Re: Trouble with M-x sunrise command on debian squeeze (solved)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[brownh]

Sorry for the question that arose as a result of my stumbling (when
you reach 76, see how you do! ;-)). It was an emacs line originally
there but subsquently removed.

Haines Brown

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

Hello Haines,

Thanks for using Sunrise. In order to avoid this kind of set up
problems you may want to try out the new ELPA installation (see
http://www.emacswiki.org/emacs/Sunrise_Commander#toc11 for more
details)

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

On 23 Paź, 03:13, brownh <bro...@historicalMaterialism.info> wrote:
(...)
> If you are implying that all menu bar commands can be run from the
> keyboard, then I have no need for it. But Sunrise-Commander is new to
> me and I don't know yet if that is true.

Yes, that's exactly the case: Sunrise (as anything else in emacs, I
guess) is completely keyboard-driven. In order to help memorizing a
useful subset of the bindings, you may be interested in the buttons
extension:

    http://www.emacswiki.org/emacs/Sunrise_Commander#toc2

to have a look at all the available bindings just press h inside any
of the panes.

(...)
> > What there is is only the menus that additional extensions (like
> > sunrise-x-modeline or sunrise-x-tabs) provide, but nothing else, I'm
> > afraid. Or is that what you're asking about? I've just checked on
(...)
> Are you saying that while the Sunrise menuing is not implemented, it
> can displayed by means of the make-frame line above? I'm also running
(...)

No, all I'm saying is that some of the extensions do display their
menus when menu-bar-mode and Sunrise are both active, but they are
not the "real deal",  they contain only specialized functions from
their specific extensions.

(...)
> I am runing the lisp stanzas in this order: a) require
> sunrise-commander and sunrisex-tree, b) the make-frame line, c) the
> add-hook lines. Is the order important? In any case, I get odd
> results:

Yes, the order is important: the "require" and the "add-hook" stanzas
are intended to be run only once at the beginning of the session, you
use the "make-frame" stanza whenever you want to open a new frame.

> First, Sunrise commander opens only one pane in its frame, which is
> a list view rather than tree view. Its buffers are:
(...)

That's weird, there should be two panes side by side. Can you please
do "M-x toggle-debug-on-error" and send me the trace that appears (if
any) when launching the new frame? Looks like the logic for bringing
up the panes is having some kind of problem.

> Second, I discovered that there are two sessions of emacs
> running. Below the one I just described there is another emacs session
> with the buffers:
(...)

That looks like two frames of the same emacs instance. What happens
when you do C-x C-c in one of them? Does the other get closed too?

(You are aware of the peculiar way the terms "frame" and "window" are
used in emacs, aren't you?)

(...)
> Now, If I do M-x sunrise _twice_ for the top session, it displays
> sunrise commander with two panes and the left one is the tree view

That's right - the sunrise function works like a switch: it turns on
and off the panes.

> (but nothing else except the scratch buffer across bottom). If I try

That's correct too: that is all there is - the bottom window is used
in Sunrise for "quick viewing" the contents of files (you press o and
the file appears underneath, if you select another file and press o
again, the previous file is closed and the new one is shown, and so
on), as a container for the command-line (after pressing C-c t in any
of the panes) and it's also used by the buttons extension to display
the command buttons.

You may get rid of the bottom window with the popviewer extension:

    http://www.emacswiki.org/emacs/Sunrise_Commander#toc5

but be aware that it is not compatible with the buttons extension (it
makes no sense to have both installed simultaneously).

> this on the bottom emacs session, I get same result, but it just takes
> one M-x sunrise. Just to be clear, it seems that the sunrise panes
> occupy the two emacs frames.

Yes, you're right again: there is only one instance of the panes and
they can be displayed in only one frame at a time -- when you open
them in one frame, they're automatically closed in the other. If you
need to keep track of multiple directories in any of the panes you
can use the tabs extension:

    http://www.emacswiki.org/emacs/Sunrise_Commander#toc7

> Third is a point that probably is irrelevant. I find that in Options,
> if I hide the toolbar and then Save options, the change does not
> stick. I can only hide the toolbar for the current session.

Maybe you need to add "(tool-bar-mode -1)" to you .emacs file?

>
> Something very strange is going on. I hope my complicated description
> is not giving you a headache. ;-)

Emacs *is* something very strange :) No worries, the whole community
is here to help (or try at least).

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[Lars Magne Ingebrigtsen]

brownh <brownh@historicalMaterialism.info> writes:

>> That's weird, there should be two panes side by side. Can you please
>> do "M-x toggle-debug-on-error" 
>
> Sorry, there's no stack trace. I normally keep a (setq
> toggle-debug-on-error t) in my emacs init files, and nothing has shown
> up.

There is no `toggle-debug-on-error' variable -- it's called
`debug-on-error'.

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

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

On 24 Paź, 00:28, brownh <bro...@historicalMaterialism.info> wrote:
(...)
> Sorry, there's no stack trace. I normally keep a (setq
> toggle-debug-on-error t) in my emacs init files, and nothing has shown
> up.
(...)

Does the same happen if you simply do "M-x sunrise" (i.e. without
launching a separate emacs frame before?)

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

On 24 Paź, 22:21, brownh <bro...@historicalMaterialism.info> wrote:
(...)
> Sorry, I sent my reply to your rather than to the group. In short,
> there is no stack trace whether I stand Sundrise Commander with M-x
> sunrise command or with Lisp code in .emacs.

This is starting to look like a tough one.  I've  just tested the
same expression: (progn (select-frame (make-frame)) (sunrise)) on
emacs 22 and 24 on linux and on emacs 23 on windows 2k - it works
exactly the same everywhere (I am evaluating it by doing M-x  and
then typing in the expression - are you doing the same?)

Are you using some kind of tiling WM? (just shooting in the dark)

> However, there is a message in the message buffer in both cases, which
> may or may not be relevant:
>
> Starting Sunrise Commander...
> Loading hl-line...done
> Omitting...
> Omitted 6 lines.
> Omitting...
(...)

No, these are all normal messages.

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

Hello Haines,

Your  .emacs-test  file  was exactly what I needed to understand your
problem. Try adding this:

    (setq inhibit-startup-screen t)

anywhere in your .emacs file. That should do the trick.

As  for  the  error in the tree view, it looks indeed like a bug, but
unfortunately I don't seem able to replicate  it.  Could  you  please
play  a  bit  more with it and try to find a method to make it happen
reliably? (clicking on some specific point in the  pane,  opening  or
closing  some specific file, whatever). I've been using this stuff in
my day-to-day work since I wrote it  without  any  glitches,  but  my
usage  pattern  is  probably a whole lot different from yours (e.g. I
hardly ever use the mouse in the tree view).

Thanks a lot for helping improving Sunrise.

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

On 25 Paź, 13:18, brownh <bro...@historicalMaterialism.info> wrote:
(...)
> to work consistently. For example, after issuing my M-x sunrise twice,
> and things look as they should, when I open folders on the tree, the
> right list view window is unaffected and continues to show the files
> and directories of my home directory as a list view. Despite my
> multiple M-x sunrise command, the buffer list looks normal.
(...)

That is the way the tree view works -- it neither is an explorer nor
pretends to be one. Of course this doesn't mean it's not possible to
do what you want: just modify a bit the snippet your already have:

    (add-hook 'sr-start-hook
              '(lambda ()
                 (let ((selected sr-selected-window))
                   (sr-select-window 'left)
                   (unless (eq major-mode 'sr-tree-mode)
                     (sr-tree-view)
                     (unless sr-synchronized (sr-sync)))
                   (sr-select-window selected))))

but the philosophy behind Sunrise is actually quite different, so if
what you're after is an explorer-style FM, be aware that Sunrise may
disappoint you after some time. What I mean by the philosophy behind
Sunrise is best described here:

    http://www.softpanorama.org/OFM/index.shtml

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)

Re: Trouble with M-x sunrise command on debian squeeze (solved)

[José A. Romero L.]

On 26 Paź, 15:26, brownh <bro...@historicalMaterialism.info> wrote:
(...)
> Yes, that did the job for some reason. While Sunrise-Commander does
(...)

I suppose it's because the way the splash screen works: it waits
until emacs is completely initiated, and then pops up in the buffer
that has focus, which in your case was one of the Sunrise panes.

(...)
> operations requiring configuration, but if problems arise, I'll get
> back to you. Otherwise, I assume things will proceed as expected.
(...)

Sure, any time.

Cheers,
--
José A. Romero L.
escherdragon at gmail
"We who cut mere stones must always be envisioning cathedrals."
(Quarry worker's creed)