M-n versus hidden threads

M-n versus hidden threads

[John H Palmieri]

This is with Quassia Gnus 0.34, GNU Emacs 20.2.1.

If I am looking at a summary buffer in which threads are hidden
(either because gnus-thread-hide-subtree is t, or because I've hit
`T H'), then hitting M-n (gnus-summary-next-unread-subject) doesn't
always do what I expect: if the point is on a line which is the start
of a (hidden) thread, then hitting M-n doesn't move the point.  More
precisely: the documentation for gnus-summary-next-unread-subject
says:

  "Go to next N'th unread summary line."

If the thread has k unread article in it, and if I give an argument k
or larger to M-n, then the point moves the appropriate number of
lines.  If I give an argument less than k, the point doesn't move.  Is
this the right behavior?  Other options: (1) M-n should count
(visible) lines in the buffer, so M-n should always move the point.
If you want to move within the thread, show the thread first, and then
use M-n to move around.  (2) M-n should count invisible lines in the
buffer, so hitting M-n four times should be like hitting C-u M-n.
Hitting M-n SPC should show the thread and select the second article
in the thread, not the first one.

I'm not sure which of these is intended, and I'm not sure which of
these is best.  Comments?  Suggestions?  Patches?  The answers to all
my questions?

-- 
John H. Palmieri, Visiting Assistant Professor
Department of Mathematics, University of Notre Dame
Email: palmieri@member.ams.org
URL: http://www.nd.edu/~jpalmier/

Re: M-n versus hidden threads

[Lars Magne Ingebrigtsen]

John H Palmieri <John.H.Palmieri.2@nd.edu> writes:

> If I am looking at a summary buffer in which threads are hidden
> (either because gnus-thread-hide-subtree is t, or because I've hit
> `T H'), then hitting M-n (gnus-summary-next-unread-subject) doesn't
> always do what I expect: if the point is on a line which is the start
> of a (hidden) thread, then hitting M-n doesn't move the point.

Do you have `gnus-summary-check-current' set?


-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: M-n versus hidden threads

[John H Palmieri]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> John H Palmieri <John.H.Palmieri.2@nd.edu> writes:
> 
> > If I am looking at a summary buffer in which threads are hidden
> > (either because gnus-thread-hide-subtree is t, or because I've hit
> > `T H'), then hitting M-n (gnus-summary-next-unread-subject) doesn't
> > always do what I expect: if the point is on a line which is the start
> > of a (hidden) thread, then hitting M-n doesn't move the point.
> 
> Do you have `gnus-summary-check-current' set?

Ordinarily, I don't have gnus-summary-check-current set; I just set it
to t, and still got the same behavior.  If the first article in the
thread is read, then hitting M-n (some apparently random number of
times--once isn't usually good enough, twice sometimes is, and three
times usually is) and then SPC selects the first unread article in the
thread.  Without hitting SPC, though, no matter how many times I hit
M-n, I can't move the cursor off of the first line of the thread.

(Sorry for the delay in responding; it was "spring" break here, so I
wasn't answering email.)

-- 
John H. Palmieri, Visiting Assistant Professor
Department of Mathematics, University of Notre Dame
Email: palmieri@member.ams.org
URL: http://www.nd.edu/~jpalmier/

Re: M-n versus hidden threads

[Lars Magne Ingebrigtsen]

John H Palmieri <John.H.Palmieri.2@nd.edu> writes:

> Without hitting SPC, though, no matter how many times I hit
> M-n, I can't move the cursor off of the first line of the thread.

Yup.  Fix in Gnus v5.6.3.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen