[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [actf-dev] Validation component
|
Hi Michael,
Thanks so much for your email.
I agree with you that you can implement most of the validation rules
with Javascript but there are some that it will be better to be
implemented in a much richer language. For example, we are interested
in checking HTTP Header information, etc.
As I have already highlighted in my email, it is quite important that
we validate pages against IE.
In our project, we are also working on validating user agents and for
that part we are quite interested in using Javaco.
Thanks.
Regards,
Yeliz.
On 6 Jul 2008, at 23:35, Michael A Squillace wrote:
- It seems like it will be difficult to extend Webelo to write new
validation rules. You have to work with XML and also encode
validation rules in a scripting language, which we think will be
difficult to encode some rules in a scripting language.
MAS: I don't see why it would be anymore difficult to write rules in
JavaScript than in, say, Java. Some would argue that its easier since
scripting is often a quicker process than writing full-blown code
to be
compiled. At any rate, it is only more sophisticated rules that need
involve scripting; for simpler tests, XML is all that is necessary.
Perhaps
you can provide me some examples of the sorts of rules that you'd
like to
encode.
At first, we
though you could just extend the validation rules by writing XML
pages but one has to encode validation rules in a scripting language.
MAS: Again, this is not true. I would need some example sof rules
you think
you'd like to encode and could provide further guidance.
- Webelo only validates pages with Mozilla DOM which we think it is
potentially a limitation because most of the assistive technologies
(such as Jaws) work with Internet Explorer. Therefore it is crucial
to validate pages for Internet Explorer.
MAS: We agree but our team's resources dictated that we work with
Mozilla
because that browser widget was already embedded in Eclipse and
provided
access to the DOM; the IE widget in Eclipse did not. Fukuda-San's
team has
an API for accessing the IE DOM but we have not yet explored that
code.
These are the two main reasons why we have decided to use aDesigner.
Can you please confirm that aDesigner validation component will
continue to be part of Actf?
MAS: I'm sure it will be but Fukuda-San will need to confirm. Also,
keep in
mind that, so far as I know, ADesigner's rules cannot be extended
without
writing your own Java code and contributing to the ADesigner
codebase. If
you're going to write code, you could just as easily implement the
requisite API that I discussed in my presentation. Of course, the
choice is
yours.
