RE: [vhdl-200x] RFC: std.textio.OUTPUT ?

> -----Original Message-----
> From: owner-vhdl-200x@eda.org [mailto:owner-vhdl-200x@eda.org] On
> Behalf Of Brent Hayhoe
> Sent: 22 January 2013 23:48
> To: vhdl-200x@eda.org
> Subject: Re: [vhdl-200x] RFC: std.textio.OUTPUT ?
> 
> 
> Whilst I agree that the current implementation is what users want and
> that we are not aiming to change that, it is important that the LRM
> reflect the reality.
> 
[MJT] 
Agreed

> Both "STD_OUTPUT" and "STD_INPUT" are special kinds of I/O in that they
> are a shared resource and have all kinds of accesses before the actual
> simulated code 'run' actually opens the resource for its own use.
[MJT] 

Maybe that's the difference - I *perceive* the operation differently :)

My mental model is of the transcript window displaying the "all kinds of accesses" *in addition* to a "tee" of a stream which carries the content of STD_OUTPUT.

> 
> As such, regardless of how this task is implemented within the
> simulator, any output to "STD_OUTPUT" will appear in the transcript as
> text appended to the existing output.
> 
> I fail to see how this is anything but "APPEND_MODE".
> 
[MJT] 
I agree, the transcript does operate in append mode.  My only difference is that I don't model the transcript window as *the* STD_OUTPUT, just one view on it, along with "other things".  And my mental model is also independent of simulator implementation - it's just my model :) It may be my mental model is in "error"?

However, I'm not one to get "all het up about it" - I can adjust my mental model to the transcript window "being the stream"...  In which case, if the simulator creates a logfile (as well as the transcript display), should that be a new one per simulation restart, or a continuously appended one?

Cheers,
Martin

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.