brucala / deneb.jl Goto Github PK

This issue is used to trigger TagBot; feel free to unsubscribe.

If you haven't already, you should update your TagBot.yml to include issue comment triggers.
Please see this post on Discourse for instructions and more details.

If you'd like for me to do this for you, comment TagBot fix on this issue.
I'll open a PR within a few hours, please be patient!

https://vegafusion.io/

Due to the shorthand string syntax, using chars like ( or : in the Encoding can give unexpected results:

julia > Encoding("Beak Length (mm):Q")
Deneb.EncodingSpec: 
{
  "x": {
    "aggregate": "Beak Length ",
    "field": "mm",
    "type": "quantitative"
  }
}

At the moment there are no escaping rules to avoid that. In the meantime one can simply use the verbose syntax without the shorthands:

> julia> Encoding(x=(field="Beak Length (mm)", type=:quantitative))
Deneb.EncodingSpec: 
{
  "x": {
    "field": "Beak Length (mm)",
    "type": "quantitative"
  }
}

For some reason the cumulative frequency distribution example stopped working:

However, if I open the chart in the Vega editor, or if I run the plot locally, it works well.

Currently saving charts relies on NodeJS (and that's the only reason Deneb.jl depends on NodeJS). An alternative (currently used by Altair) is to use vl-convert instead. Blog post about vl-convert.

One convenience about vl-convert is that it would be an standalone artifact that does not require any external dependencies, unlike the approach with NodeJS that requires to install things like Canvas.

Unlike NodeJS, vl-convert does not convert to pdf. We could keep both dependencies (vl-convert for png/svg and NodeJS for pdf) but maybe make NodeJS a weak dependency.

In the composition of a Transform with a Facet, the transform disappears:

transform = Transform(
    fold=stations,
    as=[:Station, :snow_depth],
) 
facet = Facet(row=:Station) * Encoding(x="monthdate(date):O")
transform * facet

results in:

{
    "mark": {
      "tooltip": true
    }
  },
  "spec": {
    "encoding": {
      "x": {
        "timeUnit": "monthdate",
        "field": "date",
        "type": "ordinal"
      }
    },
    "width": 500
  },
  "facet": {
    "row": {
      "field": "Station"
    }
  }
}

Charts are not being displayed properly in Jupyter.
I think in the past it work because the mime type application/vnd.vegalite.v4+json was automatically available in Jupyter lab, but v5 isn't.

While trying out Deneb, the following warning was printed:

npm notice
npm notice New major version of npm available! 8.5.5 -> 9.8.0
npm notice Changelog: https://github.com/npm/cli/releases/tag/v9.8.0
npm notice Run npm install -g [email protected] to update!
npm notice

I don't have npm installed. Is this an artifact that Deneb installs? If so, should the version be updated?