Re: Current thoughts on pw_gecos field Lassi Kortela 19 Aug 2019 19:58 UTC

>     For porting full names to Unix?
>
> No, but so that you can call parse-full-name without worrying about
> whether you are on Windows or not.  You want a full name of "John Jones,
> Jr." to return ("John Jones Jr.") rather than ("John Jones" "Jr.").

I don't think that goal makes sense. There are many different ways to
check whether one is on Windows or Unix. If someone goes to the trouble
of implementing parse-full-name, it won't be much extra effort to do:

   (define (parse-full-name info)
     (if (not unix?)
         (user-info:full-name info)
         (parse-from-gecos (user-info:name info)
                           (user-info:full-name info))))

That `if` could even be a `case`, with unixes known to have relinquished
commas and ampersands also just returning the name. Which makes a good
case for why the parsing should be hidden in the SRFI implementation.

For the computer archaeologists who want to work with finger, there can
be a separate (user-info:gecos-field) procedure that doesn't need to be
mandated by the spec.