Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:

> Also, shouldn't the final (delete-indentation ARG) be in the "else" part
> of the `if'?

It was, at least on my disk.  I did not check the patch.

>> +(ert-deftest test-org-delete-indentation ()
>> +  "Test M-^ (`org-delete-indentation') specification."
>
> I suggest to omit binding in the description. That's one thing less we
> have to keep up-to-date if it ever changes.

OK.

The attached patch is updated and seems pretty good.  It also handles tags
and entities according to org-auto-align-tags.

The second patch (0003) adds support for tables in the sense that

| a | c |
| b | d |

can be made into

| a b | c |
|     | d |

using M-^.

While this is pretty nice since it's hidden away in a uncommon key, one
thing that bothers me about it is that there's no natural key to go back
to the first situation.

While RET could be made to move content between, it may be too annoying.
Personally. I never use RET in tables for movement commands so it would
not bother me.  I wrote a preliminary patch for this, but there's many
open questions on supposed behavior IMO.

I can work more on M-^ and tables if desirable (primarily it should be
moved to org-table, I guess).  I can also work more on RET in tables, if
desirable.  Another "quirk" that I noticed is that RET works differently
when next row is a hline (insert new row) and a row (move to next row).
This is completely illogical to me.

—Rasmus

-- 
. . . The proofs are technical in nature and provides no real understanding