Re: Reproducability of XEP-0115: Entity Capabilities - 5.3 Complex Generation Example

On 04/24/2008 4:04 AM, Yann Leboulanger wrote:
> Peter Saint-Andre wrote:
>> Yann Leboulanger wrote:
>>
>>> I have another question about this example:
>>> how is it possible that the field ip_version which has no type (so
>>> should be considered as text-single according to XEP-0004) can have
>>> several <value> elements?
>> Typically the field 'type' attribute is provided in data forms of type
>> "form" only (so that the software presenting the form can show an
>> appropriate interface) but not in forms of type "submit" or "result"
>> (since the FORM_TYPE needs to specify the field types).
> 
> Indeed I re-read XEP-0004 and it says:
> "For data forms of type "form", each <field/> element SHOULD possess a
> 'type' attribute that defines the data "type" of the field data (if no
> 'type' is specified, the default is "text-single"); fields provided in
> the context of other forms types MAY possess a 'type' attribute as well."
> 
> So in this case how should a client do? It should remember from previous
> forms that a form with FORM_TYPE=jabber:bot, the field features is of
> type 'list-multi' to present the results ?

We assume that the client received a dataform of type "form" and then
submitted it (no need to include the types there because the entity
you're submitting to sent you the form in the first place), and yes I
think the client would need to remember the types it received in the
dataform of type "form" in order to properly parse the result. But we
could tighten up the spec on this score if that is desirable.

Presence of type "available" not allowed?

Hello,

[Standards list: Please Cc me, as I'm not subscribed to this list.]

Excuse me if this was already discussed.

According to RFC 3921, section 2.2.1, it is not allowed to send presence
stanzas with a type="available" attribute. Such a value, however, seems to
be widely used in effective code and in documentation, including on
documentation referenced or hosted at xmpp.org.

I see several possible responses to this fact:

1. Modify existing documentation and implementations to a more strict
behavior that would treat values not mentionned in RFC 3921 as illegal.
2. Add 'available' as a legal type for presence stanzas in future
replacements of RFC 3921.
3. Ignore this difference, and go on tolerating and documenting this value
as a possible one.

Though I'm suspecting response 3 to be chosen in fact, I think we should
avoid html-like behavior regarding respect to standards in code
implementations, and even more in normative documentation.

What do you think?
--

-- 
David Ammouial
http://da.weeno.net/

Re: Presence of type "available" not allowed?

>  1. Modify existing documentation and implementations to a more strict
>  behavior that would treat values not mentionned in RFC 3921 as illegal.

What documentation mentions the type 'available'? I know some (older)
implementations used to use 'available' as a type, and most
implementations just ignore an unknown type and treat it as
'available', which i think is a sensible thing to do. If you are a
purist, you could return an error if you encounter it, but I think
it's much more work to return errors than ignore it on the client
side.

Anyway, if you see the type somewhere, it should be reported as a bug,
and the software/documentation should be fixed.

cheers,
Remko

Re: [Standards] Presence of type "available" not allowed?

On 05/03/2008 2:51 PM, Remko Tronçon wrote:
>>  1. Modify existing documentation and implementations to a more strict
>>  behavior that would treat values not mentionned in RFC 3921 as illegal.
> 
> What documentation mentions the type 'available'? I know some (older)
> implementations used to use 'available' as a type, and most
> implementations just ignore an unknown type and treat it as
> 'available', which i think is a sensible thing to do. If you are a
> purist, you could return an error if you encounter it, but I think
> it's much more work to return errors than ignore it on the client
> side.
> 
> Anyway, if you see the type somewhere, it should be reported as a bug,
> and the software/documentation should be fixed.

I have found two instances of this bug, both in examples from XEP-0200.
They have been corrected:

http://svn.xmpp.org:18080/browse/XMPP/trunk/extensions/xep-0200.xml?r1=1465&r2=1823

Peter

--

-- 
Peter Saint-Andre
https://stpeter.im/

Re: Presence of type "available" not allowed?

I believe the correct way to do this is:

<presence><show></show></presence>

I.e., An empty <show> tag.

Take care,
Sean

On 5/3/08, David Ammouial <da-jdev <at> weeno.net> wrote:
> Hello,
>
> [Standards list: Please Cc me, as I'm not subscribed to this list.]
>
> Excuse me if this was already discussed.
>
> According to RFC 3921, section 2.2.1, it is not allowed to send presence
> stanzas with a type="available" attribute. Such a value, however, seems to
> be widely used in effective code and in documentation, including on
> documentation referenced or hosted at xmpp.org.
>
> I see several possible responses to this fact:
>
> 1. Modify existing documentation and implementations to a more strict
> behavior that would treat values not mentionned in RFC 3921 as illegal.
> 2. Add 'available' as a legal type for presence stanzas in future
> replacements of RFC 3921.
> 3. Ignore this difference, and go on tolerating and documenting this value
> as a possible one.
>

Re: Presence of type "available" not allowed?

Sean Gilbertson pisze:
> I believe the correct way to do this is:
> 
> <presence><show></show></presence>

Isn't <presence/> good enough?

--

-- 
Maciek

Re: Presence of type "available" not allowed?

Remko Tronçon pisze:
>>  1. Modify existing documentation and implementations to a more strict
>>  behavior that would treat values not mentionned in RFC 3921 as illegal.
> 
> most implementations just ignore an unknown type and treat it as
> 'available', which i think is a sensible thing to do.

On the other hand, ignoring undefined type 'invisible' and treating it 
as normal available presence may be considered evil by some people.

--

-- 
Maciek

Re: Presence of type "available" not allowed?

Re: Presence of type "available" not allowed?

On 05/04/2008 11:35 AM, Maciek Niedzielski wrote:
> Remko Tronçon pisze:
>>>  1. Modify existing documentation and implementations to a more strict
>>>  behavior that would treat values not mentionned in RFC 3921 as illegal.
>>
>> most implementations just ignore an unknown type and treat it as
>> 'available', which i think is a sensible thing to do.
> 
> On the other hand, ignoring undefined type 'invisible' and treating it
> as normal available presence may be considered evil by some people.

Which is why we need to define a real protocol for invisibility. Here is
a candidate:

http://www.xmpp.org/extensions/xep-0186.html

Peter

--

-- 
Peter Saint-Andre
https://stpeter.im/

Re: Presence of type "available" not allowed?

Dnia 2008-05-04, nie o godzinie 21:21 -0600, Peter Saint-Andre pisze:
> Which is why we need to define a real protocol for invisibility. Here
> is a candidate:
> 
> http://www.xmpp.org/extensions/xep-0186.html

I prefer: http://www.xmpp.org/extensions/xep-0126.html

--

-- 
  /\_./o__ Tomasz Sterna
 (/^/(_^^' http://www.xiaoka.com/
._.(_.)_   im:smoku <at> xiaoka.com