bug-texinfo
[Top][All Lists]
Advanced

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

Re: scrolling behaviour of newer info has changed


From: Gavin Smith
Subject: Re: scrolling behaviour of newer info has changed
Date: Fri, 7 Aug 2015 11:40:09 +0100

On 7 August 2015 at 10:28, Benno Schulenberg <address@hidden> wrote:
> When with the current (6.0) stand-alone info I just run 'info',
> am shown the top node, and type PageUp, it does not beep or
> say anything.  An older info (4.13) will beep and say there is
> no Prev or Up node here.  Also when reaching the end of the
> top node, it will not beep.

I see the same happens when displaying man pages. I think this is
because it is not trying to go forward or backward in the node
structure, which is what happens when scroll-behaviour=Continuous for
regular Info files. Lack of an error message here seems fairly
harmless although I suppose it should be added back.

> Further, when pressing and holding PageDown, newer info
> will keep the cursor on the top line when reaching the end
> of the document.  Older info would put it just beyond the
> last line.  But then, when pressing and holding PageUp,
> older info will put the cursor back on the top line, but
> newer info will put and keep it on the bottom line.  Is
> this emacsy behaviour intentional?
>
> Even if so, I don't like it.  I expect the cursor to go to the
> top when I press PageUp, so that a Tab will then bring me to
> the first reference or menu item on that page.

I can't remember when or how this changed. I think I like the newer
behaviour better. Consistency with Emacs is an argument. I'd expect
people to be using PageUp to read earlier in the file, not to move the
cursor to the beginning of the node. I'd expect them to be using Home
or "b" for that. My perception is that it is more jarring both for the
cursor to move and an error message to be displayed. Maybe that's
because the error message is a message of failure, leaving the user
wondering how to recover from this failure, and manage to do what they
intended to do. It's more relaxing to leave the user-visible state of
the program exactly as it was before the failure, except for the error
message; otherwise, the user wonders how to deal with the sudden
cursor movement as well as the error message.



reply via email to

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