Free `1' and `0' keys from the org-agenda interface ?

Free `1' and `0' keys from the org-agenda interface ?

[Bastien]

Hi all,

i'd like to have your opinion on this.

I tend to use my `org-agenda-custom-commands' in a strict order.  
I first check for a first view, then check for a second view, etc.  
It would be nice to be able to use `1' for the first view -- but
it's currently used to restrict the agenda view to the buffer.

My suggestion for a possible replacement for the `1' and `0' keys 
would be this:

 `<' : restrict to the buffer
`<<' : restrict to the region or subtree

Note: `<' could also be `@' since `C-c @' means select the current
subtree.

I think it's okay to have some kind of "incremental" restriction,
but I'm not sure about this.  And I'd definitely like to have the
[0-9] keys for my own agenda views :)

Did anyone ever meet the same need?

Thanks,

PS: I think today will be the craziest day in term of list traffic!

-- 
Bastien

Re: Free `1' and `0' keys from the org-agenda interface ?

[Adam Spiers]

Bastien (bzg@altern.org) wrote:
> Hi all,
> 
> i'd like to have your opinion on this.
> 
> I tend to use my `org-agenda-custom-commands' in a strict order.  
> I first check for a first view, then check for a second view, etc.  
> It would be nice to be able to use `1' for the first view -- but
> it's currently used to restrict the agenda view to the buffer.
> 
> My suggestion for a possible replacement for the `1' and `0' keys 
> would be this:
> 
>  `<' : restrict to the buffer
> `<<' : restrict to the region or subtree
> 
> Note: `<' could also be `@' since `C-c @' means select the current
> subtree.
> 
> I think it's okay to have some kind of "incremental" restriction,
> but I'm not sure about this.  And I'd definitely like to have the
> [0-9] keys for my own agenda views :)

Wow, what can I say?  I think those two mails arrived with 30 seconds
of each other - great minds think alike ;-)

I'm only just starting to use custom commands; however, your comment
about typically invoking custom commands in a strict order doesn't
surprise me.  I might well find my usage habits following the same
trait.

I'm curious about your suggestion of '<<' though - does that imply the
sub-keymap functionality I asked for already exists?

Re: Free `1' and `0' keys from the org-agenda interface ?

[Bastien]

Adam Spiers <orgmode@adamspiers.org> writes:

> I'm curious about your suggestion of '<<' though - does that imply the
> sub-keymap functionality I asked for already exists?

I first thought yes, since it looks like the first `<' keystroke is
waiting for the next one (at least from the description I proposed).

But in fact there is no sub-keymap involved here: it's only about
cycling through three states: 

- not-restricted
- buffer-restricted
- subtree/region-resctricted

Then three keystrokes would turn off the restriction.

But calling a sub-keymap with `<' could also be useful:

`< <' : cycle through default restricted states (see above)
`< f' : prompt for an org-agenda-file to restrict to
`< s' : prompt for a subtree to restrict

Caveat: we should better not allow interactive restriction to a
property, since this is the job of the custom-command itself...

My 2.5 cents.

-- 
Bastien

Re: Free `1' and `0' keys from the org-agenda interface ?

[Adam Spiers]

On Tue, Oct 16, 2007 at 06:53:47PM +0100, Bastien wrote:
> Adam Spiers <orgmode@adamspiers.org> writes:
> 
> > I'm curious about your suggestion of '<<' though - does that imply the
> > sub-keymap functionality I asked for already exists?
> 
> I first thought yes, since it looks like the first `<' keystroke is
> waiting for the next one (at least from the description I proposed).
> 
> But in fact there is no sub-keymap involved here: it's only about
> cycling through three states: 
> 
> - not-restricted
> - buffer-restricted
> - subtree/region-resctricted
> 
> Then three keystrokes would turn off the restriction.
> 
> But calling a sub-keymap with `<' could also be useful:
> 
> `< <' : cycle through default restricted states (see above)

Hmm, that doubles the number of keypresses to achieve a given state,
which is a bit cumbersome.  But if it's customizable, who cares? :-)

Re: Free `1' and `0' keys from the org-agenda interface ?

[Bastien]

Adam Spiers <orgmode@adamspiers.org> writes:

>> `< <' : cycle through default restricted states (see above)
>
> Hmm, that doubles the number of keypresses to achieve a given state,
> which is a bit cumbersome.  But if it's customizable, who cares? :-)

I wouldn't mind having single keys to achieve this - it's just that I
thought `1' and `0' where not the best choice.

-- 
Bastien

Re: Free `1' and `0' keys from the org-agenda interface ?

[Carsten Dominik]

On Oct 16, 2007, at 23:26, Bastien wrote:

> Adam Spiers <orgmode@adamspiers.org> writes:
>
>>> `< <' : cycle through default restricted states (see above)
>>
>> Hmm, that doubles the number of keypresses to achieve a given state,
>> which is a bit cumbersome.  But if it's customizable, who cares? :-)
>
> I wouldn't mind having single keys to achieve this - it's just that I
> thought `1' and `0' where not the best choice.

Yes, I can see that the 0 and one would be useful in other 
circumstances.
And I don't *relly* think that a double "<" would be so bad, because
don't think restricting an agenda command to a region or subtree is
used frequently, is it?

- Carsten

Re: Free `1' and `0' keys from the org-agenda interface ?

[Adam Spiers]

On Wed, Oct 17, 2007 at 05:47:58PM +0200, Carsten Dominik wrote:
> On Oct 16, 2007, at 23:26, Bastien wrote:
> >Adam Spiers <orgmode@adamspiers.org> writes:
> >
> >>>`< <' : cycle through default restricted states (see above)
> >>
> >>Hmm, that doubles the number of keypresses to achieve a given state,
> >>which is a bit cumbersome.  But if it's customizable, who cares? :-)
> >
> >I wouldn't mind having single keys to achieve this - it's just that I
> >thought `1' and `0' where not the best choice.
> 
> Yes, I can see that the 0 and one would be useful in other 
> circumstances.
> And I don't *relly* think that a double "<" would be so bad, because
> don't think restricting an agenda command to a region or subtree is
> used frequently, is it?

I have recently started using it quite a lot, e.g. to narrow to my
work or personal TODOs, which are both in files of their own.

But if sub-keymaps were to be implemented, this would be customizable
anyway, so we wouldn't need to worry about it.

Re: Free `1' and `0' keys from the org-agenda interface ?

[Andrew J. Korty]

Bastien <bzg@altern.org> writes:

> I tend to use my `org-agenda-custom-commands' in a strict order.  
> I first check for a first view, then check for a second view, etc.  

Why not just display all those views, in order, with one custom command?
For example, when I'm at home, I run C-c a h:

 ("h" "home contexts"
  ((tags-todo "+@home+@computer/TODO" nil)
   (tags-todo "+@home-@computer/TODO" nil)
   (tags-todo "+@computer-@home-@office/TODO" nil)
   (tags-todo "@phone/TODO" nil)
   (tags-todo "@any/TODO" nil))

-- 
ajk

Re: Re: Free `1' and `0' keys from the org-agenda interface ?

[Bastien]

ajk@iu.edu (Andrew J. Korty) writes:

> Bastien <bzg@altern.org> writes:
>
>> I tend to use my `org-agenda-custom-commands' in a strict order.  
>> I first check for a first view, then check for a second view, etc.  
>
> Why not just display all those views, in order, with one custom
> command?

Yes, good idea. But I still think it would be nice to be able to bind
the [0-9] keys. One advantage of these keys is that they are very easy
to memorize, and this might help when trying to go from one agenda view
to another.

-- 
Bastien