[aprssig] APRS device identifiers (tocalls) in YAML, XML and JSON

[Heikki Hannikainen]

On Thu, Oct 24, 2013 at 12:48 AM, Lynn W. Deffenbaugh (Mr)
<ldeffenb at homeside.to> wrote:
> Great first cut Hessu!  Thanks for the effort, and especially for the
> various supported formats.

I hope it's useful for a few of you. One format is never good for
everyone, so I figured it's best if I pick one to please myself and
then a few others that are likely to be acceptable for the rest.
One-way conversions are easy to do.

> I just can't believe I never noticed that everything ending in D and U are
> allocated, even in the original tocalls.txt:
>
>>        APnnnD  Painter Engineering uSmartDigi D-Gate DSTAR Gateway
>>        APnnnU  Painter Engineering uSmartDigi Digipeater

Not everything, just ones with AP, three digits and then D or U. So
that's a bug in my translation, should be APnnnD, not AP???D. The
Perl/regexp version had AP(\d{3})D. Fixed.

> Is aprsd truly only allocated numeric versions?  Or am I mis-reading the nnn
> vs ??? in the tocall: elements?
>> - tocall: APDnnn

aprsd is not allocated in tocalls.txt, but I think I've only observed
numeric versions in the wild.

> Looking further down the list, how is a * different than a ? or n,
> specifically:
>
>> - tocall: APTT*

* and ? are 'glob' style wildcards:
* for any number of any characters, ? for a single character. n stands
for a digit.

I'd actually prefer regular expressions for these myself, since they
are the Gold Standard for string matching and have a lot of
flexibility: \d{0,3} stands for somewhere between 0 to 3 digits (\d).
Bob's tocalls.txt only uses xxx and nnn, but actual use by some apps
does not match that, some only use one or two alphanumerics even
though the allocation says 'xxx'.

> What are your intended implications and uses of the classes? Particularly
> "app" vs "software"?  There's a lot of folks out there that take APRSIS32 on
> Windows mobile and/or run it on XP tablets.  So which is it?

I'd say the only use for those is to show the string (or a localized
version of it) to the end user, there shouldn't be any technical
implications as such. I suppose your next thing will be an app (for
android at first), distributed through the app stores, and aprsis32 is
a "software" of the traditional type even though it might run on
tablet-shaped WinXP computers.

>  Also is there
> any expectations that "tracker" are deaf?  Or is it acceptable that they
> might do messaging?

There's a messaging flag (messaging:1) for the ones which are
messaging capable (but do not specifically indicate it in their
transmissions, like Kenwoods transmitting Mic-E). So that can be
specified separately. The class shouldn't come with any expectations
like that, capabilities need to be indicated separately.

I'd say 'trackers' are one of those small devices with no user
interface (tinytrak, opentracker, etc).

> On 10/23/2013 5:23 PM, Heikki Hannikainen wrote:
>> https://github.com/hessu/aprs-deviceid

- Hessu