"Thomas Lösel" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Hello,
>
> I wanted to start the Blazer from my application in order to display
> help pages. I used code as follows:
Search the KBs for "WebBrowserManager" sample code. It let's you launch
various browsers with a basic
Are you by any chance having this problem (and have solved it)?
http://news.palmos.com/read/messages?id=201159#201159
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
I settled on AppNetLib but am having this problem:
http://news.palmos.com/read/messages?id=201159#201159
What did you decide?
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
If you can get the sdk-5r3 files, here is a way to move these 2 object files
into the sdk-5r4 files (MANY THANKS to Ton van Overbeek!!!):
Set aside these 2 files by moving them into a subdirectory named
"sdk-5r3":
cd /PalmDev/sdk-5r3/lib/m68k-palmos-coff
mkdir sdk-5r3; mkdir sdk-5r3
"Alan Ingleby" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Here's what I use:
>
> Boolean HelperDeviceHasNVFS()
This will surely work to detect NVFS but I have heard that some systrap*
call can be made to test the presence of the actual API command itself. I
have other reasons t
I am finally (because I have to in order to work around a bug on the OS)
call an API that is not fully backwards compatible (and also doesn't have a
glue function). So I will need to test that the API exists before I call
it, lest I crash on older HW. What is the proper way to do this?
In case it
Coincidentally, DateGlueTemplateToAscii.o is also missing (but I don't need
that one).
This can be confirmed by editing the file and scanning for the string
"DateGlue". These objects simply aren't there!
I had been using sdk-5r3 for PRC-Tools for the longest time with no
problems. Then today I u
Several years ago somebody in here passed along this code to calculate free
RAM but I have just found out that the algorithm is deficient on new devices
(TX and later, at least). Yes, I know there is an article from 2001 in the
KB (bottom of this article) which is very similar so I assume (since i
...which leads either to a timeout in the later NetLibReceive (or, in my
case NetLibDmReceive) *or* in the remote server disconnecting (because it
timed out waiting for the command we thought we sent but it never received).
My code works fine in all real devices and on the emulator but since I am
a
My app uses code from the KB that says, essentially, "if my Processor is
*x86, then I am running on POSE". I am concerned that the new "Windows
Palm" will cause this function to return "true", too (i.e. "I am running on
POSE"). Does anybody know for sure?
--
For God, Family and Republic!
My plan
The AppNetLib source/project allows http get commands to be sent "to the
internet" and receives the response (up to 32K) to a buffer. The problem I
am having is that it usually "stalls". The initial response is received but
not everything so it just sits at a "middle" stage forever. I have found
Never mind, I was doing something else wrong.
The comment is evidently correct.
When I set it to commandMask *only* it works (same for vchrMenu which also
works with no mask set).
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/d
// The following are key codes used for virtual events, like
// low battery warnings, etc. These keyboard events MUST
// have the commandKeyMask bit set in the modifiers in order
// to be recognized.
#define vchrLowBattery 0x0101 // Display low battery dialog
...
#define vchrFind 0x010A
...
vchr
I am trying to make sure I am properly locking and unlocking records in my
DBs while reading the data and as part of my testing I am occasionally
leaving some records locked and close my DB/app. I assume that a soft reset
will clear all locks on records (set the lock count for all records in all
D
"Logan Shaw" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Roger Stringer wrote:
> >> From: "Gregg Woodcock" <[EMAIL PROTECTED]>
>
> >> But what is killing me (and has ALWAYS worked fine) is code like this:
"Gregg Woodcock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Follow-Ups set to tools-forum.
Boy did I mess that up. Followups were (actually set to emulator-forum) and
evidently the new NG software reduces all crossposts down to the first NG
(so it didn&#
Follow-Ups set to tools-forum.
I am troubleshooting bad code which has become evident because "New NVFS
from garnet 5.4.9 can remove any resource (locked or unlocked) from closed
unprotected database."
Ben's example of bad code was this:
DmOpenDatabase
DmGetResource
ptr = DmLockHandle
OOPS! Actually I use resource entries like this:
ICON ""
SMALLICON ""
Are there ICON families, like BITMAP families, too?
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
"Robert Moynihan" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Then try something like this instead.
>
> BITMAPfamilyex ID bitmapID_exclaim
> compress
> begin
> BITMAP "bitmaps/exclaim.bmp" bpp 1 density 1 transparent 255 0 0
> BITMAP "bitmaps/exclaim-c.bmp" bpp 8 density 1
"Doug Gordon" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> And in the good-old Constructor you do it when you create a Bitmap Family
resource in a similary way -- check the transparency option and then assign
the color.
If only
I use cygwin/GCC and use "flat" bitmaps reference
I know this is not directly possible because all that can be done to a field
is to set the foreground/background colors and then display the text. So I
know that I'll be stuck "painting" the text somehow. The problem is that I
also want to retain the built-in text features that come with using a
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
>
> >Now we are getting someplace (BTW, shouldn't there be some "porting your
app
> >to NVFS" KB/doc somewhere?)
>
> Have you read the Palm SDK Developers Guide, downloadable from
> pluggedin.palm.com?
No. And I haven't beca
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
>
> >The following obtuse comment from another developer seems to confirm my
> >vague suspsicions but did not provide details, references, documentation
or
> >work-around:
> >
> >"New NVFS from garnet 5.4.9 can remove any
I have heard rumors of there being serious problems with Tungsten-X and NVFS
causing crashes on just about any app under certain (common) circumstances.
Yet I cannot find anything in the dev forums or KB so I am asking here. My
app (New Yanoff: http://www.PalmYanoff.com) plays no tricks of any kin
I would like my app's icon to be transparent (instead of white block
background) but as I understand it, bitmap does not support transparency
(unlike other image formats like GIF). So how is this done (I am using
Microsoft Paint on W2K if that makes the explanation easier)?
--
For God, Family and
"Bodo Wenzel" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> > Why does the state need to be tracked?
> My application runs on OS >= 3.0 that doesn't have
MenuCmdBarGetButtonData().
> So I have to detect the opened command bar via the event which occurs only
on
> OS >= 3.5.
>
> > It
"Bodo Wenzel" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
Why does the state need to be tracked?
It seems to me that once we ensure that all opened/opening command bars have
at least 1 icon, then any time we want to check the status of the command
bar (in my case this is very, very
I asked this in the BlueTooth forum a few days ago and got no response; I am
hoping somebody in this wider audience can help (it isn't really "that kind"
of a BlueTooth question anyway)...
I am having trouble getting full Palm network access through BlueTooth
working because it is apparently bypas
I asked this in the BlueTooth forum a few days ago and got no response; I am
hoping somebody in this wider audience can help (it isn't really "that kind"
of a BlueTooth question anyway)...
I am having trouble getting full Palm network access through BlueTooth
working because it is apparently bypas
I want to ignore a screen tap if the command bar is open. The problem I am
having is that my users like to abort the command bar timeout by tapping
somewhere outside the command bar (i.e. my forms main field). So I want to
ignore such taps. The regular ways to do this (monitor winEnterEvent and
"Gregg Woodcock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> The default behavior for non-editable fields is to highlight a word when
it
> is double-clicked (double-tapped). I would like to do something with the
> text depending on a word that is sing
The default behavior for non-editable fields is to highlight a word when it
is double-clicked (double-tapped). I would like to do something with the
text depending on a word that is single-clicked (single-tapped) which, so
far as I can tell, has no default behavior.
I know I should capture the pe
"John Marshall" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Gregg Woodcock wrote:
> > I know he was at Palm and then at enterprise.net
>
> Enterprise.net was merely an ISP that I used in the UK
"was at" in the sense of "had an
I know he was at Palm and then at enterprise.net but I am getting "User
Unknown" from his email at enterprise! I would like to point out a few
minor updates for some of his Palm web pages if I can contact him. Somebody
shoot me his email, please?
--
--
For God, Family and Republic!
My plan to
I am converting an app to use SSL and it appears I will have to rewrite the
original NetLib code to use an non-Dm-protected buffer. I chose the Dm
"flavors" of the NetLib Send/Receive for the benefit of the memory
protection they provide and do not want to switch over if there is some way
to get D
"Henk Jonas" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> You could put the global find code in an extra code resource, which you
> lock in your PilotMain and jump into to perform the search.
Henk, can you provide some guidance or documentation on how to do this (code
example, may
After adding some fixes to my Global Find code, I am finding that my app's
main code segment is too full. If I want GF to work when the app is not
active and I want to keep my user validation code (I assure you I do), then
there is nothing I can remove from code0 to reduce the size. I have alread
"Dave Lippincott" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Read the Handera SDK.
I would love to but I am coming into this late and, so far as I can tell,
this is no longer available!
--
For information on using the Palm Developer Forums, or to unsubscribe, please
see htt
"mguo" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]
> They can be intercepted inside your main event loop, before SysHandleEvent
has a chance to handle it.
BUT WHAT EVENT IS IT? It is NOT a keyDown...
--
For information on using the Palm Developer Forums, or to unsubscribe, ple
When the "Back" button is held in on the HandEra 330, the "Voice Pad" app is
launched. Evidently this is not done by the OS sending a keyDown event
(which is too bad). What event and parms are generated and can they be
intercepted and handled by a form's event loop (hanled=true)?
--
For infor
This is a fairly general question and I am sure there are many good ways to
handle it...
Right now I trap keyDownEvent and filter each character one-by-one. Now we
all know that keyDownEvents are not generated from they popup keyboard so
this allows users to enter invalid characters.
Currently I
How can this be done?
The only way I can see is to (illegally) examine the LineInfoType data. Is
there a legitemate way?
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/support/forums/
"Chris Faherty" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
>
> Try 5.1. Simulators past 5.1 hang for me as well on Windows 98. It's not
> your fault :-) They're broken.
There is no 5.1 (5.0 or 5.2) on the website any more; anybody have a link
for it?
--
For information on u
I have a bug which is trampling my form data ("no form to return to" and
other crashes). I know roughly where this is happening but I have gone over
all the code a million times and it looks great. I know it is NOT a dynamic
alloc problem because I have implemented the KB code to track buffer
ove
I am finding that when I open the DIA (or use a small Fonts4OS5), the number
of rows for my table do not get updated. TblGetNumberOfRows() always
returns '10' no matter what I do; even though my manual calculations says I
should be getting 16! What's worse is there is no API to force a change to
I have a ThinModem+ that whose memory has become corrupt. Normally this is
no big deal because I would just reflash the original software but this
corruption is strange in that as soon as the screen is tapped (even on a
no-notify restart), the OS crashes. Here is as far as I can get:
Reprogram Ha
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 06:34 PM 5/16/2004, you wrote:
> >I cannot believe there is no (easily located) step-by-step guide to
convert
> >a standard manual bsh:make project for use in PODS. Has anybody done
this?
>
> There's no guide, as the cur
I cannot believe there is no (easily located) step-by-step guide to convert
a standard manual bsh:make project for use in PODS. Has anybody done this?
Right now I have a doze *.c and *.h files, one *.prc, a few *.bmp, and a
makefile. I have several make targets:
pay, compiles the shareware versio
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 03:31 PM 5/10/2004, you wrote:
> Can you run your application using POSE, with all debug checks on and a
> debug version of the 4.1 ROM?
As I mentioned in the Emulator forum, under the subject, "New simulator
users: 3 si
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> The surprise was that when I tried calling the Network pseudo-panel on a
> debug 6.0.1 simulator, I got expected behavior -- it appeared, showed a
> Done button, and when I hit it, my launcher app was restarted.
>
> Maybe th
In another thread I noted a backwards-compatibility bug in Cobalt (see
"Cobalt (OS6) simulator does not return to app from LaunchWithCommand to
NetworkPanel!") and, as is all too typical for me, the thread had no
response. OK, well I'll take the other tack now and fix it for myself by
kludging aro
Here is my code:
// Allow OS events to be processed like power button, etc.
for (userCancel = false; (!done && !userCancel);)
{
//work here
do {
EvtGetEvent(&event, 0);
if ((!PrgHandleEvent(prg, &event)) && (PrgUserCancel(prg)))
{ userCancel = true; } // Don't stop; get *
I launch a Network Panel from within my app and want to return to my app. I
use this command:
LaunchWithCommand(sysFileTPanel, sysFileCNetworkPanel,
sysAppLaunchCmdPanelCalledFromApp, NULL);
This works fine and returns to my app for all OS versions except Cobalt
(OS6) on the sim
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 11:18 PM 4/24/2004, you wrote:
> >Last I heard we were screwed. Is this still the case? I *am* using
> >PalmOSGlue library...
>
> The Palm OS 5 SDK has always worked with prc-tools.
This is not true according to severa
Last I heard we were screwed. Is this still the case? I *am* using
PalmOSGlue library...
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/support/forums/
It appears at least for OS3.* it silently fails. I check Err but it is
errNone for both the Set and the Get. I can use the same exact code to
modify a more "normal" but such as dmHdrAttrBackup and it works just fine.
Is there some other way to Get/Set the Bundle programatically?
--
For informa
I want to generate a 48x48 1-bit bitmap and display it on my PDA based on
X-Face (http://www.mythic-beasts.com/tools-toys/xface/). The encoding is a
small plain text string. Surely somebody has source code for Palm to do
this, right? Care to share anybody?
--
For information on using the Palm
I just discovered that under many circumstances, the program you are using
keeps processing when the PDA is turned off so long as the PDA is plugged
into a power source (i.e. charging). For example, on my Visor Prism, any
program that is in a Progress Dialog loop, so long as it does not generate
s
I have added a few editing operations (Rot13, etc.) to the standard edit
items (cut/copy/undo) and want to note that none of my operations are
"undoable". I'd like to use the same dialog that the built-in functions do
but don't know how to do it or even whether it is safe/acceptible to do so.
I as
"Greg Lutz" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> I've been watching the responses on this thread with some amazement,
> because so far, no one has noticed the essential problem: it's a compiler
> bug all right, for not rejecting the code out of hand!
>
> It's the first sta
"John Marshall" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Gregg Woodcock <[EMAIL PROTECTED]> wrote:
> > You said, the 2 different definitions are "entirely different"
> [...]
> > In other words, the 2 are exactly the same
>
"John Marshall" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
Your attitude is among the most unwarranted rude, and condescending I have
ever experienced.
Your behavior is at such an inappropriate level that you really should seek
to address it if only for your own benefit.
It is a H
"Hildinger, Robert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> "Greg Lutz" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]
> Gregg's original code: static const char threeChars[] = {"123"};
> Code from your reply: static const char threeChars[] = "123";
>
> Gregg's
"Gregg Woodcock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Please note that in all the FrmCustomAlert calls, I forgot to add the
third
> string varaible which I intended to be ">".
MY BAD!!! It has nothing to do with "char" ->
"Hildinger, Robert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> "Gregg Woodcock" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]
> This must be a compiler bug. A "Char" is typedef'd to be equivalent to a
>
Please note that in all the FrmCustomAlert calls, I forgot to add the third
string varaible which I intended to be ">".
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/support/forums/
I am trying to be a GoodCoder (TM) and did a global replace to conform to
the Palm standard data type naming convention (e.g. "int" -> "Int16", etc.).
I use cygwin/GCC and have encountered a strange bug after this conversion.
Please don't ask why I am doing it this way because really I'm not; this
"John Marshall" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Gregg Woodcock <[EMAIL PROTECTED]> wrote:
> > OK, now that you've had your fun (which is somewhat deserved), hopefully
you
> > are willling to help me actually figure out WHY
"Ton van Overbeek" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> In article <[EMAIL PROTECTED]>, Gregg Woodcock wrote:
> >
> > Armed with the map file I can see that indeed, it is now (and actually
> > probably was in the beginn
"John Marshall" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> It might be worth changing the suggested usage to "-Wl,-T,text_64k",
> which would more reliably warn idiots who (mis)use it at compile-time,
> but the advantage is offset by the probability of said idiot mistyping
> all
"Keith Rollin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> What SDK are you using? The closest I could find in the current SDK to
what you cite is the following:
> typedef enum WinDirectionTag
> {
> winUp = 0,
> winDown,
> winLeft,
> winRight
> } WinDirectionType;
>
> Before Palm
I cannot run gcc compile with "-pedantic" as the Palm SDK are non-compliant
with strict ANSI C standards. The big offender is non-int enums declared
like this:
typdef enum { UP, DOWN, LEFT, RIGHT } DirectionType;
instead of like this:
typdef enum { UP = (int)0, DOWN, LEFT, RIGHT } Direction
"John Marshall" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> [I am unsure which of "PLEASE" in the subject line and "woefully
> inadequate" makes me less inclined to help you. Nevertheless...]
>
> Gregg Woodcock <[EMAIL PROTECTED
"Gregg Woodcock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> The only dox/info I can locate online are woefully inadequate:
>
http://drs.yahoo.com/S=2766679/K=text_64k_palmos3+/v=2/SID=w/l=WS1/R=1/H=0/*-http://prc-tools.sourceforge.net/doc/prc-tools_3.html
One more thing; I am using latest Cygwin release (i.e. latest GCC) and 3.5
SDK. Perhaps this only works for some specific late version of GCC/SKD?
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/support/forums/
The only dox/info I can locate online are woefully inadequate:
http://drs.yahoo.com/S=2766679/K=text_64k_palmos3+/v=2/SID=w/l=WS1/R=1/H=0/*-http://prc-tools.sourceforge.net/doc/prc-tools_3.html
As I understand it, this allows the primary code segment (#1) to grow (more
or less) up to 64K (instead
OK, I finally got my makefile doing what I want and here's the code (I am
using 'VERSION ID 2 "mm/dd/yy"'):
mv myapp.rcp myapp.rcp.old
awk -v awkdate='"$(shell date +%m/%d/%y)"' \
'/VERSION ID 2/ { $$4=awkdate } { print $0 }' \
myapp.rcp.old > myapp.rcp
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 04:00 PM 11/20/2003, Gregg Woodcock wrote:
> >However the beta of PilRC does not like my include lines the way 2.9
did
>
> Which beta? 3.0b2 or 3.0b4? What kind of error are
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 03:27 PM 11/20/2003, Gregg Woodcock wrote:
> >This is the best way to handle it. And while you are at it, add a
> >__VERSION__ that PilRC will copy the 'tver' 1/1000 value o
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 02:01 PM 11/20/2003, Gregg Woodcock wrote:
> >P.S. How do all you programmers handle the creation/compile/release date
in
> >your apps? Do you use a string resource in the RSC? Do
"Ben Combee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 02:47 PM 11/20/2003, Gregg Woodcock wrote:
> >Now, what I was asking is for you to give me a good reason *NOT* to do
it.
>
> The 'tver' resource type is reserved for use as spe
"Keith Rollin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 9:41 PM +0100 11/20/03, Aaron Ardiri wrote:
> > > I think this is (essentially) what I will do however I think it is
best to
> >> make it a non-1/non-1000 'tver' resource. I am using tver 1 for the
real
> >> versio
"Keith Rollin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> At 2:36 PM -0600 11/20/03, Gregg Woodcock wrote:
> >"Keith Rollin" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]
> > > Why not hard
"Keith Rollin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Why not hard code it as a string resource, and use that string to (a)
initialize your form, and (b) use it for the other part of your program?
I think this is (essentially) what I will do however I think it is best to
mak
"Keith Rollin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
Why not hard code it as a string resource, and use that string to (a)
initialize your form, and (b) use it for the other part of your program?
Because it is wastes more string space (I have to have the string resource
date
I hard-code the date my app was compiled as a label inside my "about" form.
I'd like to get at this string from another part of my code without opening
the "about" form. So I guess what I need to do is open the form resource
and then somehow extract the label control's string data. Is this possib
Sorry for posting this here but there doesn't seem to be a "news" or "admin"
group for this kind of thing...
I am glad to see the NGs back online but there is a rather serious NTTP
non-compliancy now as follows:
In response to HEAD command, this new server responds with 220 (head *AND
BODY* follow
I would like to do something smarter than just have the entire screen
redrawn every time an udpate needs to happen. My screen is a table and each
row is an entry into a pdb which has potentially thousands and thousands of
records. That means to access DmQueryRecord might take over a second for
ea
Calling NetLibOpen ALWAYS opens a connection if one is not present. I don't
want to do that. I want to do this:
If he is online already when the app starts, use the connection that is
already open AND DO NOT CLOSE IT when the app/work ends. If the connection
is NOT already open, open it when it
89 matches
Mail list logo
