Description


Uniface\'s support of XSLT 1.0 through XALAN only gives access to the limited number of functions available in the XSLT 1.0 specification. In today\'s environment where integration is high on the agenda, more tools are required to aid integration. Please can we have a supported XSLT 2.0 implementation in Uniface?

Use Case


The project worked on is to integrate with other systems. You have one API function you are using to a standard for your system SOA style. Other systems are the same. You can write code to integrate the systems, at some point a transformation is needed. XSL is a general IT skill and is more widely available than Uniface coding skills. In the case I am working in, the integration does not require a code change, but a XSL script change only where XSLT 2.0 would be involved. At this time, expense is increased as code has to change. .Net and Java provide XSLT 2.0 support

Importance


High

Type


Integration

Operating System


Not Applicable

Status


Open

3 thoughts on “XSLT 2.0 support”

  1. Why stop at XSLT2, when XSLT3 is a candidate recommendation! 😉

    Anyway the Saxon implementation is really good. I’ve only used the Java based version both as a Java standalone and through a hand-crafted JNI (java call-out hint, hint!) interface, but I see that there is now a C version in beta that can be deployed as a DLL. So in theory that could be used now.

    But I strongly concur with this wish, the difference between XSLT 1 and 2 is dramatic, and no-one should be using XSLT1 these days.

    1. I’d go for either version 2 or 3.
      We have/had JNI bridge written in C but its quite a pain based on how wide our support is for hardware/software/OS so requires much maintenance.
      Based on wish JAVA CALL-OUT, currently page 18 dated 27 July 2009, having a way for Uniface to hold Java signatures to do a direct call out would not only be amazing, but would also let us write our own XSLT 2 Java function instead as well as other functionality we need 🙂

  2. This is on our radar, but won’t get any attention until we’re further with Uniface 10, and the mobile solutions, which are our focus points. We need to look into a C++ implementation to get away from Java in the Uniface.

Leave a Reply