OK, I couldn’t figure out a way to handle the temporal axis issue, so I decided to avoid it altogether and instead use a quantitative X axis for both visuals. So, I computed the hours and minutes as integers, then composed decimal numbers from them as needed. Here’s what I came up with:
Also, I knew I’d used a “visual switching” technique a while ago, based on a “hack” discovered by Madison Giammaria last year:
- add a parameter for an expression to retrieve the frequency selected in the Power BI slicer
- add a parameter for an expression to set the “padding” of the child visuals
Vega-Lite then uses the padding values when rendering the visual.
Here’s the [params] block from my I used in this second version:
{
"params": [
{
"name": "_meeting_frequency",
"expr": "data('dataset')[0]['Frequency']"
},
{
"name": "padding",
"expr": "{ top: _meeting_frequency == 'Daily' ? 20 : -540, bottom: _meeting_frequency == 'Daily' ? -530 : 20, left: _meeting_frequency == 'Daily' ? -130 : 10, right: _meeting_frequency == 'Daily' ? 0 : 0 }"
}
]
}
Here’s the links:
Technique found by Madison Giammaria and posted to LinkedIn in July 2024:
GP’s example from July 2024 using this technique in the Deneb Showcase section of the Enterprise DNA website:
As this was only an alternative solution, I was a bit quick-and-dirty: although I tried to use several of your code blocks, I’m sure I missed several (e.g., tooltips, yOffset, etc.). There’s probably some left over development artifacts as well. Nevertheless …
I know its not what you were looking for, but I hope it helps.
Greg
eDNA Forum - Deneb Team Meetings - V2.pbix (2.4 MB)
Meetings.xlsx (13.0 KB)