|
Re: PDF output (JET vs XSLT) [message #38260 is a reply to message #38227] |
Fri, 28 December 2007 15:10 |
Paul Elder Messages: 849 Registered: July 2009 |
Senior Member |
|
|
Dave:
On JET vs XSLT:
* XSLT produces one file per invocation. JET can produce multiple files per
invocation. An XSLT stylesheet it the equivalent of a single JET template.
If you are producing more than one file from a given model, JET templates
(or any other templating language, for that matter) can be less complex.
* JET 'transformations' let you calculate values (and store them as
'temporary' attributes on the input model). This can greatly simplify
individual JET templates, which can concentrate on value substitution and
flow control. In XSLT, all calculations are directly in the stylesheet,
which can make it very hard to 'see' the structure of the generated content.
* Finally, I have trouble relating to the structure of an XLST sytlesheet.
For my taste, it is too oriented to the structure of the input at the
expense of the understanding of the output. IHMO, a good template shows the
structure of the output, with a "little bit" of markup to represent
substitutions from the model.
On generating PDF:
* I don't know of anyone generating PDF from JET. JET is primarily a tool
for generating text (which PDF is not).
* A workable solution would be to write JET templates that generate text
(HTML, RTF, postscript, ..., whatever), and then generate PDF from that
either as a separate transformation, or via a custom JET tag.
On using JET with "Structured English" and SBVR:
* I'm not familiar with this. Out of the box, JET can read/travers XML
documents, and anything that has a registered EMF resource factory. A quick
look at the SBVR spec indicates that MOF/XMI export is possible - at a
minimum, this should be usable by JET. Also, depending on how the structured
english is persisted (its unclear to me after 10 minutes with the spec), JET
may be able to read that directly.
Feel free to ask more questions.
Paul
"Dave Carlson" <dcarlson@xmlmodeling.com> wrote in message
news:fl16qe$ap1$1@build.eclipse.org...
> Hi,
> I need to make a choice of JET vs XSLT, or some other M2T technology, for
> creating PDF and XHTML output from an EMF model. There are advantages to
> using the FO capabilities of XSLT for producing PDF output. But
> advantages of JET for traversing EMF model structures. Has anyone used
> JET for PDF output? More specifically, I am looking into creating
> Structured English output from SBVR vocabularies and business rules.
>
> The current release of IBM's RSx modeling tools use XSLT for creating HTML
> and PDF reports, but I've found those stylesheets to be a pain to
> customize.
>
> Thanks for any input,
> Dave Carlson
>
>
|
|
|
Powered by
FUDForum. Page generated in 0.03086 seconds