list_screen_columns
List Screener Columns
Reads only.
Use this for every column a predicate can name, with its meaning and unit. Use get_event_analytics instead for the values of those columns on one event. List every column usable in the screener/alert predicate DSL: (namespace, column_name, data_type, description, label, unit) for ea (event_analytics), e (events), pa (performer_analytics), p (performers: p.name only) and v (venues: v.venue_name, v.city, v.state, v.country_code, v.capacity). Mirrors the SQL compiler's allowlist exactly: anything not returned here will fail to compile. description explains the column's semantics, units, and NULL conditions. Includes the event demand columns (ea.demand_*), the performer-grain demand columns (pa.demand_*, a separate performer-resource reading rather than a rollup of that performer's events), performer popularity-rank columns (pa.popularity_rank*, pa.rank_asof_date), event sale-timing/priority columns (e.presale1_date, e.onsale_date, e.monitoring_priority), the percent changes of the listing and ticket counts (ea.listings_pct_*, ea.tickets_pct_*), the median of 7 days earlier (ea.median_price_7d_ago), and e.tm_sells. Labels mark a change with Δ and a percent change with %Δ, each followed by its window: Median Δ7d is the dollar change of the median over 7 days, Median %Δ7d its percent change. A window with no mark is a window, so Absorbed 7d counts over 7 days. A predicate names namespace.column_name, never the label. Call this first when building a predicate.
Parameters
None.
The label is the column's name in the app. A label marks a change with Δ and a
percent change with %Δ, each followed by its window: Median Δ7d is the dollar change
of the median over 7 days, Median %Δ7d its percent change. A window with no mark is a
window: Absorbed 7d counts over 7 days. A predicate names namespace.column_name,
never the label.
Call this first
Before building a predicate, call list_screen_columns to discover the exact columns
and their meanings. It replaces guessing at column names.
Example
list_screen_columns answers an array of column descriptors:
[
{
"namespace": "ea",
"column_name": "median_price_pct_7d",
"data_type": "numeric",
"description": "Percent change in median_price_current vs the last reading seven UTC days earlier.",
"label": "Median %Δ7d",
"unit": "%"
}
]