RE: UV write fatal error

[Glenn Herbert]

Ok.  So there WAS a bscan done on the T25 file prior to attempting the 
write.  Since this is done from within BASIC, my guess is that the active 
scan is causing some type of error.  Try this;  after the scan has verified 
that the data item does NOT exist, close and reopen the data file, THEN 
attempt your write (yes I know that this may invalidate the notion that the 
record does not exist).  This would test to see if the active scan of the 
T25 file is interfering with the low level write, as the close of the data 
file would  automatically terminate the scan (there is no other way from 
within BASIC).   I wrote the code for bscan back in 1992 so I don't recall 
all different interactions involved, but it could be a long-standing bug 
that never appeared or not reported.  In any case, if you could try that, 
I'd be interested in the results.

At 12:11 PM 02/17/2004, you wrote:
No, the value does not get written to the type 25 file. The data is not
pre-existing. The BSCAN is done to verify this fact before attempting
the write.
Thanks.

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Glenn Herbert
Sent: Tuesday, February 17, 2004 9:00 AM
To: [EMAIL PROTECTED]
Subject: RE: UV write fatal error
Does the data value actually get inserted into the type 25 file if you
check after the write fails?  Also, is the data pre-existing within the
type 25 file and this is an update to that record?  Sounds almost like a
memory allocation or rotating file pool problem.
The on error branch (actually ANY branch) would only be followed if the
run-machine wasn't crashing underneath you, which, if you are being
dropped back to the command prompt, is happening.
At 11:33 AM 02/17/2004, you wrote:
>
>
>-----Original Message-----
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
>On Behalf Of Ray Wurlod
>Sent: Friday, February 13, 2004 12:13 PM
>To: [EMAIL PROTECTED]
>Subject: RE: UV write fatal error
>
>Permissions?  All the preceding steps (Open, BSCAN) can be done without
>write permission. Write, needless to add, cannot.
>
> > >-----Original Message-----
> > >From: Jimi DeRouen [mailto:[EMAIL PROTECTED]
> > >
> > >I am attempting to write a record to a type 25 file and apparently
> > >getting a fatal error. The platform is UV 9.5.1.1b Pick flavor on
> > >W2K
>
> > >SP4. I have constructed the WRITE statement to include all possible
> > >(I
> > >think) branching, yet the program goes to the command prompt when
> > >the
>
> > >WRITE is executed without following any of the branches. The WRITE
> > >statement is as follows:
> > >
> > [SNIP]
>
>I have no problem writing other values. It is only this particular data
>value that seems to produce the error. Why wouldn't the program follow
>the on error branch?
>--
>u2-users mailing list
>[EMAIL PROTECTED]
>http://www.oliver.com/mailman/listinfo/u2-users
--
u2-users mailing list
[EMAIL PROTECTED]
http://www.oliver.com/mailman/listinfo/u2-users
--
u2-users mailing list
[EMAIL PROTECTED]
http://www.oliver.com/mailman/listinfo/u2-users
--
u2-users mailing list
[EMAIL PROTECTED]
http://www.oliver.com/mailman/listinfo/u2-users