property x0shift, x1shift, y0shift, y1shift for adjusting the shape coordinates

[my-tien]

This property adjusts the shape coordinates if xref/yref references a (multi-)category axis.

A shape at position x0/x1 = "A"

with x0shift = 0.5

Disclaimer

I am required to add that…

the software is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. in no event shall the authors or copyright holders be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the software or the use or other dealings in the software.

[archmoj]

We should also add these to newselection and test it.

[archmoj]

Is this also working for a path?

[my-tien]

No, this should only do something when using x0, x1, y0, y1

[archmoj]

Interesting PR for shapes and selections.
But I started wondering perhaps we should add a categoryshift option to Cartesian axes instead?

[archmoj]

I am wondering if adding a categoryselectionstart, categoryselectionend, categoryshapestart and categoryshapeend to cartesian axes is a better option?

[archmoj]

Let's ask @alexcjohnson to review 🔎

[my-tien]

I am wondering if adding a categoryselectionstart, categoryselectionend, categoryshapestart and categoryshapeend to cartesian axes is a better option?

Here is a PR with this suggestion for shapes only: #7010 (I actually only included selection in this PR, because it reuses shape properties.)

[archmoj]

Considering #7010 (comment) discussion,
two attributes are needed for start and the end i.e. instead of xshift which is already implemented we need x0shift and x1shift.

[archmoj]

Thanks for the simplification.
Please update the PR title and description.
Also zzz_shape_shift_vertical image test is failing now.

[archmoj]

@stephprobst Do you have a use case for date axes?
If so it could be possible to add these options for other axis types.

[stephprobst]

@stephprobst Do you have a use case for date axes? If so it could be possible to add these options for other axis types.

@archmoj No, I don't think this is relevant for other axes. The rationale for this property is to allow exact positioning of a shape based on the axis value. For categorical axes this is needed, because it allows the combination of categorical and numerical values (e.g. country "Canada" plus a shift of 0.5 to position the shape at the end of the categories interval on the axis). On numerical or date axes this isn't really needed, since the user can directly select the precise numerical or date value (e.g. "2023-12-31" to position a shape at the end of year 2023).

My proposal would be to skip the implementation for date and numeric axes.

[archmoj]

When shape.editable is set to true, clicking the shape and editing the position of a vertex result in the wrong positioning of the updated shape.

[archmoj]

Please also test texttemplate on your mocks to show the slopes similar to those in the text_on_shapes_texttemplate mock.
Thank you!

[my-tien]

Please also test texttemplate on your mocks to show the slopes similar to those in the text_on_shapes_texttemplate mock. Thank you!

I noticed that I didn't account for the shift for the texttemplate. Fixed that and added a test shape with slope and xcenter.

[my-tien]

When shape.editable is set to true, clicking the shape and editing the position of a vertex result in the wrong positioning of the updated shape.

Fixed

[archmoj]

Thanks @my-tien for the PR 🥇
💃