Re: Heads-up: test failures

[Nick Dokos <ndokos@gmail.com>]

Colin Baxter <m43cap@yandex.com> writes:

> On Mon, Feb 13 2017, Nick Dokos wrote:
>
>> Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:
>>
>>> What is the failure in each case?
>>
> ---- snip ----
>
> Ran 711 tests, 711 results as expected (2017-02-14 07:58:32+0000)
> 8 expected failures
>
> org-mode release_9.0.5-282-g2e32709
> emacs-25.1.1; 3.2.0-4-686-pae #1 SMP Debian 3.2.84-1 i686 GNU/Linux
>
>

It might be that emacs 26 shows the failures (both Marco Wahl and I are
running that, but Colin Baxter runs emacs 25 and does not see them).

I simulated the move-down test by hand:

,----
|   ;; Preserve children visibility.
|   (org-test-with-temp-text "* Headline
| - item 1
|   - sub-item 1
|     sub-body 1
| - item 2
|   - sub-item 2
|     sub-body 2"
|     (let ((org-cycle-include-plain-lists t))
|       (search-forward "- sub-item 1")
|       (org-cycle)
|       (search-forward "- sub-item 2")
|       (org-cycle))
|     (search-backward "- item 1")
|     (org-move-item-down)
|     (search-forward "sub-body 1")
|     (should (org-invisible-p2))
|     (search-backward "sub-body 2")
|     (should (org-invisible-p2)))
`----

The   (search-forward "sub-body 1") ends up in a place where (org-invisible-p2) returns t,
but the (search-backward "sub-body 2") end up in a place where it returns nil (C-u C-x = shows
that it's on the RET, at the end of the sub-item 1 line).

Maybe something about such text properties got changed in recent emacs?

-- 
Nick