lynx-dev
[Top][All Lists]
Advanced
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: LYNX-DEV SHOW_CURSOR

From: Laura Eaves
Subject: Re: LYNX-DEV SHOW_CURSOR
Date: Sat, 10 May 1997 15:30:02 -0400 (EDT) 
> Date: Sat, 10 May 1997 13:06:53 -0500 (EST)
> From: Foteos Macrides <address@hidden>
>...
>       The behavior of positioning the cursor at the current link was
> added as an option years ago for the benefit of users whose terminals
> do not support what FANCY_CURSES designated at that time:  support for
> enough character attributes to (1) highlight all links, and (2) make
> the current link recognizable through a distinguishable form of
> highlighting (e.g., bold for links, and a change to reverse when the
> link is current).  It was also intended as an aid for braille.  The
> command line -show_cursor was added for the situation when the image
> was compiled for "fancy" curses, but the user's terminal (or the
> termcap/lib entry for it) didn't "cooperate", or was coupled to braille.

... and that assumes the default is FALSE.  (I might add,
braille/speech/screen magnification software all benefit from being able to
track the cursor to the current link.)
Thank you for this background information.
What you're saying is that if lynx is compiled with fancy curses or slang,
you want the cursor to be hidden (SHOW_CURSER:FALSE) by default?
(If it isn't compiled that way, it behaves as if the default is TRUE.)
If an ISP compiles a new version with slang or ncurses and hadn't done so
previously, the behavior would change for unsuspecting users.
Obviously there are some people who don't want to see the cursor along with
their highlighted text.
So you don't want to remove the option altogether.
But my second question still stands: Would it be agreeable to make the default
TRUE for future releases?

> <HTML>
> <NOTE CLASS="warning">
> Much of what Laura posted recently concerning cursor positioning, how
> LINKS_ARE_NUMBERED versus NUMBER_AS_ARROWS modes behave in various
> situations, how they affect the 'l'ist output or "Follow link (or
> page) number:" features, what's intended to be communicated about
> them in the online 'h'elp, etc., are voluminous, self-assured
> expressions of misunderstanding, misinterpretation, wrong guesses,
> or abject ignorance.  I don't know if anyone has the spare time to
> address all that point-by-point, but I caution others to apply grains
> of salt when reading those messages.
> </NOTE>
> </HTML>

I tried my assumptions on ac-0.23 and when NUMBERS_AS_ARROWS is set,
the numbers indeed act like arrows, thus making the number-based commands
(123, 123g, 123p) inaccessible in this mode.
Also, 123 (no g) still behaves like goto URL, which is backward compatible with
older lynxes, but not with the current documentation.
If you were planning on changing the behavior, of course, that's another matter.
(If a link is invisible, you can't position the current link on it before
following it -- but if you eliminate invisible links altogether, which it
seems you have in one of your modes, then it wouldn't matter if you changed
the semantics of the 123 command.  In fact, it would probably be an
improvement.)

I don't remember saying anything about the list command.
The fixes I sent for Lynx_users_guide.html had to do with a typo +
the claim that ^A and ^E don't work in NUMBERS_AS_ARROWS mode.
(I tried them and they do work.)

Anyway, I was just reviewing the latest diffs.
If I made any errors, please excuse me.
--le

;
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.
;
reply via email to
[Prev in Thread] Current Thread [Next in Thread]