Re: Some testing issues

[Stephen Berman]

On Sat, 08 Jul 2017 09:18:08 +0200 martin rudalics <address@hidden> wrote:

>> And lastly, in one case, the test succeeds
>> without the set-window-buffer call when run interactively, but fails
>> without it when run in batch mode from the shell.  Any idea what's going
>> on here?  (If anyone wants to take a closer look, comments in the test
>> file point out the problematic cases.)
>
> The interactive call might succeed because command_loop_1 makes the
> selected window's buffer current in
>
>       set_buffer_internal (XBUFFER (XWINDOW (selected_window)->contents));
>
> Supposedly, testing ‘recenter’ doesn't make much sense without a frame.

True, though when I step through the code in the interactive test, the
condition (pos-visible-in-window-p shown) is true, so recenter is not
called; so the question to me is why this is not the case in the batch run.

>> A second problem concerns trying to reference a message in a test.  The
>> function todo-toggle-view-done-items outputs a message if the category
>> contains no done items, and the test captures this by calling
>> current-message immediately after todo-toggle-view-done-items.  This
>> works when running the test interactively, but in a batch run,
>> current-message is nil (though the message is output in the shell).
>> Why?  (I avoid the failure in batch mode by testing a different but
>> effectively equivalent condition.)
>
> ‘current-message’ returns "the string currently displayed in the echo
> area" and there's no echo area in batch mode.

Yes, as Noam also pointed out; I didn't think carefully enough about the
use of current-message.

>> Finally, one of the tests involves a display overlay that hides the item
>> header, which as a consequence, when using todo-mode, prevented the
>> cursor from appearing where the code put point, resulting in a display
>> bug.  I fixed this (in master 264dd81) by moving point to the first
>> visible position after the overlay.  But when running the test, the
>> overlay evidently does not inhibit point, but since the test, following
>> the bugfix code in todo-mode.el, assumes it does, it fails.  It seems
>> that the test environment does not reflect the same display mechanisms
>> that using the code does.  I haven't found a solution or workaround for
>> this, so the test is currently marked ":expected-result :failed".  I
>> hope someone can come up with a better alternative.
>
> You need a frame to display something and there's none in batch mode.
> Hence you can't test display features in batch mode.

Thanks for clarifying that for me.  I guess I'll have to leave such
features untested, or is there an alternative?

Thanks for the feedback.

Steve Berman