How to calculate X when dealing with truncate-lines non-nil.

How to calculate X when dealing with truncate-lines non-nil.

[Keith David Bershatsky]

In developing my own feature requests to draw crosshairs (#17684) using multiple fake cursors (#22873), I have reached a roadblock in understanding why

move_it_to (&it, PT, it.last_visible_x, it.last_visible_y - 1, -1, MOVE_TO_POS | MOVE_TO_X | MOVE_TO_Y);

is giving me an it.current_x of non-existent coordinates to the imaginary right of the visible window when truncate-lines is non-nil.

For example, my screen is 1920 x 1080.  When PT is somewhere to the right of the initial visible window, I'm getting X coordinates to a tune of 8,000 and so forth.

How can I obtain the correct X, Y, HPOS, VPOS when dealing with a non-nil truncate-lines situation?

[Everything works as expected when truncate-lines is nil.]

Thanks,

Keith

Re: How to calculate X when dealing with truncate-lines non-nil.

[Keith David Bershatsky]

I believe this issue is resolvable mathematically as follows:

    int x;
    if (it.line_wrap == TRUNCATE)
      x = it.current_x - it.first_visible_x;
      else
        x = it.current_x;

Unless anyone sees a problem with this method, the issue is resolved.

Thanks,

Keith

Re: How to calculate X when dealing with truncate-lines non-nil.

[Eli Zaretskii]

> Date: Wed, 27 Sep 2017 20:16:11 -0700
> From: Keith David Bershatsky <esq@lawlist.com>
> 
> In developing my own feature requests to draw crosshairs (#17684) using multiple fake cursors (#22873), I have reached a roadblock in understanding why
> 
> move_it_to (&it, PT, it.last_visible_x, it.last_visible_y - 1, -1, MOVE_TO_POS | MOVE_TO_X | MOVE_TO_Y);
> 
> is giving me an it.current_x of non-existent coordinates to the imaginary right of the visible window when truncate-lines is non-nil.
> 
> For example, my screen is 1920 x 1080.  When PT is somewhere to the right of the initial visible window, I'm getting X coordinates to a tune of 8,000 and so forth.
> 
> How can I obtain the correct X, Y, HPOS, VPOS when dealing with a non-nil truncate-lines situation?

You didn't show enough of your code to answer the question.  Please
show everything from the call to init_iterator or start_display to the
above call to move_it_to.

Re: How to calculate X when dealing with truncate-lines non-nil.

[Eli Zaretskii]

> Date: Thu, 28 Sep 2017 10:31:58 -0700
> From: Keith David Bershatsky <esq@lawlist.com>
> 
> I believe this issue is resolvable mathematically as follows:
> 
>     int x;
>     if (it.line_wrap == TRUNCATE)
>       x = it.current_x - it.first_visible_x;
>       else
>         x = it.current_x;
> 
> Unless anyone sees a problem with this method, the issue is resolved.

This could be a solution in some cases.  The real answer depends on
what exactly does your code do, something that you didn't disclose.

In any case, I think that the 'it.line_wrap == TRUNCATE' condition is
not needed: if you want the window-relative X coordinate, you should
always subtract the first_visible_x value: if it's zero, no harm will
be done.

Re: How to calculate X when dealing with truncate-lines non-nil.

[Keith David Bershatsky]

Thank you, Eli, for helping me to simplify the code for the window-relative X coordinate.  I have implemented the suggested revision.

I now have a working draft of crosshairs for all three situations (i.e., TRUNCATE, WINDOW_WRAP, and WORD_WRAP) by using a combination of window-relative X *and* the absolute IT X.  Drawing/erasing multiple fake cursors is done with the former, whereas decision making is based on the latter.

Keith

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

DATE:  [09-29-2017 09:25:44] <29 Sep 2017 19:25:44 +0300>
FROM:  Eli Zaretskii <eliz@gnu.org>
> 
> > Date: Thu, 28 Sep 2017 10:31:58 -0700
> > From: Keith David Bershatsky <esq@lawlist.com>
> > 
> > I believe this issue is resolvable mathematically as follows:
> > 
> >     int x;
> >     if (it.line_wrap == TRUNCATE)
> >       x = it.current_x - it.first_visible_x;
> >       else
> >         x = it.current_x;
> > 
> > Unless anyone sees a problem with this method, the issue is resolved.
> 
> This could be a solution in some cases.  The real answer depends on
> what exactly does your code do, something that you didn't disclose.
> 
> In any case, I think that the 'it.line_wrap == TRUNCATE' condition is
> not needed: if you want the window-relative X coordinate, you should
> always subtract the first_visible_x value: if it's zero, no harm will
> be done.