Now as I read it a few times, I can make sense out of it.  Just a long day
yersterday.  Although it can be a little bit more clearly explained for
challenged people like myself:-)

I guess maybe reverse thinking...

***
If `PrintError' is also on, then it will be skipped
since the 'die' will print the message, unless '__DIE__' handler
is defined in which case 'PrintError' is done before 'RaiseError'.
***

Sounds like the same thing, but when reading for the first time can make
more sense, since in most cases '__DIE__' is probably not defined, so that
condition is more commone and should come first.  Anyone??

Ilya Sterin





-----Original Message-----
From: Tim Bunce
To: Sterin, Ilya
Cc: 'Tim Bunce '; '''Neil Lunn ' ' '; ''''[EMAIL PROTECTED]' ' ' '
Sent: 05/09/2001 6:00 AM
Subject: Re: executing atomic transactions in DBI

On Tue, May 08, 2001 at 11:06:58AM -0600, Sterin, Ilya wrote:
> In the docs shouldn't this be changed...
> 
> ***
>        If `PrintError' is also on, then the 
>       `PrintError' is done before the `RaiseError' 
> ***
> 
> to 
> 
> ***
> 
>        If `PrintError' is also on, then the 
>       `PrintError' is done before the `RaiseError'
>        though 'RaiseError' will override 'PrintError' 
>        and the error will be printed as die() rather than
>        warn().
> ***
> 
> Or something like that...

Err, why, exactly?

Tim.

Reply via email to