Yes.
 
The OPAQUE is included in the WBXML to allow for application specific optimisations, e.g. transporting binary data.
 
There is no "general" definition of how to interpret OPAQUE data, it is ALWAYS application specific.
 
You are how-ever right in saying that from an application point of view it MAY be possible to say that OPAQUE is in CDATA, but often it is not that simple.
 
E.g. for SyncML 1.2 the Meta tag is used to indicate the content stored in the data field
 
<Meta><Format>b64</Format></Meta>
<Data>...b64-data...</Data>
 
Here the Data tag MAY be encoded using the OPAQUE field, provided that the Meta/Format tag is changed accordingly, so if the Data element is coded as OPAQUE (meaning the data becomes base64-decoded), the value of the Format element MUST be changed to "bin" instead of "b64".
 
Such "local" specifications can apply for every application (content-type).
 
Reg. the 3rd paragraph: It was more of a "implementation" specific suggestion to implement logic on the end-application (meaning AFTER the WBXML encoder/decoder) instead of doing business logic directly on the WBXML content (since a simple string can be encoded in many ways).
 
In general, many applications (at least SyncML DM) uses OPAQUE to code strings (where they just as well could've used STR_I). So to tie the OPAQUE to CDATA would not be a portable solution.
 
BR. Rasmus
 
On 8/24/05, Philippe Laporte <[EMAIL PROTECTED]> wrote:
Hi,
   Thanks for your answer. Yes, libwbxml will report CDATA as text.

Yes, an XML file can be converted in many different ways into an WBXML
document, when it comes to inlining strings or referencing the string
the string table, but these are all standardized, and hence deterministic.

The decoder can handle the different encodings because they are valid
with respect to the WBXML specs. What is not deterministic is if the
decoder sees OPAQUE, what's in it.

Or?

I don't quite get the third paragraph. Do you mean the WBXML OPAQUE content?

Really there should be standardized way of determining what is the
OPAQUE...no? One could say, for example, if the WBXML is SyncML, then
the OPAQUE is CDATA, no?

Best Regards,

Philippe Laporte
Software

Gatespace Telematics
Första Långgatan 18
41328 Göteborg
Sweden
Phone: +46 702 04 35 11
Fax:   +46 31 24 16 50
Email: [EMAIL PROTECTED]



Rasmus Lock Larsen wrote:

> I see your point. Actually, the wbxmllib will not decode an WBXML
> document encoded from an XML containing CDATA into the same XML (since
> the CDATA tags are not added when decoding OPAQUE WBXML).
>
> My key-point is, that you CAN'T ensure interoperability when talking
> about WBXML (since there is an almost unlimited number of WBXML
> documents representing the same XML document), so as long as your
> decoder can handle the different combinations of encodings, the
> interoperability should be ensured (in XML that is).
>
> Never write any logic (except in the decoder) that uses the WBXML
> content, since the WBXML content is so ambiguous, you will most likely
> increase the code-complexity and still not have a 100%
> interoperability coverage.
>
> Br. Rasmus
>
> On 8/22/05, *Philippe Laporte*
> < [EMAIL PROTECTED]
> <mailto:[EMAIL PROTECTED]>> wrote:
>
>     Hi,
>         Thanks for your answer.
>
>     My question is about how to ensure interoperability between different
>     WBXML servers and clients when it comes to encoding of CDATA.
>
>     Thanks a lot,
>
>     Philippe Laporte
>     Software
>
>     Gatespace Telematics
>     Första Långgatan 18
>     41328 Göteborg
>     Sweden
>     Phone: +46 702 04 35 11
>     Fax:   +46 31 24 16 50
>     Email: [EMAIL PROTECTED]
>     <mailto:[EMAIL PROTECTED] >
>
>
>
>     Philippe Laporte wrote:
>
>     > Hi,
>     >    Looking at the code in libwbxml it appears that CDATAs are
>     encoded
>     > into opaque for WBXML.
>     >
>     > I couldn't find the section in the WBXML specification document that
>     > specifies this.
>     >
>     > Is it specified at all, or is it in any way standardized to do
>     this?
>     >
>     > Similar question for not encoding in    <![CDATA[    and     ]]>
>     > (although of course it makes a lot of sense).
>     >
>     > So the complementary question is what is the heuristics for knowing
>     > what's in an OPAQUE section?
>     >
>     > Thanks,
>     > Philippe
>     >
>     >
>     > -------------------------------------------------------
>     > SF.Net email is Sponsored by the Better Software Conference & EXPO
>     > September 19-22, 2005 * San Francisco, CA * Development Lifecycle
>     > Practices
>     > Agile & Plan-Driven Development * Managing Projects & Teams *
>     Testing
>     > & QA
>     > Security * Process Improvement & Measurement *
>     http://www.sqe.com/bsce5sf
>     > _______________________________________________
>     > Wbxmllib-devel mailing list
>     > [EMAIL PROTECTED]
>     <mailto: [EMAIL PROTECTED]>
>     > https://lists.sourceforge.net/lists/listinfo/wbxmllib-devel
>
>
>
>     -------------------------------------------------------
>     SF.Net email is Sponsored by the Better Software Conference & EXPO
>     September 19-22, 2005 * San Francisco, CA * Development Lifecycle
>     Practices
>     Agile & Plan-Driven Development * Managing Projects & Teams *
>     Testing & QA
>     Security * Process Improvement & Measurement *
>     http://www.sqe.com/bsce5sf
>     _______________________________________________
>     Wbxmllib-devel mailing list
>     [EMAIL PROTECTED]
>     <mailto:[EMAIL PROTECTED]>
>     https://lists.sourceforge.net/lists/listinfo/wbxmllib-devel
>
>
>
>
> --
> Med venlig hilsen / Best regards
> Rasmus Lock Larsen
> [EMAIL PROTECTED] <mailto: [EMAIL PROTECTED]>




--
Med venlig hilsen / Best regards
Rasmus Lock Larsen
[EMAIL PROTECTED]

Reply via email to