RE: Some draft code for mobileOK Basic Tests RI

> >
> > 1) What work is there on an XMLized version of CSS and is there some
> > code out there already that does this transform?
> Write XPath Queries into the meta-xml-doc and transform it to an XSLT.

Yes but you need the css as XML before you can do that, I think

> 
> > 2) What work is there on an XMLized version of HTTP headers, or is
> > it as
> > simple as
> >
> > <HTTPResponse result="true" start="20:00:00.31" end="20:00:05.29">
> > 	<status>200 OK</status>
> > 	<header name="Content-Type">text/plain; charset="UTF-8"</header>
> > 	<header name="ETag">ds90ewkldskljewr09kjnerfdmnerf</header>
> > 	<payload>&lt;html&gt;&lt;body&gt;&lt;p&gt;Escaped encoding, what
> > a pain&lt;/p&gt;&lt;/body&gt;&lt;/html&gt;</payload>
> > </HTTPResponse>
> Write XPath ...
> 
> >
> > 3) What should the character encoding of the meta doc (and indeed the
> > result doc) be, given that the input doc could be in any encoding?
> meta/result doc: UTF-8 ;-)
> 
> If all meta informations about the incoming document should be in one
> big XML (like the HTTPResponse), like
> the parsing informations about correct/incorrect encoding, encoding
> type,  XML declaration informations, ... in an XMLized version,
> the work can be done by some XPath queries.
> 
> OK, only to check if I am right:
> We have a incoming, meta and result doc.
> The preprocessor transforms the incoming markup, HTTP headers and
> parsing information in one big XML.
> The validation processor (XSLT) uses the meta doc to check the XML
> and generate a result doc.
> Is this correct?
> 
I think these are the Meta doc contents:
- pre-processor version and name
- original URI requested
- http headers used in request
- details of 3xx redirects and final response
	- time start, time end, result
	- http headers
	- payload of doc (untidy)
	- tidied payload of doc
	- external resources
		- encoding and validity check on images, plus intrinsic sizes
		- CSS as XML ???

Received on Wednesday, 7 February 2007 13:19:40 UTC