Help with xAP_Item in Library Module

[Jim Duda]

I'm finally getting around with playing with xAP, and having fun!

I'm trying to get a xAP_Item to work inside a library module.

Can anyone help me understand why this doesn't work?

The following code is in a library module.  I get the xAP_data startup
messages in the log file.  I also get the check_for_xAP_data check:
messages.  However, $check is always empty, so, I'm not getting anything
from the xAP_ref.

If I use similiar code in a normal code module, all works fine.

I've been struggling with this for a few days.  Stumped.

use strict;

package TEST;
use xAP_Items;

my $xAP_ref = undef;

sub startup {
   &::MainLoop_pre_add_hook(  \&TEST::check_for_xAP_data, 1 );
   if (not defined $xAP_ref) {
     &main::print_log ("TEST::check_for_xAP_data startup");
     $xAP_ref = new xAP_Item('xap-hbeat.alive');
     my $friendly_name = "xAP_test";
     &main::store_object_data($xAP_ref, 'xAP_Item', $friendly_name,
$friendly_name);
   }
}

sub check_for_xAP_data {
    # First, check the xAP channel for messages
#    my $check = $xAP_ref->state_now( );
#    &main::print_log ("TEST::check_for_xAP_data check: $check");
    if ($xAP_ref->state_now) {
      # dump some debugging information
      foreach my $key (keys %{$xAP_ref->{'xap-hbeat'}}) {
        &main::print_log ("TEST::xAP_ref:: receive key: $key data:
$xAP_ref->{'xap-hbeat'}{$key}") if $main::Debug{xap_echo};
      }
    }
}

1;



This code module works fine, I get the hbeat debug messages in the loop.

use strict;

use TEST;
use xAP_Items;

# noloop=start
&TEST::startup( );
# noloop=stop

$xAP_ref = new xAP_Item('xap-hbeat.alive');

#my $check = $xAP_ref->state_now( );
#print_log ("test.pl::check_for_xAP_data check: $check");
if ($xAP_ref->state_now) {
   # dump some debugging information
   foreach my $key (keys %{$xAP_ref->{'xap-hbeat'}}) {
     print_log ("test.pl::xAP_ref:: receive key: $key data:
$xAP_ref->{'xap-hbeat'}{$key}");
   }
}


07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: hop data: 1
07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: uid data: ffe90000
07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: v data: 12
07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: interval data: 60
07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: port data: 49152
07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: pid data: 6816
07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: class data:
xap-hbeat.alive

Thanks,

Jim



-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
________________________________________________________
To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365

[Gregg Liming]

Jim Duda wrote:
> I'm finally getting around with playing with xAP, and having fun!
>
> I'm trying to get a xAP_Item to work inside a library module.
>
> Can anyone help me understand why this doesn't work?

Well, first off, the xAP_Items lib also uses a pre_add_hook to check for
pertinent UDP traffic and then set_now on pertinent objects.  You're not
guaranteed order of hook modules; so, that alone could explain it.

... not to be dismissive; but, I don't understand why you would want to
use such a coding pattern for this anyway.  There's not a single
xAP-enabled module that does so.  The only reason to ever use a
pre_add_hook is for "conduits" (i.e., code that attaches to specific
sources of data such as a socket or serial connection) to acquire and
parse (usually, semi-raw) data.  The xAP_Item lib already does that.
So, trying to intercept or do something different is pretty pointless.
Perhaps, you could shed light on what you're really trying to do?

> The following code is in a library module.  I get the xAP_data startup
> messages in the log file.  I also get the check_for_xAP_data check:
> messages.  However, $check is always empty, so, I'm not getting anything
> from the xAP_ref.
>
> If I use similiar code in a normal code module, all works fine.

Yes--because it is much more correct.

> I've been struggling with this for a few days.  Stumped.
>
> use strict;
>
> package TEST;
> use xAP_Items;
>
> my $xAP_ref = undef;
>
> sub startup {
>    &::MainLoop_pre_add_hook(  \&TEST::check_for_xAP_data, 1 );
>    if (not defined $xAP_ref) {
>      &main::print_log ("TEST::check_for_xAP_data startup");
>      $xAP_ref = new xAP_Item('xap-hbeat.alive');
>      my $friendly_name = "xAP_test";
>      &main::store_object_data($xAP_ref, 'xAP_Item', $friendly_name,
> $friendly_name);
>    }
> }
>
> sub check_for_xAP_data {
>     # First, check the xAP channel for messages
> #    my $check = $xAP_ref->state_now( );
> #    &main::print_log ("TEST::check_for_xAP_data check: $check");
>     if ($xAP_ref->state_now) {
>       # dump some debugging information
>       foreach my $key (keys %{$xAP_ref->{'xap-hbeat'}}) {
>         &main::print_log ("TEST::xAP_ref:: receive key: $key data:
> $xAP_ref->{'xap-hbeat'}{$key}") if $main::Debug{xap_echo};
>       }
>     }
> }
>
> 1;
>
>
>
> This code module works fine, I get the hbeat debug messages in the loop.
>
> use strict;
>
> use TEST;
> use xAP_Items;
>
> # noloop=start
> &TEST::startup( );
> # noloop=stop
>
> $xAP_ref = new xAP_Item('xap-hbeat.alive');
>
> #my $check = $xAP_ref->state_now( );
> #print_log ("test.pl::check_for_xAP_data check: $check");
> if ($xAP_ref->state_now) {
>    # dump some debugging information
>    foreach my $key (keys %{$xAP_ref->{'xap-hbeat'}}) {
>      print_log ("test.pl::xAP_ref:: receive key: $key data:
> $xAP_ref->{'xap-hbeat'}{$key}");
>    }
> }
>
>
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: hop data: 1
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: uid data: ffe90000
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: v data: 12
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: interval data: 60
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: port data: 49152
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: pid data: 6816
> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: class data:
> xap-hbeat.alive
>
> Thanks,
>
> Jim
>
>
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> ________________________________________________________
> To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365
>

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
________________________________________________________
To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365

[Jim Duda]

Gregg,

 > Well, first off, the xAP_Items lib also uses a pre_add_hook to check for
 > pertinent UDP traffic and then set_now on pertinent objects.  You're not
 > guaranteed order of hook modules; so, that alone could explain it.

Yes, I had come to that realization, however, I had then tried
post_add_hook, and had the same behavior, which confused me.  I
certainly do accept this explanation, it makes purpose sense.

 > ... not to be dismissive; but, I don't understand why you would want to
 > use such a coding pattern for this anyway.  There's not a single
 > xAP-enabled module that does so.  The only reason to ever use a
The only reason to ever use a
 > pre_add_hook is for "conduits" (i.e., code that attaches to specific
 > sources of data such as a socket or serial connection) to acquire and
 > parse (usually, semi-raw) data.  The xAP_Item lib already does that.
 > So, trying to intercept or do something different is pretty pointless.
 > Perhaps, you could shed light on what you're really trying to do?

For a long time now, I've been wanting to play with replacing the
proxy stuff with xAP infrastructure.  I do a lot with proxies as I
have multiple MH boxes in my house on a local LAN.

I have three code modules, xAP_speak, xAP_play, and xAP_w800
which perform what the proxy function does for those three applications.
  I've used this simple project as a way to learn about xAP.

I'm now tinkering with if there is a way to perform this stuff
down in the lower library modules, to make it more embedded and
transparent.  I'm trying to understand what is the best use case.

I have been playing around with X10_W800.pm.  I have a modified version
which sends W800 messages in a xAP schema.  I was struggling to get
the "receive" side to work. I couldn't find a way to pull the messages
from the xAP item inside the code module.  I put together the TEST.pm
as a smaller test case.  I couldn't figure out why it wouldn't work,
hence the posting.

I'm coming up to speed on xAP.  I know there was always some grand
plan about using xAP for proxy support, in lieu of the socket
method used now.  I realize there is some small start to be found
in xAP_send.pl (that's where I cloned the add_hook from for X10_W800).

I'd like to help development something with xAP.  Proxies are something
I use and can test.

Suggestions?

Jim



Gregg Liming wrote:

> Jim Duda wrote:
>> I'm finally getting around with playing with xAP, and having fun!
>>
>> I'm trying to get a xAP_Item to work inside a library module.
>>
>> Can anyone help me understand why this doesn't work?
>
> Well, first off, the xAP_Items lib also uses a pre_add_hook to check for
> pertinent UDP traffic and then set_now on pertinent objects.  You're not
> guaranteed order of hook modules; so, that alone could explain it.
>
> ... not to be dismissive; but, I don't understand why you would want to
> use such a coding pattern for this anyway.  There's not a single
> xAP-enabled module that does so.  The only reason to ever use a
> pre_add_hook is for "conduits" (i.e., code that attaches to specific
> sources of data such as a socket or serial connection) to acquire and
> parse (usually, semi-raw) data.  The xAP_Item lib already does that.
> So, trying to intercept or do something different is pretty pointless.
> Perhaps, you could shed light on what you're really trying to do?
>
>> The following code is in a library module.  I get the xAP_data startup
>> messages in the log file.  I also get the check_for_xAP_data check:
>> messages.  However, $check is always empty, so, I'm not getting anything
>> from the xAP_ref.
>>
>> If I use similiar code in a normal code module, all works fine.
>
> Yes--because it is much more correct.
>
>> I've been struggling with this for a few days.  Stumped.
>>
>> use strict;
>>
>> package TEST;
>> use xAP_Items;
>>
>> my $xAP_ref = undef;
>>
>> sub startup {
>>    &::MainLoop_pre_add_hook(  \&TEST::check_for_xAP_data, 1 );
>>    if (not defined $xAP_ref) {
>>      &main::print_log ("TEST::check_for_xAP_data startup");
>>      $xAP_ref = new xAP_Item('xap-hbeat.alive');
>>      my $friendly_name = "xAP_test";
>>      &main::store_object_data($xAP_ref, 'xAP_Item', $friendly_name,
>> $friendly_name);
>>    }
>> }
>>
>> sub check_for_xAP_data {
>>     # First, check the xAP channel for messages
>> #    my $check = $xAP_ref->state_now( );
>> #    &main::print_log ("TEST::check_for_xAP_data check: $check");
>>     if ($xAP_ref->state_now) {
>>       # dump some debugging information
>>       foreach my $key (keys %{$xAP_ref->{'xap-hbeat'}}) {
>>         &main::print_log ("TEST::xAP_ref:: receive key: $key data:
>> $xAP_ref->{'xap-hbeat'}{$key}") if $main::Debug{xap_echo};
>>       }
>>     }
>> }
>>
>> 1;
>>
>>
>>
>> This code module works fine, I get the hbeat debug messages in the loop.
>>
>> use strict;
>>
>> use TEST;
>> use xAP_Items;
>>
>> # noloop=start
>> &TEST::startup( );
>> # noloop=stop
>>
>> $xAP_ref = new xAP_Item('xap-hbeat.alive');
>>
>> #my $check = $xAP_ref->state_now( );
>> #print_log ("test.pl::check_for_xAP_data check: $check");
>> if ($xAP_ref->state_now) {
>>    # dump some debugging information
>>    foreach my $key (keys %{$xAP_ref->{'xap-hbeat'}}) {
>>      print_log ("test.pl::xAP_ref:: receive key: $key data:
>> $xAP_ref->{'xap-hbeat'}{$key}");
>>    }
>> }
>>
>>
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: hop data: 1
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: uid data: ffe90000
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: v data: 12
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: interval data: 60
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: port data: 49152
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: pid data: 6816
>> 07/11/08 09:39:02 PM test.pl::xAP_ref:: receive key: class data:
>> xap-hbeat.alive
>>
>> Thanks,
>>
>> Jim
>>
>>
>>
>> -------------------------------------------------------------------------
>> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
>> Studies have shown that voting for your favorite open source project,
>> along with a healthy diet, reduces your potential for chronic lameness
>> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
>> ________________________________________________________
>> To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365
>>
>
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> ________________________________________________________
> To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365
>
>

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
________________________________________________________
To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365

[Gregg Liming]

Jim Duda wrote:

> For a long time now, I've been wanting to play with replacing the
> proxy stuff with xAP infrastructure.  I do a lot with proxies as I
> have multiple MH boxes in my house on a local LAN.

I assume that you realize/know this; but, just in case... the proxy
solution is inherently client-server whereas xAP (and xPL for that
matter) is peer-centric.

> I have three code modules, xAP_speak, xAP_play, and xAP_w800
> which perform what the proxy function does for those three applications.
>   I've used this simple project as a way to learn about xAP.

I'm guessing/hoping that you've adopted existing schemas where it
applies.  Any data reported via a w800 could be mapped into a BSC
message and speaking has a schema as well.  You would want to map each
device reportable by a w800 as a distinct endpoint rather than pass it's
ID (e.g., A1, K2) as a parameter in the message payload.

> I'm now tinkering with if there is a way to perform this stuff
> down in the lower library modules, to make it more embedded and
> transparent.

While technically possible, that's definitely not in keeping w/ the
spirit/intent of either xAP or xPL.  In particular, at least half of the
issue is to adopt standard schemas that translate to rather meaningful
events.  Although BSC tries to do this to a large degree over a lot of
devices, there is really no similarity between something like speaking
and digital I/O.  Since an appropriate use of schema is required for
each device type, it is unlikely that any generic schema be appropriate
for all and therefore used/applied transparently.

In addition, xAP's current spec has a UID limitation of 256 per
"gateway". That implies that you then have to create "virtual devices"
to handle situations where that number could be greater.  IMO, this is a
huge pita.  Fortunately, a forthcoming spec will resolve this
limitation.  If this is a show-stopper, then use xPL as it is free of
the UID sillyness.

> I'm trying to understand what is the best use case.

IMO, distinct gateways and/or device support should be added as needed
to accomodate varying device types.  BSC already exists and covers a
huge set of possibilities.  Likewise, so does speak, display and a
number of others.  To be honest, I don't know of a widely used schema
that does not already have built-in support.

> I have been playing around with X10_W800.pm.  I have a modified version
> which sends W800 messages in a xAP schema.

Not to "beat a dead horse"; but, this is hopefully a standard schema
vice one that you've invented.

> I was struggling to get
> the "receive" side to work. I couldn't find a way to pull the messages
> from the xAP item inside the code module.  I put together the TEST.pm
> as a smaller test case.  I couldn't figure out why it wouldn't work,
> hence the posting.

There are lots of examples in mh's lib dir: BSC.pm, ZoneMinder_xAP,
OneWire_xAP all implement item libs that can send and receive xAP
messages.  The design pattern is to create a private xAP item in some
item's constructor and tie it to the item.  Then, the set method is
overwritten and tested against setby.  If setby is the tied xAP item,
then you grab data from the xAP item. Otherwise, you assume its a direct
call.

> I'm coming up to speed on xAP.  I know there was always some grand
> plan about using xAP for proxy support, in lieu of the socket
> method used now.  I realize there is some small start to be found
> in xAP_send.pl (that's where I cloned the add_hook from for X10_W800).

I don't think that there was ever a concept of simply replacing the
proxy conduit w/ a xAP conduit.  That just doesn't make sense and is an
abuse of xAP (or xPL for that matter).

> I'd like to help development something with xAP.  Proxies are something
> I use and can test.
>
> Suggestions?

Have your w800 support handle BSC fully.  Then you don't have to worry
about anything else as there is built-in support for BSC.

Gregg

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
________________________________________________________
To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365

[Jim Duda]

Gregg,

Thanks for the detailed discussion.  I obviously have much to learn
here as I don't know much about the proper usage of xAP schemas,
BSC, and other items you've brought to my attention here.  I have
a good deal of information to digest.

I assume a detailed description of the standard shemas and
specifically "BSC" can be found the xap automation web site?
I have yet to read all that material in detail.  I think I need
to do that first.

I realize that the standard proxy stuff is more client-server
based.  I had extended the proxy stuff to be more peer centric
(I believe), when I created proxy_client_server.pl.  I realize
the name is rather awkward, but my intention and how I use it
is really more peer-centric.  Each MH node on my LAN uses
proxy_client_server to offer up messages for a device it has
locally attached and listens for messages from other nodes to take
advantage of resources the other peers have available locally
attached.

My naive understanding of xAP is that it is intended to be
used in a similar way, each node broadcasts information for
resources it has access to and listens for information other
peers have access to.

I will go off and understand more of xAP and get back on
the list with more knowledge.

Thanks,

Jim

Gregg Liming wrote:

> Jim Duda wrote:
>
>> For a long time now, I've been wanting to play with replacing the
>> proxy stuff with xAP infrastructure.  I do a lot with proxies as I
>> have multiple MH boxes in my house on a local LAN.
>
> I assume that you realize/know this; but, just in case... the proxy
> solution is inherently client-server whereas xAP (and xPL for that
> matter) is peer-centric.
>
>> I have three code modules, xAP_speak, xAP_play, and xAP_w800
>> which perform what the proxy function does for those three applications.
>>   I've used this simple project as a way to learn about xAP.
>
> I'm guessing/hoping that you've adopted existing schemas where it
> applies.  Any data reported via a w800 could be mapped into a BSC
> message and speaking has a schema as well.  You would want to map each
> device reportable by a w800 as a distinct endpoint rather than pass it's
> ID (e.g., A1, K2) as a parameter in the message payload.
>
>> I'm now tinkering with if there is a way to perform this stuff
>> down in the lower library modules, to make it more embedded and
>> transparent.
>
> While technically possible, that's definitely not in keeping w/ the
> spirit/intent of either xAP or xPL.  In particular, at least half of the
> issue is to adopt standard schemas that translate to rather meaningful
> events.  Although BSC tries to do this to a large degree over a lot of
> devices, there is really no similarity between something like speaking
> and digital I/O.  Since an appropriate use of schema is required for
> each device type, it is unlikely that any generic schema be appropriate
> for all and therefore used/applied transparently.
>
> In addition, xAP's current spec has a UID limitation of 256 per
> "gateway". That implies that you then have to create "virtual devices"
> to handle situations where that number could be greater.  IMO, this is a
> huge pita.  Fortunately, a forthcoming spec will resolve this
> limitation.  If this is a show-stopper, then use xPL as it is free of
> the UID sillyness.
>
>> I'm trying to understand what is the best use case.
>
> IMO, distinct gateways and/or device support should be added as needed
> to accomodate varying device types.  BSC already exists and covers a
> huge set of possibilities.  Likewise, so does speak, display and a
> number of others.  To be honest, I don't know of a widely used schema
> that does not already have built-in support.
>
>> I have been playing around with X10_W800.pm.  I have a modified version
>> which sends W800 messages in a xAP schema.
>
> Not to "beat a dead horse"; but, this is hopefully a standard schema
> vice one that you've invented.
>
>> I was struggling to get
>> the "receive" side to work. I couldn't find a way to pull the messages
>> from the xAP item inside the code module.  I put together the TEST.pm
>> as a smaller test case.  I couldn't figure out why it wouldn't work,
>> hence the posting.
>
> There are lots of examples in mh's lib dir: BSC.pm, ZoneMinder_xAP,
> OneWire_xAP all implement item libs that can send and receive xAP
> messages.  The design pattern is to create a private xAP item in some
> item's constructor and tie it to the item.  Then, the set method is
> overwritten and tested against setby.  If setby is the tied xAP item,
> then you grab data from the xAP item. Otherwise, you assume its a direct
> call.
>
>> I'm coming up to speed on xAP.  I know there was always some grand
>> plan about using xAP for proxy support, in lieu of the socket
>> method used now.  I realize there is some small start to be found
>> in xAP_send.pl (that's where I cloned the add_hook from for X10_W800).
>
> I don't think that there was ever a concept of simply replacing the
> proxy conduit w/ a xAP conduit.  That just doesn't make sense and is an
> abuse of xAP (or xPL for that matter).
>
>> I'd like to help development something with xAP.  Proxies are something
>> I use and can test.
>>
>> Suggestions?
>
> Have your w800 support handle BSC fully.  Then you don't have to worry
> about anything else as there is built-in support for BSC.
>
> Gregg
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> ________________________________________________________
> To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365
>
>

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
________________________________________________________
To unsubscribe from this list, go to: http://sourceforge.net/mail/?group_id=1365