[Emacspeak] Re: Stage part of a hunk in magit

[Tim Cross <theophilusx@xxxxxxxxxxx>]

Tim Makarios <emacspeak.correspondence(a)freespoken.nz> writes:

> On Thu, 2021-05-13 at 14:07 +1000, Tim Cross wrote:
>> In what way does it not work or does it behave differently when
>> emacspeak is running? Are there any errors? 
>
> If I've got multiple new consecutive lines in a file tracked by Git, and I
> want to stage only, say, the first two, I can bring up the diff in Magit,
> move to the first new line, press C-<SPC>, move to the second line, and
> press s to stage only those two marked lines.  This works if Emacspeak isn't
> loaded, but if Emacspeak is loaded, then it always stages the entire hunk
> (as determined by Git), regardless of where the mark is set.  I see no error
> messages, nor any difference in the output to the magit-process buffer.
>
> Also, without Emacspeak loaded, Magit uses different highlighting for the
> lines between the mark and point (inclusive) when I'm doing this, but I
> can't see any difference in Magit's highlighting when Emacspeak is loaded. 
> I realize that many on this list may not be able to appreciate the
> difference in highlighting, but I thought it might be useful diagnostically,
> since the Magit manual says (in the Getting Started chapter):
>> If the region looks like it does in other buffers, then it doesn’t select
>> Magit sections that can be acted on as a unit.
>

OK, now I understand the issue.

I just tried a simple example in my setup and it works as expected.
However, I run emacspeak with spacemacs, which is a slightly unusual
setup. The good news is that it obviously can work with emacspeak
running. 

I will try to see if I can reproduce your issue in my test environment.
I'm just in the middle of upgrading that environment to Ubuntu
21.04, so it might be a day or so before I can test it out.

My wild baseless guess is that when you are setting the region, it isn't
actually marking the region. The fact you don't see the highlighting
would also indicate this may be the case. I would start by verifying
C-<space> is actually calling the same function and is setting the
region. It might also be worth trying with transient mark mode enabled
and disabled to see if that makes any difference. 
-- 
Tim Cross