Composing predicates in alt.condition() is tedious

[dechamps]

I want to OR two predicates in an alt.condition() call.

Initially I just assumed I could use a similar syntax as for transform_filter():

(alt.Chart(data=None)
    .mark_circle()
    .encode(color=alt.condition({'or': [
                    alt.FieldEqualPredicate(field='field1', equal='value1'),
                    alt.FieldEqualPredicate(field='field2', equal='value2'),
                ]}, alt.value('red'), alt.value('green'))))

But that doesn't work:

SchemaValidationError: Invalid specification
        altair.vegalite.v4.schema.channels.ColorValue, validating 'additionalProperties'
        Additional properties are not allowed ('or' was unexpected)

Then I thought maybe the operators would work:

(alt.Chart(data=None)
    .mark_circle()
    .encode(color=alt.condition(
            alt.FieldEqualPredicate(field='field1', equal='value1') |
            alt.FieldEqualPredicate(field='field2', equal='value2'),
        alt.value('red'), alt.value('green'))))

But nope:

TypeError: unsupported operand type(s) for |: 'FieldEqualPredicate' and 'FieldEqualPredicate'

After fumbling around in the dark for some time, this is the best I came up with:

(alt.Chart(data=None)
    .mark_circle()
    .encode(color=alt.condition(alt.LogicalOrPredicate(**{'or': [
                    alt.FieldEqualPredicate(field='field1', equal='value1'),
                    alt.FieldEqualPredicate(field='field2', equal='value2'),
                ]}), alt.value('red'), alt.value('green'))))

This actually works, but the syntax is contrived and this approach doesn't seem documented. Should the API be improved, or am I missing something here?

[jakevdp]

That works; a clearer option is probably this:

(alt.Chart(data=None)
    .mark_circle()
    .encode(color=alt.condition(
        (alt.datum.field1 == 'value1') | (alt.datum.field2 == 'value2'),
        alt.value('red'), alt.value('green')))
)

Adding operator support to other predicates is tracked here: #695

But it hasn't been a priority because vega expressions like the one above cover the use-cases pretty well.

[joelostblom]

@dechamps I am going through Altair issues to find those that have been resolved and can be closed. It looks to me like this issue has been solved so I am closing it, but please feel free to reopen and add a comment if there is something you don't think is resolved yet.