From: "Jeff" <fyberoptic1979@...>
Jul 30, 2007
Going by spew.txt, changing the data at 0x20000 can let you add new
app en=
tries to the MS. You do so by changing the first digit of this
series of =
bytes:
02 00 02 00 01 00 01 5e 1c 18 01 01 01 20 a7 19
Well this works =
fine for one or two extra apps. But upon trying to
add a third app, I ran=
across a problem. I get three icons, but my
third icon is now the same a=
s my first one (on mine this is the Hello
World app). And running this th=
ird one doesn't work, it simply
returns to the main menu. The first two a=
pps run fine though.
Has anyone deciphered this series of bytes to know wh=
at they're for?
Or is there the possibility that the MS just simply can't=
handle more
than 2 apps?
From: "Jeff" <fyberoptic1979@...>
Jul 30, 2007
I forgot to mention that if you clear out all the bytes except the
first o=
ne where you specify your number of apps, all the app icons
end up the sam=
e (your first app's), and none of them will execute
(aka just returning to=
the main menu). Could be a piece to the
puzzle.
From: "Cyrano Jones" <cyranojones_lalp@...>
Jul 30, 2007
Or is there the possibility that the MS just simply can't handle more
an 2 apps?
I was looking into my notes on the app directory (pg #08),
and=
I found out a bit more since I wrote the instructions
in "spew.asm" file.=
The first 4 bytes in page #08 are a header, where the
first byte is a =
count of the number of apps in dataflash.
After the 4 byte header, there =
are six byte records, one
for each loaded app. I don't know what all the =
bytes
mean, but the second byte is the dataflash page number
where you pu=
t the app, and the 6th byte is an app number.
I think there is a max of 5=
extra apps.
App number prob should be at least #18, but I don't know
fo=
r sure. I don't know what would happen if you used the
number of some oth=
er app (built in apps have numbers, too).
I think the highest the app nu=
mber is allowed to be is #1D,
or maybe #1E. (there is an internal app tab=
le with 30 rows,
but I don't remember right now if the first is app 00 or 0=
1.)
Going by the data you posted, the first dataflash app is
#18, so I am =
gonna guess that the last internal app is #17.
AFAICT, just about every dif=
ferent screen type is a
different "app", even if the screens seem to be pa=
rt of
what you would normally call the same "app".
0308:4000 nn <-- nu=
m of recs
0308:4001 ??
0308:4002 ?? <--?????
0308:4003 ??
0308:4004=
xx 00 xx xx xx 18 <-- data for 1st loaded app
0308:400A xx 01 xx xx xx=
19 <-- data for 2nd loaded app
0308:4010 xx 02 xx xx xx 1a <-- data fo=
r 3rd loaded app
0308:4006 xx 03 xx xx xx 1b <-- data for 4th loaded app=
0308:4000 xx 04 xx xx xx 1c <-- data for 5th loaded app
=
^^ ^^
|| ||
page who
=
My notes say 0308:4002 is "hex editor state", but I'm
not sure that's rig=
ht. Just leave it whatever it is.
I filled in typical values above for th=
e page & app
numbers, but I don't see any reason they need to be
consecut=
ive. The xx's are not "dont care", I just
don't know what they are for. =
Seems to work if
you just leave them 00.
From: "Jeff" <fyberoptic1979@...>
Jul 31, 2007
Jones" <cyranojones_lalp@...> =
wrote:
t byte is a count of the number of apps in dataflash.
header, there are six byte records, one
ow what all the bytes
mber
I ha=
d a thought earlier. Since obviously the data for the tv
listings and wha=
tnot had to be updated when you dialed in for your
mail, do you suppose pa=
rt of those six byte records could specify a.)
whether (or where) the MS s=
hould execute code while the connection is
open for the app to fetch its o=
wn data, and/or b.) for the MS to then
parse the data it received after th=
e connection is complete? I don't
have any of the actual Yahoo apps to po=
ke around in, but I figure the
data has to be parsed or at least stored so=
mewhere for these apps to
find.
r the page & app
consecutive. The xx's are not "dont care", I just
are for. Seems to work if
This worked like a=
charm, by the way. Thanks!
From: "Cyrano Jones" <cyranojones_lalp@...>
Aug 1, 2007
had a thought earlier. Since obviously the data for the tv
d whatnot had to be updated when you dialed in for your
pose part of those six byte records could specify a.)
the MS should execute code while the connection is
fetch its own data, and/or b.) for the MS to then
ived after the connection is complete?
I suppose anything is possible. =
My hunch is that the
mailstation passively receives the data, rather than
=
fetching it. I would think, since it is binary data,
that it would be at l=
east base64 or uuencoded.
o poke around in, but I figure the
red somewhere for these apps to
If by data you mean the TV listin=
gs, or weather, it is
stored right in the app itself, as one big graphic im=
age.
I don't know if they reload the whole 16k page each time
they update =
data, or just the image data (what I named
icon2 in helloapp, maybe not the=
best name).
This worked like a charm, by the way. Thanks!
Great! Is the foldertab d=
isplaying right???