Re: Text box throws an Error 7 Out of Memory with less than 27K in it!



"Steve Richfie1d" <Steve@xxxxxxxxxxxxxxxxxxxxx> wrote in message
news:43681rF1lserdU1@xxxxxxxxxxxxxxxxx
> Bob,
>
> There seems to be a basic (no pun intended) disconnect here:

I'd have to agree with Bob here (not necessarily on his delivery though :-),
the textbox was reliable in it's acceptance of exactly 32k of text. Can you
use this code to add the text and see if you still get the error:

Private Sub AddText(ByVal TB As TextBox, Text As String)
On Error GoTo Fail
Dim s As String
1 s = TB.Text
2 s = s & Text & vbCrLf
3 While Len(s) > 32768
4 s = Mid$(s, 10000)
Wend
5 TB.Text = s
Exit Sub

Fail:
MsgBox Erl & " " & Len(TB.Text) & " " & Len(Text) & " " & Len(s)
End Sub



>
>> I could fire up VB5 and build a trivial test app to pad a textbox out to
>> those exact numbers and then concatenates that many more, but I already
>> know *exactly* what would happen.
>
> So do I - it would almost certainly work great and as described in the
> literature. However, on a breakpoint I hovered over Len(txtLog) and it
> gave me the length, then I hovered over what was being concatenated and
> counted the characters on the screen, then F8'ed the simple statement to
> concatenate the two strings and got the error I posted. Hence, it *IS*
> real.
>
> OBVIOUSLY SOMETHING is happening here that at least some of the time
> fails. Further, when I changed the logic to chop at 20K instead of 30K the
> problems went away. I'm just wondering if the problem REALLY went away, or
> just stopped crashing in present testing only to cause problems after we
> start shipping. Things like this could give you nightmares when you don't
> understand WHY they failed - which is my present situation.
>
>> When you see something happening which you think should be impossible,
>> don't conclude that the impossible is happening.
>
> Why not - I see the impossible at least weekly! Just read some of my prior
> postings here - others have been pretty successful at identifying the
> obviously impossible things that I have been running into. There is a LOT
> in VB that doesn't work as advertised, or as you would think is
> reasonable. Some things even work differently in the IDE than they do in
> EXEs!
>
> BTW, this application is VERY unusual, with 4 different real-time AcitveX
> controls coordinating speech synthesis and recognition in real time
> conversations. With all this going on, damn near anything is possible! I
> am apparently pushing VB *WAY* beyond where it was ever intended to go,
> e.g. I must continue servicing events while in my error handler to be able
> to debug this program!
>
>> It's SO much more
>> likely that you're missing something that really *is* happening.
>
> Nope! Again, just read some of my prior postings here. Most are obviously
> impossible, but VERY real.
>
>> Therefore, describing the impossible to the newsgroup and asking for a
>> solution is mostly just a waste of everyone's time.
>
> It certainly hasn't been a waste of MY time, as Michael and Tony have dug
> down to exactly WHY some of my impossible happenings have happened.
>
>> Post [minimal] code which demonstrates the problem.
>
> Hey, I used to work supercomputer compiler support at CDC. When dealing
> with someone who might actually FIX a problem (no chance with VB5),
> minimal code is important. Unfortunately, there is no central public bug
> repository to check before diving into such efforts to avoid wasting time.
>
> Many real-world problems just don't reduce to minimal code, ESPECIALLY
> where something is blowing memory, stacks, etc., which is SO common in
> Microsoft's crappy C coded systems.
>
> My favorite such bug at CDC was a benchmark that ran instantaneously. The
> customer looked at the code that the compiler produced, and there was the
> outer DO loop (it was coded in FORTRAN) structure but the 10 pages of
> complex code within the DO loop had evaporated! Some debugging showed what
> the problem was - they didn't output or return any of what they were
> computing, so the compiler statement-by-statement had worked its way
> backwards through the program deciding that each statement was extraneous.
> I reported back why the compiler was working OK and if they just wanted to
> waste time (benchmark) that they needed to output something for their
> efforts, and submitted an internal low-priority bug report that empty DO
> loops weren't being eliminated!
>
> BTW, remember when the ANSI C standard was held up for a year? That was
> myself and 2 other guys trying to get critical bugs fixed in the
> specification. They were never fixed, and now are in part responsible for
> the general unreliability of present C coded systems, including the
> security holes that Microsoft keeps patching. Who knows, the loss of that
> political battle years ago could underlie this very problem!
>
>> Most of the time - while you're in the process of isolating just enough
>> to make it reasonable to bring it here and post - you'll FIND what's
>> really happening,
>
> I did that by breakpointing and checking absolutely everything involved in
> the simple concatenation statement that failed.
>
>> and you'll see how to fix it. If you did that first, you wouldn't even
>> have to ask the question.
>
> Again, WRONG! This program definitely does NOT live in the simplistic
> world that you clearly envision.
>
> My recommendation: Hold your objections while Michael or Tony or ??? have
> a crack at this.
>
> Steve Richfie1d


.



Relevant Pages

  • Re: Bug in latest IAR MSP430 compiler optimization???
    ... It has been assigned bug id EW20095, and it will be fixed in the ... The tests compiler companies do are quite extensive. ... register allocation when combined with post-incs and rollup. ... IAR use an industry standard test suite like this (that you don't ...
    (comp.arch.embedded)
  • Re: Salford compiler: Assigning a value to a constant expression is invalid
    ... a bug, it is not usually wise, I think, to assume that it's a compiler bug. ... -|> I'd suggest submitting a self-contained source code file to Salford Support. ... So all I had to do in that case was send the routine and ...
    (comp.lang.fortran)
  • Re: timeouts
    ... If I report a bug in the GCC C compiler, for example, then I can ... Stephen> a support contract for it. ... option when exclusively developing open source software. ...
    (comp.lang.ada)
  • Re: Malcolms new book - Chapter 1 review
    ... Aside from the diagnostics the compiler ... the code has a bug that should ... The book next offers two macros (mentioning two more, ...
    (comp.lang.c)
  • Re: Text box throws an Error 7 Out of Memory with less than 27K in it!
    ... those exact numbers and then concatenates that many more, but I already know *exactly* what would happen. ... there is no central public bug repository to check before diving into such efforts to avoid wasting time. ... Many real-world problems just don't reduce to minimal code, ESPECIALLY where something is blowing memory, stacks, etc., which is SO common in Microsoft's crappy C coded systems. ... The customer looked at the code that the compiler produced, and there was the outer DO loop structure but the 10 pages of complex code within the DO loop had evaporated! ...
    (microsoft.public.vb.enterprise)