What is difference between domain "expr" and domainRaw "expr"?

[jwoLondon]

I understand that since VL 5.13 it is possible to specify a scaling domain with domainRaw so that it may be parameterised. But since domain can also take an expression as an argument, it is not clear to me what extra domainRaw brings. Can anyone explain the difference between the two forms, and why both are needed?

[lsh]

A reason to use domainRaw is that it doesn't overwrite the inferred domain, and when domainRaw evaluates to null, the inferred domain is used. A minimal(ish) example is the following snippet:

{
  "$schema": "https://vega.github.io/schema/vega-lite/v5.json",
  "description": "A simple bar chart with embedded data.",
  "data": {
    "values": [
      {"a": "A", "b": 28},
      {"a": "B", "b": 55},
      {"a": "C", "b": 43},
      {"a": "D", "b": 91},
      {"a": "E", "b": 81},
      {"a": "F", "b": 53},
      {"a": "G", "b": 19},
      {"a": "H", "b": 87},
      {"a": "I", "b": 52}
    ]
  },
  "mark": {"type": "bar", "clip": true},
  "encoding": {
    "x": {"field": "a", "type": "nominal", "axis": {"labelAngle": 0}},
    "y": {
      "field": "b",
      "type": "quantitative",
      "scale": {"domainRaw": {"expr": "ydomain"}}
    }
  },
  "config": {
    "signals": [
      {
        "name": "ydomain",
        "update": "null",
        "on": [
          {"events": [{"type": "mousedown"}], "update": "[20, 50]"},
          {"events": [{"type": "mouseup"}], "update": "null"}
        ]
      }
    ]
  }
}

On mousedown, the chart will zoom into a predefined region, and on mouseup it snaps back.

Screen.Recording.2023-09-01.at.1.10.34.PM.mov

With a little extra work, making custom zoom, brushing, and filtering actions becomes a little easier. To do the same thing without domainRaw would require getting the domain extent and storing it, and much more state management.