[oXygen-user] Learning Xml structure for XSLT authoring

[Adrian Buza]

Hello Wendell and Karl,

Wendell,

If we decide to implement a solution that adds a performance penalty or 
clutters the interface, I can assure you that an option will be 
available to turn that feature off.

Additionally there is another idea to provide content completion in XSLT 
from the associated DTD/schema(if one is available) of the XML file. 
Something similar could be used to implement your proposal("XSLT's 
content completion could aggregate the XML files it sees there"), a 
schema could be created dynamically from the XML files(similar to the 
Tools > Generate/Convert Schema) and that schema could be used for CC.

BTW Wendell, I believe you have accidentally sent the reply only to me 
without CC-ing to the list.


Karl,

We could use an Oxygen PI in the XSL content to associate an XML with 
the stylesheet, similar to the association of a stylesheet with the XML.
e.g.
<?oxygen type="text/xml" href="myfile.xml"?>


I've logged all your suggestions and the improvements of this feature 
will be discussed internally.

Regards,
Adrian

Adrian Buza
oXygen XML Editor and Author Support


Karl Stubsjoen wrote:
> How about this features:
>
> Similar to learning an Xml document, when editing an XSLT an option to
> learn *any* Xml (ie., the Xml document I happen to have open).  A
> dialog box appears on Learn, (better:  Associate) Xml to XSLT and then
> you can pick one.
>
> Thanks!
> Karl..

Wendell Piez wrote:
> Hi Adrian and Karl,
>
> On 2/10/2012 3:51 AM, Adrian Buza wrote:
>> Currently there is no other way to have content completion(CC) in the
>> XSLT with the XML structure other than creating and associating a
>> transformation scenario with the XSLT. If you create and associate a
>> scenario with the XML file that does not affect the CC in the stylesheet
>> mentioned in the scenario.
>> The problem with detecting all scenarios that use an XSLT file is that
>> it's possible to have a scenario that uses an XSLT but it's associated
>> with various XML files. So in that case which XML would you want?
>> One possibility is to have all of them, but that could prove to be
>> excessive in some situations(e.g. very large XML files).
>>
>> I've logged this to our issue tracking tool as a feature request and it
>> will be analyzed and implemented in a future version of Oxygen.
>
> I'm not sure I'd actually like that. I'd certainly like to be able to 
> turn it off, or (maybe better) for a feature like that to be 
> configurable.
>
> On the other hand, I'm not sure I understand the requirement, since 
> it's now perfectly possible to have one scenario bound to a document 
> A, which uses stylesheet X, and another bound to stylesheet X, which 
> uses document A. Then when you edit stylesheet X oXygen knows about 
> document A.
>
> The kernel of this idea, however, namely to have content completion in 
> a stylesheet informed by more than one XML document at a time, is very 
> interesting. Maybe if XSLT transformation scenarios were able to 
> process entire subdirectories (or a logical folder in a project) in 
> batch, then the XSLT's content completion could aggregate the XML 
> files it sees there. (Yes this would have performance implications.)
>
> Cheers,
> Wendell
>
>> Karl Stubsjoen wrote:
>>> I realize that if I setup a transformation scenario between the XSLT
>>> and the Xml then when authoring the XSLT I get intellisense on the Xml
>>> source but only if I create a transformation scenario where the XSLT
>>> is used as the document to build a scenario from.  So what if I create
>>> a transformation scenario and I use my source Xml as the starting
>>> point (typical for me)?  I don't get the intellisense when in the
>>> stylesheet.
>>>
>>> Besides the above, is there another way for the Xslt to learn the 
>>> Xml source?
>>>
>>> Thanks,
>>> Karl..
>>>
>>>
>>>
>>
>> _______________________________________________
>> oXygen-user mailing list
>> oXygen-user at oxygenxml.com
>> http://www.oxygenxml.com/mailman/listinfo/oxygen-user
>>
>>
>