TalvaStudio Beta 1

[Didier PH Martin <martind@netfolder.com>]

Hello

I was quite silent these last weeks, this is because we where working on
talvastudio: the first web based XML development environment. It is still
not perfect and it still need improvements, however, it is probably the time
that we get some feedback from you fellow XML developers. If you want to
beta test it just signin at:
http://www.talvastudio.com/scripts/asp/login/signin.asp

Note:
The document is very imperfect and a lot is still missing. We are now
writing it. So I'll give you some hints before using the tool.

a) the interface is like a windows explorer. It behaves roughly the same
way. The workspace (i.e the explorer) is dynamically created from a jsp
script that produces an XML document (i.e. a data model) then this document
is transformed into DHTML with an XSLT script. You'll see that with XSLT you
can do quite powerful things since even the javascript code is produced with
the XSLT stylesheet. Thus, the workspace is a DHTML application produced
with XSLT. Anyway, almost everything is XML based.

b) create and edit the files in the workspace. all references like for
instance, the references for the stylesheets can be relative. So everything
is relative to your account folder (exactly like it would be for a usual web
site). Thus if you have a folder named stylesheets under the root folder if
the file is in the root folder you can reference a stylesheet located in the
"stylesheet" folder with the realtive url: stylesheet/test.xsl

c) to adapt the stylesheets to devices you add a new attribute to the
stylesheet processing instructions: the media attribute. Here is an example:
media="screen,html" or media="handheld,wml". Thus, when the server
recognizes the user agent, it matches the stylesheet identified with the
media attribute. The media recognition is automatically performed by the
server. It actually recognizes some hundreds of them but when a new one
performs a request and if it is not recognized, we get its header in a log
file and then, after analysis, we add this new device to the list.

d) You'll meet the pipeline processing metaphore when you select a JSP or an
XML document. For JSP documents, you'll see a JSP editor identified by the
"script" tab, then the "script output" identified by a tab having the same
name, this last context is used to debug your JSP script, if it works OK you
should see the produced XML document. If not, an error message appears to
tell you on which line you got the error. Then you have the "stylesheet"
tab, this context allows you to edit the stylesheets. the stylesheets for
each device is selectable with a drop box. So you can edit all the
stylesheets associated to a document in this context. The system
automatically parsed the processing instructions and knows which stylsheet
is associated to which device. Then the next tab is the "stylesheet output".
In this context the source XML or HTML document produced by the stylesheet
is displayed. finally the "view" tab allows to see the result. We are
working to add more device emulators.

I am sure that you'll have questions or comments. Do not hesitate to email
me your comments or question to martind@netfolder.com

Thanks for you feedback and I hope this tool will be useful to the XML
community.

Didier PH Martin