Re: Is there really any interest in a patch to allow cvs 1.11.6

[Larry Jones]

Paul Edwards writes:
> 
> I think VMS has some scheme where even numbers are
> errors, odd numbers are OK, so although I don't recall
> seeing it implemented that way, I would expect to see
> 0 get translated into 1 on the way out.

Exactly.  VMS has highly structured 32-bit "Condition Value"s.  The top
four bits are used to control processing (for example, the low-order bit
indicates whether the corresponding message has been displayed to the
user yet or not), the next 12 bits identify the facility reporting the
condition (the top bit is set for customer-definined facilities and
clear for system-defined facilities), the next 13 bits identify the
specific condition (conditions with the top bit clear are global,
conditions with the top bit set are facility-specific), and the last
three bits are a severity code with the very last bit being the
"success" bit.  The defined severities are:

        0 - Warning
        1 - Success
        2 - Error
        3 - Informational
        4 - Fatal

Exit statuses 0 and 1 are mapped since the normal VMS interpretation of
those values is exactly opposite the normal conventions, other values
are used as-is.  I don't recall exactly what statuses they're mapped to,
but 0 is mapped to something with a severity of 1 and 1 is mapped to
something with a severity of 2.  Likewise, I don't remember exactly how
EXIT_SUCCESS and EXIT_FAILURE are defined, but they have the obvious
severities.

CVS has returned just success or failure since 1996, so I don't see any
reason to change it now.

-Larry Jones

What this games needs are negotiated settlements. -- Calvin