugging mode to show the source and the stylesheet documents side by side and to show also the results and special debugging views.

Debugging can be done using Saxon 6.5.5 or Xalan 2.5.1 transformation engines

All debugger capabilities are provided (step into, step over, step out, run, run to cursor, run to end, pause, stop).

Breakpoints can be set on both XML or XSL documents used in the transformation.

The current stack containing both stylesheet and source nodes can be seen at any time during debugging. The advantage of this approach is that the source scope on which a stylesheet instruction is being executed (the last red colored node on the stack) is easily visible.

Entering and leaving node events are logged and made available in a view to show the history of the processor execution.

XPath expressions are evaluated automatically during debugging. The debugger will update their values dynamically as the processor changes its source context.

Once a node from an XML entity or an imported/included stylesheet is processed by the transformation engine, will open that resource in order to highlight the current node location.

The list with all the templates used during the transformation and their corresponding number of calls are presented in the Templates view. This allows to spot unused templates and also most frequently used templates as they are ordered based on the hits count.

The debugger lists all variables and parameters visible in the current context together with their scope and actual value.

The output is dynamically presented as it is generated by the transformation process.

The output can be seen both as text/xml and as rendered XHTML.

A mod file was added in docbook/dtd in the distribution that adds XInclude support to the DocBook DTD. There are also document templates (New from templates action) that allow easy creation of a DocBook document with XInclude support. Also a ready-to-use sample is provided in the samples/docbook directory giving you a head start when working with DocBook and XInclude.

The re-use of transformation scenarios increased dramatically by allowing variables in the transformation configuration dialog box. For instance, it is possible to configure a scenario once and use it to convert any DocBook document to a PDF file. Another example is if you have an XML file called test.xml and a stylesheet called test.xsl and you want the output to be test.html (all in the same directory), then you can generically define a transformation scenario using editor variables, specify the XSL (${cfd}/${cfn}.xsl) and the result file (${cfd}/${cfn}.html), and re-use the same scenario to convert sample.xml using the sample.xsl stylesheet into sample.html.

It is possible to define complex scenarios that involve more than a single stylesheet, thus a single action can execute multiple XSLT transformations and eventually also a FOP transformation.

The transformation scenarios dialog box allows to specify to fetch the XSL automatically from the xml-stylesheet processing instruction. If an XML file does not have an associated transformation scenario then a default scenario will be created on the Apply Transformation action configured to use the stylesheet specified by the xml-stylesheet processing instruction.

Presents the elements defined in a target schema (Options/Preferences/Content Completion/XSL) as possible substitutes for xsl:element. Also, these elements will present the possible attributes, and where available, possible attribute values according to the target schema.

It is possible to select the XSLT engine to be used for XSLT validation (Saxon, Xalan or the one plugged in through JAXP). Saxon SA is bundled with Oxygen so Schema Aware XSLT 2.0 is available out of the box.

The support for tagging text was improved, the surround in tag action has a keyboard (configurable) shortcut, CTRL+E by default, and offers a combo box with the allowed elements where an element will be selected automatically based on the typed prefix. This together with the surround in last tag action (CTRL+/) can be productively used to markup unstructured text.

The Content Completion Assistant driven by a Relax NG schema will offer only the elements that can be children of the current element instead of all schema specified elements. In the future the Content Completion Assistant will display only the elements that keep the document valid in that context - as it does for XML Schemas and DTDs.

If a schema or a DTD is not specified for a document, Oxygen will automatically learn its structure and use it to drive the Content Completion Assistant.

After inserting elements through the Content Completion Assistant the cursor is positioned differently depending whether the inserted element has attributes or not. Sometimes it is useful to have the cursor always between tags regardless the element attributes. This is possible by enabling the "cursor position between tags" option.

The Outline view used to select and expand automatically the node containing the document change. With version 4.0 the user can select from tree synchronization options for the Outline view selection: the node at cursor position, the node containing the document change or no update. Also on the contextual menu there are available expand and collapse actions.

When Home key is pressed for the first time on a line the cursor moves at the beginning of the text on that line. When Home is pressed again the cursor moves at the beginning of the line, that is position 0. This behavior is repeated at every Home key pressing.

Along with Open file at cursor, a new Open resource at cursor action is also provided that looks for fully specified Java classes and opens their source files. This is useful, for instance, when editing/browsing configuration files that contain references to Java classes.

The Search/Replace dialog box has a filter that allows the action to be performed only on the selected project files.

The Tree Editor can now be launched as a standalone application. It can also be configured to enter in a read-only mode if the user wants to use it only for browsing.

When working across operating systems (for instance, sharing a Linux drive to Windows machines and editing the same file from both Windows and Linux) it is possible for different line separator characters to appear. It is also possible to ignore the system default line separator and choose from Unix (\n) and Windows (\r\n) line separators in the Preferences dialog box.

To easily view a file in the configured browser a contextual action, Open in browser, was added in the Project view.

When a file with an unknown extension is being opened, a dialog box asking for what editor (XML, XSL, XSD, etc.) should be used appears. It is possible to permanently associate the file extension with the selected editor by just selecting the "associate file type with the editor" option in the same dialog box.

New plugins are available to help text processing: convert text to uppercase, to lowercase, capitalize lines, capitalize the first letter on every selected line, form words to capitalize the first letter of every word and form sentences to capitalize the first letter of the first word in every sentence. All these work on the current editor selection.

In the Project contextual menu an action has been added: "Copy location". This allows copying the URL of the selected file into clipboard.

A tool tip has been added to the Results panel showing the whole message content.

The font used in editors can be configured to enable text anti aliasing and also to make it bold.

Adding directories/files to the current project now supports multiple selections. There are also two contextual actions on the editor tabs that allow adding the current edited file and all the edited files to the current project.

There are some Emacs key bindings that have been added: cut next word - CTRL(META) + D, cut previous word - CTRL + W (only on Mac), cut to end of line - CTRL + K.

The editor can work with files located on HTTP servers that use digest authentication.

This is an internal change that is not necessary visible at first but it speeds up the editor and also makes it more robust. The Content Completion Assistant will work even if part of the document before the cursor position is not well formed.