[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: The w3c test objects.





Libby Miller wrote:




(A) I have seen these before. Some minor nits:

Many of the VALARMs have an ATTACH property that is not valid:

ATTACH;VALUE=URI:Ping

Extra CR LF's in them between components.



Sorry, could you clarify? the ics files in that directory are all generated from tools which export iCalendar.


Nothing serious, but several of the components end with:


CR LF CR LF

and not just

CR LF

I do not know of a parser that can not handle it, but it is not per the 2445 ABNF.

Some VEVENTs have no SUBJECT or DESCRIPTION.




Descriptions and summaries are optional, as I understand it - is that incorrect?

Yes incorrect. When they have the METHOD property for example. See the restriction
tables in RFC-2447. For example SUMMARY in a VEVENT PUBLISH is not
optional, but may be empty.


(B) RDF issues I have noticed in the examples:

I noticed that the RDF format drops the 'VALUE' parameter completely.

VALUE (as in VALUE=BINARY, VALUE=DATE) is handled in various ways within
the RDF model.
So in RDF whether something is text or a uri is a crucial distinction
handled by the RDF syntax. Whether something is a DATE or a DATETIME is
handled using different RDF properties.

I think the main thing is that we can roundtrip between the RDF and
iCalendar versions, so we aren't losing information - inevitably there
will be differences in the syntactic representation that are difficult
to explain as we are translating between two different representational
languages.


How do you handle VALUE=x-user-type ?

Example:

iCAL:

x-foo-property;VALUE=x-user-type:1234

RDF:

?

 The RDF-cal has a 'class' tag, that is not the same as the iCAL
'class' tag.

hm, it's not intended to be different. All we have done is create uris
to represent the possible values for class, i.e. public, private,
confidential become

<class rdf:resource='http://www.w3.org/2002/12/cal/ical#private'/>

etc

this just fits in better with the way RDF works - it's much better at
matching URIs than strings.

Okay, perhaps because it looked different from the way the other property values
were translated I may have missed it. Looking closer...


As it appears that the URL specified is specific to iCAL, why is CLASS treated
as a special property type?


 The examples drop the CLASS property nd its PUBLIC, PRIVATE,
 and CONFIDENTIAL values completely.

The RDF-cal does not distinguish between property values and parameters.



this distinction doesn't really matter to RDF :) We don't think that matters as long as we can get back to the iCalendar format as required.

Maybe ... still thinking about this.



I have yet to see an example of how multivalued values are treated.
Is the XML tag replicated or is the value just put in 'as is'?



Which properties were you thinking of? If there were (say) two descriptions then you would just repeat the tag. I don't think we have come across any iCalendar data with repeated tags yet. RDF has an underlying model of object->property->object, so the syntax would depend on whether the repeated tag translated into an RDF object or property.


Multi valued means two or more values for ONE property, not the same property repeated twice. CATAGORIES for example allows one or more values separated by a comma.

This example is one property with two valules:

CATEGORIES:BUSINESS,HUMAN RESOURCES

Thanks again for taking a look at what we've been doing - it's very
useful to get such detailed feedback from the iCalendar community.


And again I think that any WebDAV work should be separate from a specification
on translation of date from iCAL from and to XML.


--

Doug Royer                     |   http://INET-Consulting.com
-------------------------------|-----------------------------
Doug@xxxxxxxxx                 | Office: (208)520-4044
http://Royer.com/People/Doug   |    Fax: (866)594-8574
                              |   Cell: (208)520-4044

We Do Standards - You Need Standards

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature