future new features

[Jacques Durand]

Hello users: here is a forum for posting suggestions for new features,
and trying to prioritize them.
I will post soon a few starters that are on the TO-DO list of the
implementer. Feel free to post yours. Then we can discuss, refine and
prioritize.

Jacques D.

[Stephen Green]

The big things which come to mind which might
or might not be directions of growth for TAMEL
and TAMELizer are:

1. any implementation of W3C EARL in reports

2. any further implementation of TAML TA Set
such as nested TA Sets

3. any implementation of TAML TA References

4. any potential for supporting some kind of profile
of Schematron (say with an internal transformation
- perhaps to import from and export to Schematron
schema)

5. extensions e.g. to add java classes to, for
instance, test things like file size which can't be
expressed in XPath

6. File size could be an issue since TAMELizer is
XSLT-based and has problems with large XML files
so handling of large XML files would be another one
to consider.

7. ability to include TA predicates/prerequisites for
performing W3C XML Schema validation (especially
important as a potential prerequisite to most TAs)

Best regards
---
Stephen D Green

2009/11/5 Jacques Durand <jda...@gmail.com>:

[Stephen Green]

8+9. How to handle TAs which, although outside
of the scope of TAMELizer and TAMEL (or is it
TAML-X?), might, if written using prose predicates
or prerequisites, be necessary as prerequisites
to executable TAs. E.g. these could be turned
into questionaires whose results/answers get
passed in to the execution as parameters. This
is my favourite because it seems to me to be a
business case for TAMELizer in that many kinds
of TA can be bundled together (not just for XPath
tests) and those which are XPath executable can
be combined with all kinds of others using the
same TAML markup. So maybe this is an item of
potential improvement for the profile (TAML-X) too
- that it be able to include prose TAs along with
XPath ones in such a way that prose ones can be
handled as prerequisites to XPath ones. Maybe
also the other way around - that XPath TA execution
reports also be potentially output as a set of
parameters to a questionaire report which guides
any further test requirements for prose TAs which
have XPath TAs as their prerequisites.

Best regards
---
Stephen D Green

2009/11/6 Stephen Green <stephen...@gmail.com>:

[Jacques Durand]

inline <JD>

On Nov 5, 11:28 pm, Stephen Green <stephengreen...@gmail.com> wrote:
> The big things which come to mind which might
> or might not be directions of growth for TAMEL
> and TAMELizer are:
>
> 1. any implementation of W3C EARL in reports

<JD> Yes - especially if we submit TAML-X to W3C... We still have to
define what an "EARL binding" would look like for TAML-X reports,
which should be done I think in TAMEL-X document itself e.g. in an
appendix or as a binding. So I think there are a few steps to achieve
before we can look at an implementation in tamelizer.

>
> 2. any further implementation of TAML TA Set
>  such as nested TA Sets

<JD> before nested sets, a first step could be an awareness of
"simple" TA sets by the tamelizer engine for:
- resolving the "shared" items declared in the set header, to
determine the actual content of each TA.
- maybe for enabling/disabling entire sets of TAs by an @enable
attribute at set level? (instead of TA level).
Then a second step would be the ability to refer to a full TA set by
the set name, in a Prerequisite (e.g. a TA set that is associated with
a conformance profile that is to be expected as a prerequisite), or in
a Predicate (meaning by a "summary" TA that will assess the
conformance of a target - e.g. a document - to an entire set of TAs).

> 3. any implementation of TAML TA References
>

<JD> Need to evaluate the importance level of this by a use case?
One requirement that may or may not be related along this line would
be the ability for the engine to handle TAs scattered over several
documents.

> 4. any potential for supporting some kind of profile
>  of Schematron (say with an internal transformation
>  - perhaps to import from and export to Schematron
>  schema)

<JD> Indeed - that should not be too hard to import. Exporting to
Schematron would only work if no prerequisites ?

>
> 5. extensions e.g. to add java classes to, for
>    instance, test things like file size which can't be
>    expressed in XPath

<JD> some externally-defined XPath functions I guess.

>
> 6. File size could be an issue since TAMELizer is
> XSLT-based and has problems with large XML files
> so handling of large XML files would be another one
> to consider.
>

<JD> no idea how to handle this... There must be some techniques that
advanced XSLT processors will use, that we'll automatically benefit
from.

> 7. ability to include TA predicates/prerequisites for
> performing W3C XML Schema validation (especially
> important as a potential prerequisite to most TAs)

<JD> yes, certainly a high priority feature. In fact some "instance-
of" statement or the like, is already supported in schema-aware
versions of XSLT processors (e.g. Saxon-SA). I think Tom Rutt did test
that for WS-I test suites. There may be a few challenges to overcome,
as it seems that some of these processors need the schemas to be used
to be "declared" somehow. The code generator needs to add that.

>
> Best regards
> ---
> Stephen D Green
>

> 2009/11/5 Jacques Durand <jdan...@gmail.com>:

>
>
>
>
>
> > Hello users: here is a forum for posting suggestions for new features,
> > and trying to prioritize them.
> > I will post soon a few starters that are on the TO-DO list of the
> > implementer. Feel free to post yours. Then we can discuss, refine and
> > prioritize.
>

> > Jacques D.- Hide quoted text -
>
> - Show quoted text -