Title: Message
What
is the code around line 12041, say lines 12035-12050. It would be good to see
what the compiler is complaining about :)
Jeff LoucksMobile
253-691-8812
-Original Message-From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
SysGetStackInfo() reports the space allocated for use as a stack. The value
of the stack pointer (SP or a7) tells you where you are in the stack.
I believe SysGetStackInfo() returns the address of the first and last bytes
allocated to the stack, so actual stack size would be (end - start + 1).
To: Palm Developer Forum
Sent: 1/12/2006 9:17 PM
Subject: Re: Fixed string in code.
Well the problem was solved with inline assembly thanks to Jeffry
Loucks,
but for completeness sake the code was very similar to this. The code
wasn't
specified to be in inline, although it was small enough
I routinely place constants in code resources, such as tables and data
blocks that I want literally formatted and quickly referenced. It's also
easy to create any kind of literal resource using the resource compiler.
Normally, however, I don't care about the data being visible while not
sysErrNotAllowed
Os trap # 841 - sysTrapOEMDispatch
Check for any device specific calls that aren't appropriate for the T5.
Jeff Loucks
Mobile 253-691-8812
-Original Message-
From: Jan Slodicka [mailto:[EMAIL PROTECTED]
Sent: Wednesday, December 07, 2005 2:05 PM
This may seem a simple question to most, but there's a catch...
You must be careful which method you use to retrieve a ROM or OS version.
My Treo 650's disagree internally. For example:
My Sprint Treo 650 (CDMA) displays Palm OS Garnet v. 5.4.8 in the
launcher, and SysGetOSVersionString()
Katie,
Several problems that could cause a crash:
1. You pass in a (char **), implying you expect an allocated return buffer,
or a reference to some internal static buffer. Instead, you pass back the
address of an 'auto' buffer, one that temporarily exists on the stack only
as long as the
Title: Message
Try
the following:
1.
Call NetLibOpenCount() to determine if NetLib is already
open.
2.
If not, call NetLibOpenIfCloseWait() to see if NetLib was recently closed and
may not have already disconnected.
3.
If not already open and not in close-wait state, the network is
You have incremented appInfoP, so it no longer points to the originally
allocated memory. Save the pointer someplace to use in free().
Jeff Loucks
Mobile 253-691-8812
-Original Message-
From: Jim Duffy [mailto:[EMAIL PROTECTED]
Sent: Tuesday, November 08, 2005 3:57
Errr.. I meant originally locked and later unlocked.
Jeff Loucks
Mobile 253-691-8812
-Original Message-
From: Jeffry Loucks [mailto:[EMAIL PROTECTED]
Sent: Tuesday, November 08, 2005 4:05 PM
To: Palm Developer Forum
Subject: RE: Reading AppInfo block in DB
You
Returns 0 if the socket is closed, else -1 (error) or 0 for number of bytes
received.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent:
On a stream socket, the data never ends. Keep reading until you get all the
data you expect (if you know the length), or until the socket is closed.
For example, if you connect to a service that sends packets of known length,
read the socket until you have received a complete packet, and then
Try using the dialer helper. First, check for the existence of the dialer
helper. Second, use the helper to place the call:
Boolean DialPhone(char *pPhoneNumber, char *pDisplayName)
{
if (VerifyHelper(kHelperServiceClassIDVoiceDial,0))
return
Err FlpFToA (FlpDouble a, Char *s)
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: cbruner [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 12, 2005 12:55 PM
To: Palm Developer Forum
Try using the dialer helper. First, check for the existence of the dialer
helper. Second, use the helper to place the call:
Boolean DialPhone(char *pPhoneNumber, char *pDisplayName)
{
if (VerifyHelper(kHelperServiceClassIDVoiceDial,0))
return
pdat = (struct pdata *) MemPtrNew(sizeof(struct pdata));
pdat-idat = (struct idata *)MemPtrNew(5 * sizeof(struct idata));
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: Gnadinger,
Has anyone been able to call PhnLibCardInfoEx() on a Treo650 CDMA without
crashing?
I get the fatal error TILUtilsCDMAHtc.c, Line:69, Not NV read response.
This call works on Treo600.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile
For CDMA, the number returned is the ABS(dBm). A value of 113 corresponds to
-113dBm.
According to GSM 07.07 Section 8.5, convert from dBm to rssi [0..31]:
0 -113 dBm or less
1 -111 dBm
2...30 -109... -53 dBm [ (113 - s)/2 ]
31 -51 dBm or greater
To convert from rssi to bars
The netErrWouldBlock is normal if you read an empty non-blocking socket. It
means your call would have blocked to wait for data. All you do is come back
to it another time and try again. If there is data, you will receive it,
else you will get another netErrWouldBlock.
It is common to keep
ago
here in the posts?
Can't really remember, but if not, this is of course the best way to do it.
Regards
Henk
Jeffry Loucks wrote:
The netErrWouldBlock is normal if you read an empty non-blocking socket.
It
means your call would have blocked to wait for data. All you do is come
back
Check out socket notices. You set the socket to send a notice on one or more
events, and then go off and do other things until the notice (notification)
arrives.
OS5 supports only the 'notification' type of notice, on any combination of
the following events:
netSocketNoticeErr
Which NetLib call? netErrParamErr (0x1204) normally means a parameter value
is unacceptable. It's either an invalid value, or is a valid value that is
not acceptable in that particular instance.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL
.
This work has served me well and I'm offering it to the community (in a form
yet to be determined).
Thanks,
jeff
Jeffry Loucks a écrit, le 25/08/2005 03:48 :
I have received several requests for the background and thread stuff, so
I am assembling some documentation and a preliminary install. Please
I have received several requests for the background and thread stuff, so
I am assembling some documentation and a preliminary install. Please
allow me a few more days to make sense of it.
There have been offers of webspace, as well. Very much appreciated.
Thanks,
jeff
--
For information on
I've been using a homegrown thread manager in both foreground and background
for several years now. I could be convinced to make it available, if there's
sufficient interest.
Features:
1. Works on PalmOS 3.5 and later (including OS5 - haven't tried OS6).
1. Any number of threads, in any number
Docs say SslReceive() and SslRead() are the same, except SslReceive()
returns info on the socket (sender).
A similar issue I've heard from other Ssl users; are you setting the send
timeout, as well? It seems they have to be the same, or at least set
concurrently. Try setting the send and receive
First, you increment 'recordindex' before using, so you skip strArray[0] and
potentially write past the allocated array.
Second, since you skipped strArray[0], LstDrawList() tells the list draw
code to dereference a string pointer that is garbage and probably crashes
doing so.
Stanislav,
You can change the stack size at least two ways;
1) change the requested stack size at build time in the 'pref' resource
2) change the actual stack at run time
The first is simple. For example, if using PilRC, the following is supposed
to work:
SYSAPPLICATIONPREFERENCES ID 0 PRIORITY
You are not setting dbPtr back to NULL after closing the database, so the
second time through your Open() routine, the database is not reopened, dpPtr
remains pointing to garbage and Open() returns no error.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908
The inbound socket activity has already awakened the device and deferred the
socket notice. The deferred notification will be broadcast during
EvtGetEvent(), which your app is probably blocked in. You need to make sure
you defer sleep (see sleep event/notifications and management) long enough
to
Devices without Flash ROM do not have serial numbers available in ROM.
Devices without Flash use a mask ROM that cannot be modified, thus cannot be
programmed with a serial number.
Treo devices may have a serial number, but do not respond to the
SysGetROMToken() call. You must use the
There are several ways, depending on the device and telephony support.
Handspring uses the PhnLib. First, use PhnLibGetOwnNumbers() to get the list
of phone numbers. Then, depending on phone type CDMA or GSM, extract the
number (phnAddrFldPhone) from the first entry in the list. For CDMA, use
PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: Jeffry Loucks [mailto:[EMAIL PROTECTED]
Sent: Thursday, June 09, 2005 1:17 PM
To: Palm Developer Forum
Subject: RE: How to get Palm Phone Number
There are several ways, depending on the device
Alexandre,
Use socket notices.
1. Register to receive a notification, either procedural or app launch.
2. Prepare a socket to do what you need to do (stream or datagram).
3. Prime the socket to send a socket notice (notifications are the only type
supported by OS5).
4. When data arrives (or an
If you invoke modal UI objects, the event loop is out of your hands. Nasty
part of doing business with events (especially custom events) or using an
event loop.
There are several ways to prevent a device from going to sleep. I think all
of them have already been mentioned. If
One way would be to save the socket handle in a feature. Then, any app could
find the socket handle in the feature and use it. Be sure to properly manage
NetLib, so that it doesn't close out from under you.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908
Open a UDP socket, bind it, set up a code fragment somewhere (feature
memory, or whatever) and register it to receive a system notification of
your choice, and set the socket to send a notification to the code fragment
on socket receive/error. As packets are received on the socket, a system
Set the dmHdrAttrRecyclable attribute bit in the app database attributes
using DmSetDatabaseInfo(). You could do this while the app is running, or in
response to an alarm set for the expiration date.
dmHdrAttrRecyclable:
The database is recyclable. Recyclable databases are deleted when they are
I realize the question appears to be simply a play on words, but I'm serious
:)
If nobody knows if the call is synchronous, has anyone observed anything
that would indicate it is not?
Thanks.
-Original Message-
From: Jeffry Loucks
To: Palm Developer Forum
Sent: 5/10/2005 4:58 PM
Any one know if the new DmSyncDatabase() call is synchronous? I'd like to
know if, upon return, the database is completely flushed to flash or is it
simply flagged to be flushed. I suspect it is synchronous, but I don't know
what's behind the scenes, so I'm not making any assumptions.
Do you know the HAL id, as well?
Thanks
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: Douglas Handy [mailto:[EMAIL PROTECTED]
Sent: Friday, April 29, 2005 4:39 AM
To: Palm Developer
The bottom to top stroke, beginning in the Grafitti area and ending near the
top of the screen is known as the 'ronamatic' stroke. The OS generates a
KeyDownEvent with the value 0x010E (see below). The default OS' response to
the ronamatic character is to generate a KeyDownEvent with the value
I see a lot questions like what event is generated when such-and-such
happens?
Here are some suggestions:
1. Hook your device up to a debugger and watch what happens when you do
such-and-such.
2. Or, use the emulator and log events while you do such-and-such.
3. Or, write an app to simply dump
A7 points to the current top of a stack (always valid). The stack is
allocated in the dynamic heap.
A6 is the 'frame' pointer (sometimes valid). The frame exists on the stack.
A5 is the 'globals' pointer (sometimes valid). Globals is your apps' data,
and is allocated in the dynamic heap.
Sorry Ben, but with great respect I must disagree and beat a dead horse.
Your answer ignores the question.
1) The SD card solved the non-volatile issue several years ago without
breaking a lot of stuff. An embedded SD card would have been wonderful. NVFS
is a not.
2) An SD card (or something
Doug,
You are correct, it answers one interpretation of the question.
The question was What did NVFS buy us? Seems like its just breaking a lot
of stuff
IMHO, the question was not what was intended, but what did we get.
We got a non-determinant and unreliable implementation. I'm all for
It's likely that code flagged to run after system reset, is in a bad state
and is causing a repeating fatal exception. Thus, the flashing logo
(repeated resets).
Yes, I know, duh. But, it may be something you or the other package is doing
improperly on reset. Is any of your software flagged to
Free the memory when you no longer need it, but no sooner :)
If your callback registration comes and goes, you may want to leave the
memory allocated. If, after unregistering, you do not know when you'll need
it again, free it up. If it's a small amount of memory, you could leave it
for the next
Globals (globals, statics, and multi-segment jump tables) are stored in a
chunk of dynamic heap that is only allocated for the app when it is launched
normally. This memory is available to the app via M68K register A5. The
compiler knows this and generates code that accesses globals relative to
Sorry, for the length of this response. I had to leave for my son's
birthday. I would have liked to have been in the thick of this. After all, I
started it :)
I have a theory that it's a little more complex than that. I think
what happened is that the T5 was supposed to be an OS 6 device, but
Excuse me ahead of time - flames ahead...
It wasn't supposed to break anything; the implementation that shipped
on the T5 is a bit buggy, but palmOne and PalmSource are working on
improving it and making it work much better with new software.
Okay, I've been holding my tongue for a long
Check out the following:
UInt32 companyID,deviceID,HALID;
FtrGet(sysFtrCreator,sysFtrNumOEMCompanyID,companyID);
FtrGet(sysFtrCreator,sysFtrNumOEMDeviceID,deviceID);
FtrGet(sysFtrCreator,sysFtrNumOEMHALID,HALID);
The values returned define the platform. For
The 68EZ328 indeed supports a mulu instruction, but you'll notice it is a
16x16 operation. In other words, a mulu.S. You were trying to use a mulu.L,
which is not supported by the EC000 core.
Your original callback uses a global value to locate the multiply code
[move.l [EMAIL
So, just how many databases are ok to be open during MemHeapCompact()?
In the following list of open databases (on a Treo 650), my app is
responsible for the top three, leaving fifteen others that are out of my
control. How am I to make sure to close all databases before calling
MemHeapCompact?
Date/time math is simply:
1. convert dates/times to a common base unit of time since a reference
epoch, eg. days or seconds since 1904
2. add or subtract or compare converted dates
3. convert result back to desired date/time format
Works for all time, so to speak...
What I did not find (or
TILUtilsCDMAHtc.c, Line:69, Not NV read response
...
TIL - Telephony Interface Library (I think)
Utils - Utility functions
CDMA - what kind of radio, could also be GSM
Htc - Could be the contract manufacturer, HTC builds the Treo 650
Yes, that makes sense. What made me wonder about NVFS is
the complete instruction set and
interact directly with the microprocessor. Of course
we have to build a bootloader, drivers and a lot of
things that together make the HAL. What we want to
know is if there is any limitation concerning access
to the chip a very low level.
Thank you,
Leo
--- Jeffry
Anxiously downloaded and applied the 1.08 firmware update for Treo 650.
Firmware update went smoothly.
Our app makes heavy use of NetLib. It also hammers the storage heap and uses
a lot of small records, so have been severely impacted by the NVFS problems.
The new firmware greatly reduced the
Leo,
Have you ever written a hardware layer for an embedded microprocessor
system? If so, then you probably had access to the specifications, including
schematics and sample drivers. Do you have these specifications for any of
the Palm-ish devices you wish your OS to be compatible?
If you are
No, not 'difficult', but maybe not that interesting :0
Oh, sorry, that was rude... What I mean is that PDA apps generally don't do
things that, by their nature, aren't readily accomplished on PDAs. No, that
cuts the PDA a bit too much slack. Instead, developers have adapted to the
limitations of
Uhm, I thought that generating a report on the PDA and sending it to a
PC or to another PDA is something that PDAs are normally used for?
Quite right, but not necessarily in destination format. PDAs are commonly
used to acquire data of many forms, and sometimes this data is viewable, to
some
Yes, the hardware will do whatever you tell it, assuming it can. Beware the
'halt-and-catch-fire' opcode :)
But, that doesn't mean the device will end up acting like a PDA. There is
almost nothing the device does for itself (in hardware), so the hardware
must be closely managed or you don't get a
palmOne_SDK_4_1_Headers_Public_4_1_Build_31
68K/System/HsExt.h
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
Nigel Grant wrote:
Hi Guys
Can someone tell me what is name of the header
If a couple of megabytes isn't reliable, why on earth would I jeopardize 2
Gigabytes?
Pardon my attitude, but the PalmSource FFS gives me the creeps. I don't know
where my data is, when it's safe and when it's not. They could offer 2
Terabytes and it wouldn't make any difference.
Here's hoping
Yes, that's probably some of what we've seen, too.
Our company has put T5/650 dev on hold. The NVFS has thrown us too big a
curve. We have a DM intensive app, with many small and large records, lots
of notifications and intensive NetLib activity. It was hard enough on the
much more reliable
First off, you allocate 500 bytes and then tell NetLibReceive() you have
room for 1000 bytes. Second, you are passing the address of the pointer to
your buffer, not the address of the buffer. Try:
#define kBufSize 1000
void *str = MemPtrNew(kBufSize);
Int16 result =
EvtGetEvent() will return no later than your timeout (give or take). On
the real device, something is demanding attention at flank speed.
EvtGetEvent() has no way to tell you it isn't returning an event. It always
returns an event, just sometimes it's a nilEvent. Better yet, the nilEvent
may mean
Oh yes, forgot to suggest something.
If you wish to do timing with EvtGetEvent(), keep track of the next system
tick that you care about and pass the difference between now and the desired
system tick as the timeout to EvtGetEvent(). On return, check to see if
you've reached your desired tick.
I suspect one or more comm stack (or other driver) is needing to poll.
EvtWakeup() can be called by anything and will wake up the foreground. If
there are no real events, EvtGetEvent() will return a nilEvent.
EvtSetNullEventTick() can be called by anything to set the minimum ticks
until an
I've got some stuff I want to post and the stupid List Mangler keeps telling
me I've got unsubscribe commands and unsupported attachments in my post. I
don't have any of the sort, but I guess it sees something it doesn't like.
Is there a guideline somewhere about List Mangler idiosyncrasies?
The normal (old?) Palm way to retrieve the device serial number (assuming it
has one) follows. If the first byte == 0xFF, there isn't a serial number:
{
UInt8 *p;
UInt16 size;
err = SysGetROMToken(0,sysROMTokenSnum,p,size);
}
Handspring does it a different way. I suppose if the call returns an
The Palm SDIO Bluetooth card is implemented in two parts; an SDIO-serial
controller internally wired to the Bluetooth chipset with tx/rx/rts/cts. If
you want to part with your Bluetooth card, you could open it up and patch
into the serial port. If you're careful, you wouldn't necessarily break the
Am I missing something? (don't answer that :)
I regularly use the serial (RS232C) port on T-W, T-C, T-T, T-3, Kyocera7135
(unless the phone is on), Treo600, and Treo650. I realize the T-W and T-T
are old, but the rest are current.
I don't have any Zire devices, so no experience.
Past devices
You need to make sure your registered functions don't move around on you,
because the OS will not automatically adjust them. You do this by opening
the application database (and overlays, as necessary) and locking the
resources. The OS does this for you when you run your app, but unlocks and
Some other details...
Your app will need to know when it should unlock itself, eg HotSync and
deletion attempts. I have my app run on reset, and register for HotSync
start notification and delete protected database notification (I set the
protect attribute on my app).
Jeff
versha khar,
YMMV, but it can be as simple as:
1. Using NetLib within your app, open a socket to your website URL port 80.
2. Send GET / HTTP/1.0, or whatever query you need to send.
3. Receive on the socket until you get the entire answer.
4. Close the socket.
5. Parse the answer into your
PhnLibGetOwnNumbers() is defined in the Handspring telephony header file
HsPhoneNetwork.h, and is part of the Handspring SDK. The SDK I have came
from the Handspring developers site (www.handspring.com/developers), which
now redirects you to http://www.palmone.com/us/.
Jeff
at Palm development and having a great deal of
difficulty determining what technologies to use for
particular development. I have found no reference to a Handspring SDK,
could somebody give me some more specific
information?
Thanks,
John
Jeffry Loucks wrote:
PhnLibGetOwnNumbers() is defined
This is what we've encountered while looking for
phone identity on Treo 6xx:
1. If the phone is type hsAttrPhoneTypeCDMA (see
HsAttrGet(hsAttrPhoneType)):
// Treo 600 returns ESN
// Treo 650 crashes!
char *pEsn;
PhnLibCardInfoEx(telRefNum,0,0,0,pEsn,0,0,0,0);
Yes, and sometimes it even works! Here's what I've done:
// Try to get our phone number - sometimes this even works!
// The own phone number is the first in the own numbers list.
// NOTE: The phone number is not always available right away.
// For example, a GSM phone number may not be available
You are confusing resource index with resource ID. Look closely at the
following definitions and you will see that DmDetachResource() requires the
index (UInt16) of the resource to detach from the database, while the
DmAttachResource() requires the resource ID (DmResID) of the resource to
attach
I came into this thread late, so please excuse me if I cover ground already
covered.
There is a piece of code floating around the net, maybe even around here
somewhere :), that will extract a ROM image from an actual device. Older
ROMs can be extracted from older devices. Older devices are
DmFindDatabase returns a LocalID, which is a UInt32. You have dbID defined
as a UInt16, which is too small. You either ignored the error/warning about
it or have them turned off.
typedef UInt32 LocalID;// local
(card relative) chunk IDLocalID
CtlGetLabel() is defined as returning (const char *). The compiler will
complain that you are assigning a (const char *) to a (char *). The 'const'
keyword is the issue. The 'const' tells the compiler to treat the value as
non-writeable, in this case the data the returned pointer points to.
I apologize ahead of time for my verbosity. Those who know me rarely involve
me in a conversation without having a pre-arranged escape plan.
Okay, the curmudgeon's gotta speak.
Generally used to make a null pointer clear, NULL has been simply eye
candy
for the value 0.
But recently, I've been
Is it possible that more were deleted than were retained? I, too, was
deleted after several years. Maybe it would be useful to ask how many
were NOT removed, and see what they had in common.
-Original Message-
From: Ben Combee
I have collected all the missing account reports and
Okay, the curmudgeon's gotta speak.
Generally used to make a null pointer clear, NULL has been simply eye candy
for the value 0. NULL is often confused with NUL, which is simply eye candy
for the character value of 0. In addition, NULL has been used as a scalar in
logical or symbolic math
?
Jeffry Loucks wrote:
Okay, the curmudgeon's gotta speak.
Anybody want to talk about true and false?
Well, not exactly. But I have a question: I recently wrote a set of
fixed-point math functions (24 to the left and 8 to the
right) and defined ceil(a) as (a 8) + !!(a 0xff).
So my question
Any help would be appreciated.
I've asked this question mostly before, but so far nobody admits to knowing
anything about the subject.
I need to programatically retrieve the ESN, IMSI(MIN) and phone number from
the Treo 6x0 CDMA phones for use in identifying the phone and subscriber to
a
The following retrieves the phone type, either hsAttrPhoneTypeCDMA or
hsAttrPhoneTypeGSM:
UInt32 temp;
HsAttrGet(hsAttrPhoneType,0,temp);
It is a Handspring function with the following definition:
#define sysTrapHsSelector sysTrapOEMDispatch
#define hsSelAttrGet
The following is for the Treo 600. I haven't figured out the Treo 650 yet
(looking for documentation).
Determine phone type with HsAttrGet(hsAttrPhoneType,0,temp), eg:
{
UInt32 temp;
HsAttrGet(hsAttrPhoneType,0,temp);
if (temp == hsAttrPhoneTypeCDMA)
I've got to stop replying while I'm so tired.
Correct my last to read CDMA instead of CDMS, and all samples of 600
instead of 650.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: Jeffry
PROTECTED]
Mobile 253-691-8812
-Original Message-
From: Jeffry Loucks [mailto:[EMAIL PROTECTED]
Sent: Tuesday, December 14, 2004 3:39 PM
To: Palm Developer Forum
Subject: RE: Retriving ESN on CDMA Treo devices
The following is for the Treo 600. I haven't figured out the Treo 650 yet
Hi all,
I have a wireless app for a Treo600 that needs to retrieve (via API) some
kind of phone ID, like a phone number, IMEI/ESN, IMSI/MIN. I have been able
to find the phone number , IMSI and IMEI for a GSM phone, but can only find
the ESN for a CDMA phone. I'd like to get the phone number and
I don't see my pics show up anywhere on my WinXP machine after I HotSync my
Treo600.
I use an SD card to retrieve the pics.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
-Original Message-
From: Manpreet
Does anyone know how to programatically retrieve the identity info (MSID,
MIN or MDN) on a CDMA phone? The GSM library is fairly well documented and
works as expected, but I don't see/know a way to access the info with the
CDMA library. Any help would be appreciated.
Thanks.
What kind of 'crash' happens? The mechanism is stable, and I use it a lot.
Several thoughts;
1. Which Treo and what version of OS? I assume Treo 6x0 and OS5, since
socket notices were not announced until OS5.
2. I assume you are using the system notification mechanism of socket
notices, since
Be careful with the aftermarket power supplies. This applies to cables we've
found in several 'reputable' stores, not just the internet. While the cables
worked just fine, the wall warts produced neither sufficient voltage nor
current to power or charge the device. I ended up splicing in some
Connect a terminal emulator to the Palm serial port, open the console port
and write to it. What you write will be displayed on the terminal emulator.
Some issues:
1. Use a serial HotSync cable.
2. Make sure the terminal emulator serial port is configured the same as the
Palm serial port (baud,
If you must rely on the sort index, leave 'deleted' records in place and
simply mark them deleted. You can reuse the records by linking them together
in a 'deleted' list.
Jeff Loucks
Work 425-284-1128 [EMAIL PROTECTED]
Home 253-851-8908 [EMAIL PROTECTED]
Mobile 253-691-8812
1 - 100 of 114 matches
Mail list logo
