[SL-1711] [Feature] Support resolution of time dimension grain when using Dimension() object syntax

[Jstein77]

Is this your first time submitting a feature request?

• I have read the expectations for open source contributors
• I have searched the existing issues, and I could ncot find an existing issue for this feature
• I am requesting a straightforward extension of existing metricflow functionality, rather than a Big Idea better suited to a discussion

Describe the feature

Currently we only support ambiguous resolution of time grains if you use the TimeDimension() object syntax. For example:

select
*
from {{semantic_layer.query(
metrics = ['revenue'],
where = "{{TimeDimension('metric_time')}} > 2024-01-01"
)
}}

Will resolve to the lowest granularity for metric time. If instead, I use the Dimenson() object syntax this query will error

select
*
from {{semantic_layer.query(
metrics = ['revenue'],
where = "{{Dimension('metric_time')}} > 2024-01-01"
)
}}

We should be able to resolve to the lowest available granularity for either syntax.

This is currently blocking implementing supporting date_part in Tableau.

Describe alternatives you've considered

No response

Who will this benefit?

No response

Are you interested in contributing this feature?

No response

Anything else?

No response

_{From SyncLinear.com | SL-1711}

[aiguofer]

One thing to add is that the use of {{ Dimension... }} and {{ TimeDimension... }} in filters inconsistent.

We need to be able to create a filter with:

• No grain, no date part ("exact date" in tableau)
• Only grain (DATE_TRUNC query in tableau)
• Only date part (EXTRACT query in tableau)

Currently, it seems like each syntax supports some (different) subset of these scenarios. I think either one or both of the different syntax should be able to cover all 3 scenarios.

[Jstein77]

@jordan.stein to see if this is fixed

[Jstein77]

Good one to followup on when caching is done.