Hi René,
sorry for the late response.
On Tue, 2022-03-08 at 14:01 +0100, rene mayer wrote:
> I've a multiple choice set up with the question in front of the
> multiple choice check boxes.
> The code below results in having the question first aligned to the
> check boxes.
>
> begin{choicegroup}[layouter=rotated]{}
> \groupaddchoice[text=choice 1]{überhaupt nicht}
> \groupaddchoice[text=choice 2]{sehr stark}
> \choiceline{1. Kopfschmerzen}
>
> I would like to have the check-boxes first followed by the question
> left aligned to the check box.
> How to do this?
I don't think I have any support for this at the moment.
This is really complicated conceptually. The "sdapsarray" is
conceptually a matrix with row and column headers. What you want to do
here, is to display the row header on the right rather than the left
side.
>From a technical perspective, we have the layouter that allows hooking
into the process. In there, the "row" macro deals with building a line.
This is where the change would need to happen.
I suspect, you would need to effectively create a new layouter. Then,
override various functions:
* row (i.e. \_sdaps_array_row:NNNN):
Just change the order of boxes, pretty simple, see RTL code
* rowhead (i.e. \_sdaps_array_rowhead_default:Nw):
Change the alignment, also simple, see RTL code
Would need some refactoring to make the code nice. Or, one would need
to add an option that toggles the behaviour and makes the layouters
behave as if the writing direction was flipped in some cases (RTL vs.
LTR).
That said, this actually shows a possible workaround. i.e. change the
language to RTL and reverse the the order of the choices. But, to be
honest I am not sure how to do that right now (in principle, if@RTL
needs to evaluate to TRUE).
Though, I just had a quick go it trying it by defining
\makeatletter
\newif\if@RTL
\makeatother
And setting it using
\makeatletter
\@RTLtrue
\makeatother
And, now I am getting an error while compiling the document, so I guess
something might be broken with RTL support actually. I don't think I
have regression tests for that currently :-/
Benjamin
--
To unsubscribe, send mail to [email protected].
