Jabber impl

Hi Emil,

I have few questions  and suggestions about the
implementation and testing.

1. I'm looking at IcqStatusEnum should I create one for the
Jabber 
implementation ?
    The statuses are almost the same (except OCCUPIED)

2. Lets export some generic Protocol Provider Tests.
    Currently I used your sip tests and the changes so I can
used for 
the jabber implementation were minimal.
    The only tricky thing we must think are how to run this
generic test 
over particular implementation.   

More to come )

damencho

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Another question came up. This time its for the contact
list.
I was looking at the icq implementation.
When subscribing (adding buddy to the list ) and parent is
not specified 
the first persistent group is taken - which is the parent
for the new entry.
But what happens if such group doesn't exist ?

damencho

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Hi Damencho,

I used the first persistent group in icq because the server
would not 
allow you to add contacts outside of a group.

Theoretically, when no parent is specified we should add
contacts inside 
the root group. If Jabber allows that - then this is the way
to go. If 
not - then we could use some default group (as we do in
ICQ). If no 
default group is available then we could create one (e.g.
SIP Communicator).

Does this answer your question?

Emil

Damian Minkov wrote:
> Another question came up. This time its for the contact
list.
> I was looking at the icq implementation.
> When subscribing (adding buddy to the list ) and parent
is not specified 
> the first persistent group is taken - which is the
parent for the new entry.
> But what happens if such group doesn't exist ?
> 
> damencho
> 
>
------------------------------------------------------------
---------
> To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
> For additional commands, e-mail: dev-helpsip-communicator.dev.java.net
> 
> 

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Yes thanks ,
apparently a new group must be created if no group exists .
Jabber (or smack lib itself) does not create a user if you
do not 
specify a parent group.

damencho


Emil Ivov wrote:
> Hi Damencho,
>
> I used the first persistent group in icq because the
server would not 
> allow you to add contacts outside of a group.
>
> Theoretically, when no parent is specified we should
add contacts 
> inside the root group. If Jabber allows that - then
this is the way to 
> go. If not - then we could use some default group (as
we do in ICQ). 
> If no default group is available then we could create
one (e.g. SIP 
> Communicator).
>
> Does this answer your question?
>
> Emil

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Hi again,
And something I forgot - will we implement the possibility a
contact to 
exist in two groups ?


damencho

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Hi again and again,
Other interesting thing is that you can have unfiled
entries, entry that 
doesn't belong to any groups )

damencho

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Hi again,
And something I forgot - will we implement the possibility a
contact to 
exist in two groups ?


damencho

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Hi again and again,
Other interesting thing is that you can have unfiled
entries, entry that 
doesn't belong to any groups )

damencho

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Damencho,

An old mail that I've forgotten to reply to.

Damian Minkov wrote:
> 1. I'm looking at IcqStatusEnum should I create one
for the Jabber 
> implementation ?
>     The statuses are almost the same (except OCCUPIED)

Yes, you could do that.

> 2. Lets export some generic Protocol Provider Tests.
>     Currently I used your sip tests and the changes so
I can used for 
> the jabber implementation were minimal.
>     The only tricky thing we must think are how to run
this generic test 
> over particular implementation.   

Indeed this is going to be tricky. I was thinking of storing
all generic 
tests in the slick.protocol package. Activators of specific
protocol 
packages could instantiating (with necessary parameters)
those that they 
plan to run. We could wait until we have more generic tests
until we do 
that though.

Emil

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net

Damencho,

An old mail that I've forgotten to reply to.

Damian Minkov wrote:
> 1. I'm looking at IcqStatusEnum should I create one
for the Jabber 
> implementation ?
>     The statuses are almost the same (except OCCUPIED)

Yes, you could do that.

> 2. Lets export some generic Protocol Provider Tests.
>     Currently I used your sip tests and the changes so
I can used for 
> the jabber implementation were minimal.
>     The only tricky thing we must think are how to run
this generic test 
> over particular implementation.   

Indeed this is going to be tricky. I was thinking of storing
all generic 
tests in the slick.protocol package. Activators of specific
protocol 
packages could instantiating (with necessary parameters)
those that they 
plan to run. We could wait until we have more generic tests
until we do 
that though.

Emil

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribesip-communicator.dev.java.net
For additional commands, e-mail: dev-helpsip-communicator.dev.java.net