[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: <OK> Re: [Groff] Escape characters in -man output

From: M Bianchi
Subject: Re: <OK> Re: [Groff] Escape characters in -man output
Date: Fri, 8 Jul 2005 09:43:35 -0400

Forgive me, but I'm going to toss by two-bytes into this discussion.

It seems to me that this thread has been largely conducted in the "solution
space", where this and that solution is thrown in and reactions are generated.
It might be better to discuss the wide-character and escape sequence problems
in the "problem space" where the attributes of ideal (or good) solutions are
discussed and a set of requirements for the implementations are developed.

I repeatedly find that when I am thrashing through the code, coming up with one
hacked solution after another, only to discover that it creates another
problem, that I'm not thinking through the problem clearly enough.  Stepping
back and coming up with a set of needs to be addressed and the attributes that
the solution needs to have, often in very abstract terms, usually leads to a
much better result.

For example, there may be incompatibilities between the (often) contrasting
goals of:

        preserving current functionality, including the "bugs" that are so long
        standing that they almost seem features

        accommodating the wide and varied character sets of the world's many

To my mind, it would not be unreasonable to say that a new groff architecture,
"giroff" (for Gnu International Run OFF), that takes input that looks very much
like groff but doesn't necessarily use all the pre-and post-filters of groff
COULD be a good way to go.  But I think it is too early to make that decision.
However, the unspoken idea that the same architecture has to work under a very
different world view is also premature.

Bottom line suggestion:  Try to come up with a statement of the problems and
the desirable characteristics of the solutions.  Then look to see what needs to
be done to achieve those goals.

Aside: I remember how thrilled I was when I discovered that the old "all macro,
diversion, string, etc. names are two characters" had been removed!  I thank
the sane person who decided that was the way to go, and the clean extension
that made it fit into the existing architecture with a minimum of fuss.

 Mike Bianchi
 Foveal Systems

 973 822-2085   call to arrange Fax


reply via email to

[Prev in Thread] Current Thread [Next in Thread]