Skip to content

include task

The include task allows to combine several input files into an interal XML structure.

Task attributes

Attribute name Default value Description
root root The root tag of the XML structure.

Task elements

Element name Multiplicity Description
input [1..n] The file name/pattern to include.
output [0..1] The name of the output file. Optional.

Note that you can use Unix-style file name globbing to include several files with one pattern.

Each input element can have the following attributes:

Attribute name Default value Description
tag content The name of the tag under which included markdow content will be put.
format extension of the filename The format of the include file. One of ['kt', 'md'].
encoding utf-8 The encoding of the input file.

Each output element can have the following attributes:

Attribute name Default value Description
encoding utf-8 The encoding of the output file.

Task behaviour

The specified include files will be read, parsed, and combined to one XML structure.

Example

pipeline:

    include: root="top"
        input: tag="sub" "document.md"
        input: "**/*.kt"

will result in an XML structure like

<top>
    <sub>
        <!-- content of docment.md, converted into xhtml -->
    </sub>
    <!-- content of the matching .kt files, converted into xhtml -->
</top>