Frame navigation broken in terminal

Frame navigation broken in terminal

[Jose A. Ortega Ruiz]

Hi,

In yesterday's CVS (configured with --without-x and run with -q),
'other-frame' seems broken: i can create new frames (and the display
switchs to the new ones as they're created), but cannot switch
to old ones with other-frame (there's no error, but emacs stays in the
same frame, the last one created). 'select-frame-by-name' seems to
work, though.

(BTW, should I use report-emacs-bug for these kind of ongoing
development glitches?)

Cheers,
jao
-- 
"When you come to a fork in the road, take it"
 -Yogi Berra, baseball coach.

Re: Frame navigation broken in terminal

[Jose A. Ortega Ruiz]

"Jose A. Ortega Ruiz" <jao@gnu.org> writes:

> Hi,
>
> In yesterday's CVS (configured with --without-x and run with -q),
> 'other-frame' seems broken: i can create new frames (and the display
> switchs to the new ones as they're created), but cannot switch
> to old ones with other-frame (there's no error, but emacs stays in the
> same frame, the last one created). 'select-frame-by-name' seems to
> work, though.
>

A follow-up on this problem: the origin seems to be that
`frame-visible-p', which should return t for all frames in a tty,
returns t only for the selected frame, and as a consequence
`other-frame' has no effect. That is, if i evaluate

(frame-visible-p (next-frame (selected-frame)))

(with more than one frame in my tty session) i get nil, instead of t
(next-frame is correctly returning a frame value). Unfortunately,
frame-visible-p is a C function, and i don't know enough about the C
side of emacs to see what's wrong.

hth,
jao

Re: Frame navigation broken in terminal

[Stefan Monnier]

> In yesterday's CVS (configured with --without-x and run with -q),
> 'other-frame' seems broken: i can create new frames (and the display
> switchs to the new ones as they're created), but cannot switch
> to old ones with other-frame (there's no error, but emacs stays in the
> same frame, the last one created). 'select-frame-by-name' seems to
> work, though.

Should be fixed now,

> (BTW, should I use report-emacs-bug for these kind of ongoing
> development glitches?)

Yes, that would be fine,


        Stefan