Keypress frames confirmed!

Gary Reuter (garyr@cartman.icaxon.com)
Mon, 10 May 1999 20:58:02 -0400 (EDT)

Hi again!
I was right about the keypad entry... The frames of type 0x0C involve keypad input.
I find this discovery promising: this could lead to functionality that NCDS does
NOT have! And very easy to implement (I think...).
The GUI could be an image of the phone and you just click around to press
the keys.... at least for a "quick-n-dirty" implementation.
(Pavel, I'm also worried about having something that looks identical to a commercial product...
couldn't they try to sue us? :( )

Frame description:

1E 00 0C 0C 00 ln 00 01 00 42 01 kp 01 sq ck-sm

1-3rd oct: put in by FB61_TX_SendMessage()
4th octet: type 0x0C
6th octet: length -- All my tests so far have been with ln = 0x08
7-9th oct: FB61_FRAM_HEADER
10th octet: I think this might be a subtype... more tests needed.
11th octet: not sure yet -- could be length of keypress...
12th octet: kp keypad pressed... see below
13th octet: not sure what this is...

kp:

Hex Keypad
01 1
.. .. (short form for (0x01..0x09) = (1..9)
09 9
0A 0 zero
0B #
0C *
0D Pwr power button
0E Tlk Talk button
0F End End button
10 Vl+ Volume up
11 Vl- Volume down
12 Bksp ** Backspaces phone/msg entry, but not CLEAR button
13 unknown
14 unknown
15 unknown
16 unknown
17 Up up arrow (scroll up)
18 Dn down arrow (scroll down)
19 Menu Menu / Options (button at lower left of screen)
1A Name Names / Exit (button at lower right of screen)

1B onwards: don't know but they do produce a beep and light up the
keypad as if a key had been pressed.

I've also discovered that the 'req2' in my patch isn't necessary when the
timeout is long enough... (just comment it out and turn your phone on and
off again with only req1 and a nice long pause before the SwitchPower
exits).

Sending the "end" key from the main menu locks the keyboard. I haven't
yet tried putting several keypresses together to unlock the keyboard
(which would be necessary for anything else to happen using this
function).

Funny thing to try: comment out the send for req2 and change req1 to send
the a scroll keypress (0x17 or 0x18)... now watch your phone book scroll
by!!!

-Gary

PS: This particular frame type has been quite fun to play with so far, partly because
it doesn't require much programming ability (!!!), and also because NCDS doesn't seem to use
these frames very often.

Has anyone tried all the AT commands (best source by Nokia is the DataCard AT command specs)?
>From my experiments, definately not all of them are implemented in the NCDS softmodem... but
from this discovery, I'm tempted to believe the fbus protocol can do alot more than NCDS alone
can.