Thanks for this. I have adjusted my approach to suit and it works well.
> -----Original Message-----
> From: [EMAIL PROTECTED] [SMTP:[EMAIL PROTECTED]]
> Sent: Thursday, June 17, 1999 6:40 PM
> To: [EMAIL PROTECTED]
> Subject: Re: When to lock a chunk?
>
>
>
> Try it this way:
>
> typedef struct {
> long value;
> VoidHand text;
> } MyType;
>
> MyType MyArray[10];
>
> MyArray[0].value = 5;
> MyArray[0].text = MemHandleNew(StrLen("Bob") + 1);
> CharPtr s = MemHandleLock(MyArray[0].text);
> StrCopy(s, "Bob");
> MemHandleUnlock(MyArray[0].text);
>
> ...
>
> VoidHand h = FldGetTextHandle(someField);
> CharPtr s1 = MemHandleLock(h);
> CharPtr s2 = MemHandleLock(MyArray[0].text);
> StrCopy(s1, s2);
> MemHandleUnlock(h);
> MemHandleUnlock(MyArray[0].text);
>
> Of course, you should also make sure that the handle returned by
> FldGetTextHandle is large enough to hold the string you're copying into
> it.
>
> -- Keith Rollin
> -- Palm OS Emulator engineer
>
>
>
>
>
>
> Jason Garrett <[EMAIL PROTECTED]> on 06/16/99 09:21:28 PM
>
> Please respond to [EMAIL PROTECTED]
>
> Sent by: Jason Garrett <[EMAIL PROTECTED]>
>
>
> To: [EMAIL PROTECTED]
> cc: (Keith Rollin/HQ/3Com)
> Subject: When to lock a chunk?
>
>
>
>
> I have a programming dilemma....
>
> I have defined an array
>
> typedef struct {
> long value;
> CharPtr text;
> } MyType;
>
> MyType MyArray[10];
>
> To set the text property, I ...
>
> MyArray[0].value = 5;
> VoidHand h = MemHandleNew(StrLen("Bob") + 1);
> CharPtr s = MemHandleLock(h);
> StrCopy(s, "Bob");
> MyArray[0].text = s;
> MemHandleUnlock(h);
>
> Now, I get away with that. But, when I want to use 'text' again later...
>
> VoidHand h = FldGetTextHandle(someField);
> CharPtr s = MemHandleLock(h);
> StrCopy(s, MyArray[0].text); <<<<<<<<<<<<< Error: Accessing unlocked
> memory
> MemHandleUnlock(h);
>
> I get the indicated error when using POSE (which is great, because it
> picks
> up errors that
> my have gone undetected!)
> Can someone please correct me, and get me over this problem. There is a
> strong chance that
> I have taken a poor approach to this, and there is a much better and
> correct
> method.
>
> Please! and Thanks!
>
> Jason
>
>
>
>
>
>
