Re: brltty and simultaneous grade 2 translation

["John J. Boyer" <director@chpi.org>]

Dave,
I prefer a pointer and a length also. I think it would be a good idea to
have a global variable containing the length of the Braille display, so it
wouldn't have to be passed. I can pass back an aray of the position in the
input buffer of each character in the output buffer, but I feel a bit
uncomfortable about doing cut and paste in Grade 2 because Grade 2 isn't 1
for 1. I would probably turn off Grade 2 before doing cut and paste.
Perhaps it would be better if we take this discussion of-list and work out
the programming details and the format of the translation tables between us.
John

Computers to Help People, Inc.
http://www.chpi.org
825 East Johnson; Madison, WI 53703


----- Original Message -----
From: "Dave Mielke" <dave@mielke.cc>
To: <blinux-list@redhat.com>
Sent: Monday, November 26, 2001 12:45
Subject: Re: brltty and simultaneous grade 2 translation


> [quoted lines by John J. Boyer on November 26, 2001, at 11:39]
>
> >I am assuming that brltty would pass a pointer to a position in a string.
>
> I have a thing against null-terminated strings, and would prefer to pass
in an
> address and a length.
>
> >The Grade 2 translation routine would then translate until it reached the
> >end of the string or until it had enough to fill the Braille display.
>
> Yes, although that'd mean that an additional parameter, i.e. an integer
set to
> the braille window width, would also need to be passed to the translation
> routine.
>
> >If it
> >found that the last word would not fit on the display it would backtrack
to
> >the beginning of that word and just leave a few blanks at the end of the
> >display.
>
> That's definitely the simplest, most predictable, and least error prone
way to
> do it.
>
> >In any case, it would pass back a pointer to the place in the input
> >string where translation should resume, or perhaps an integer count of
the
> >number of characters in the input string that it translated.
>
> I'd prefer an integer count. This count should include any trailing
blanks.
>
> >If the pointer into the input string indicating where the translation was
to
> >begin happened to be in the middle of a word the translator would just
> >process the rest of the word, whatever that might turn out to be, instead
of
> >trying to find the beginning of the word and probably producing some
nasty
> >bugs.
>
> That's certainly a good starting point. If, with it working that way,
there are
> no complaints, then it'll prove itself to be an adequate solution.
>
> >I think that cut and paste should simply not be attempted if the Grade 2
> >translation is on. Brltty should refuse to carry it out and should
display a
> >message asking the user to turn grade 2 translation off.
>
> I'd sure like it to work, as I dislike denying functionality based on
> implementation restrictions. It could easily be made to work if the
translation
> routine were to pass back an offset into the input string for each
character in
> the output buffer.
>
> --
> Dave Mielke           | 2213 Fox Crescent | I believe that the Bible is
the
> Phone: 1-613-726-0014 | Ottawa, Ontario   | Word of God. Please contact me
> EMail: dave@mielke.cc | Canada  K2A 1H7   | if you're concerned about
Hell.
>
>
>
> _______________________________________________
> Blinux-list mailing list
> Blinux-list@redhat.com
> https://listman.redhat.com/mailman/listinfo/blinux-list
>