Bug Report: TAB on empty headline cycles through the wrong levels

Bug Report: TAB on empty headline cycles through the wrong levels

[Ryan C. Thompson]

I have discovered a possible bug with the new 6.33 feature of cycling 
empty headline levels with TAB. IT seems that org will always assume 
that the initial level of the headline is the "base" level. Pressing TAB 
once will always go one level deeper than that, to the "child" level, 
TAB a second time will go one level shallower than the base level, and 
so on.

the problem is that I expect the "base" level to be based on the 
previous headline's level, not on the initial level of the empty 
headline itself. The problem would not usually show itself, because if 
you press meta+RET  at the end of a headline, you get a new empty one at 
the same level as the previous headline. However, if you press meta+RET 
at the *beginning* of a headline (before the stars), you get a new empty 
headline at the same level as the *next* headline, which might not be 
the same level as the previous headline.
Here's a simple test case:

* Head1
** Head2


Put the point at the very beginning of the second heading, and press 
meta+RET, then TAB. You should get this:


* Head1
*** [point is here]
** Head2

The following is what I would *expect* to get:

* Head1
* [point is here]
** Head2


So basically, the base level for depth cycling should be determined from 
the previous headline, not the new blank headline. At least, that's what 
I expected, and I can't think of any reason you would want to put a 
level-3 headline under a level-1 headline.

-Ryan

Re: Bug Report: TAB on empty headline cycles through the wrong levels

[Carsten Dominik]

Hi Ryan,

here I do not agree with your assessment.
Putting the cursor at the beginning of a headline and pressing M-RET  
is a specific exception of the normal workings of the command, and it  
create a sibling above the entry.

If you want to make use of the indentation cycling, I think you should  
go to the end of the subtree above by pressing LEFT, and then use M- 
RET annd then TAB to get your level right.

- Carsten

On Nov 27, 2009, at 11:36 PM, Ryan C. Thompson wrote:

> I have discovered a possible bug with the new 6.33 feature of  
> cycling empty headline levels with TAB. IT seems that org will  
> always assume that the initial level of the headline is the "base"  
> level. Pressing TAB once will always go one level deeper than that,  
> to the "child" level, TAB a second time will go one level shallower  
> than the base level, and so on.
>
> the problem is that I expect the "base" level to be based on the  
> previous headline's level, not on the initial level of the empty  
> headline itself. The problem would not usually show itself, because  
> if you press meta+RET  at the end of a headline, you get a new empty  
> one at the same level as the previous headline. However, if you  
> press meta+RET at the *beginning* of a headline (before the stars),  
> you get a new empty headline at the same level as the *next*  
> headline, which might not be the same level as the previous headline.
> Here's a simple test case:
>
> * Head1
> ** Head2
>
>
> Put the point at the very beginning of the second heading, and press  
> meta+RET, then TAB. You should get this:
>
>
> * Head1
> *** [point is here]
> ** Head2
>
> The following is what I would *expect* to get:
>
> * Head1
> * [point is here]
> ** Head2
>
>
> So basically, the base level for depth cycling should be determined  
> from the previous headline, not the new blank headline. At least,  
> that's what I expected, and I can't think of any reason you would  
> want to put a level-3 headline under a level-1 headline.
>
> -Ryan
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Please use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode

- Carsten