xslt

Convert XML files using XML Stylesheets (XSLT) transformations to a specified schema

Command line usage

/*^babelfsh
xslt
   --xml XML --xslt XSLT [--dont-filter-duplicate-codes]
   [--dont-merge-duplicate-codes-by-designation]
   [--dont-rewrite-html-to-markdown]
   [--filter-concepts-without-display-and-designations] [--help]
^babelfsh*/

Command line argument reference

Argument name	Description
--xml XML Required	The relative path to the XML file to be converted
--xslt XSLT Required	The relative path to the XSLT file for converting to the target schema
--dont-filter-duplicate-codes --filter-duplicate-codes	Filter out codes with duplicate codes
--dont-merge-duplicate-codes-by-designation --merge-duplicate-codes-by-designation	Merge duplicate codes (their designations)
--dont-rewrite-html-to-markdown --rewrite-html-to-markdown	Rewrite HTMl in string fields to Markdown, default false
--filter-concepts-without-display-and-designations --dont-filter-concepts	Filter out concepts that have neither display nor designation, default false
--help -h	show this help message and exit

BabelFSH XML Schema

The XML schema for the XSLT transformations' output is defined in an XSD file available in the repo. Here's an annotated snapshot:

<?xml version="1.0" encoding="utf-8"?>
<xs:schema
        elementFormDefault="qualified"
        xmlns:xs="http://www.w3.org/2001/XMLSchema"
        xmlns="urn:mii-termserv:xmlns:babelfsh:codesystem"
        targetNamespace="urn:mii-termserv:xmlns:babelfsh:codesystem" xmlns:xd="http://www.w3.org/2001/XMLSchema">

    <xs:element name="CodeSystem">
        <xs:annotation>
            <xs:documentation>
                This element is the root node of the target schema. All output documents must feature a CodeSystem node.
                No metadata other than the name of the target CodeSystem is required.
            </xs:documentation>
        </xs:annotation>
        <xs:complexType>
            <xs:sequence>
                <xs:element name="concepts" type="concepts"/>
            </xs:sequence>
            <xs:attribute name="name" type="xs:string" use="required"/>
        </xs:complexType>
    </xs:element>

    <xs:element name="code" type="xs:string"/>
    <xs:element name="display" type="xs:string"/>
    <xs:element name="definition" type="xs:string"/>
    <xs:element name="concept" type="concept"/>
    <xs:element name="concepts" type="concepts"/>
    <xs:element name="designation" type="concept_designation"/>
    <xs:element name="property" type="concept_property"/>

    <xs:complexType name="concepts">
        <xs:annotation>
            <xs:documentation>
                The output CodeSystem must define a list of concept nodes
            </xs:documentation>
        </xs:annotation>
        <xs:sequence>
            <xs:element maxOccurs="unbounded" ref="concept"/>
        </xs:sequence>
    </xs:complexType>

    <xs:complexType name="concept" mixed="true">
        <xs:annotation>
            <xs:documentation>
                Every concept must have, at a minimum, have a code, but at least the display should also be present.
            </xs:documentation>
        </xs:annotation>
        <xs:sequence>
            <xs:annotation>
                <xd:documentation>
                    Using xs:ref, the above primitive types are referenced 
                </xd:documentation>
            </xs:annotation>
            <xs:element ref="code"/>
            <xs:element ref="display" minOccurs="0"/>
            <xs:element ref="definition" minOccurs="0"/>
            <xs:element name="designation" type="concept_designation" minOccurs="0" maxOccurs="unbounded"/>
            <xs:element name="property" type="concept_property" minOccurs="0" maxOccurs="unbounded"/>
        </xs:sequence>
    </xs:complexType>

    <xs:complexType name="concept_designation" mixed="true">
        <xs:annotation>
            <xs:documentation>
                This follows the FHIR Designation datatype
            </xs:documentation>
        </xs:annotation>
        <xs:all>
            <xs:element name="language" type="xs:string" minOccurs="0"/>
            <xs:element name="use" type="designation_use" minOccurs="0"/>
            <xs:element name="value" type="xs:string"/>
        </xs:all>
    </xs:complexType>

    <xs:complexType name="designation_use" mixed="true">
        <xs:annotation>
            <xs:documentation>
                This follows the FHIR Coding datatype
            </xs:documentation>
        </xs:annotation>
        <xs:all>
            <xs:element name="system" type="xs:string"
                        minOccurs="0"/>
            <xs:element name="version" type="xs:string"
                        minOccurs="0"/>
            <xs:element name="code" type="xs:string" minOccurs="0"/>
            <xs:element name="display" type="xs:string"
                        minOccurs="0"/>
            <xs:element name="userSelected" type="xs:boolean"
                        minOccurs="0"/>
        </xs:all>
    </xs:complexType>

    <xs:complexType mixed="true" name="concept_property">
        <xs:annotation>
            <xs:documentation>
                The datatypes follow the allowed datatypes for FHIR CodeSystem properties within concepts.
            </xs:documentation>
        </xs:annotation>
        <xs:choice>
            <xs:element name="valueCode" type="xs:string"/>
            <xs:element name="valueCoding">
                <xs:complexType mixed="true">
                    <xs:all>
                        <xs:element name="system" type="xs:string" minOccurs="0"/>
                        <xs:element name="version" type="xs:string" minOccurs="0"/>
                        <xs:element name="code" type="xs:string" minOccurs="0"/>
                        <xs:element name="display" type="xs:string" minOccurs="0"/>
                        <xs:element name="userSelected" type="xs:boolean" minOccurs="0"/>
                    </xs:all>
                </xs:complexType>
            </xs:element>
            <xs:element name="valueString" type="xs:string"/>
            <xs:element name="valueInteger" type="xs:int"/>
            <xs:element name="valueBoolean" type="xs:boolean"/>
            <xs:element name="valueDateTime" type="xs:string"/>
            <xs:element name="valueDecimal" type="xs:decimal"/>
        </xs:choice>
        <xs:attribute name="code" type="xs:string" use="required"/>
    </xs:complexType>
</xs:schema>

Example XSLT file

An example XSLT file is available in the resources for the ORPHAcodes nomenclature. That plugin utilizes the XSLT plugin to then post-process the results of the transformation, so that is a suitable example. Note, however, that the XSLT for ORPHAcodes uses two xsl:param parameters that are provided when the stylesheet is executed, which at this point in time is not implemented for the core XSLT plugin. Nevertheless, the XSLT file should illustrate the output format nicely.