Event count

Counts how often the Bull bar definition fired, and on which weekdays. It is the smallest complete study: one gate in scope, one collect, three results.

What it measures

Every bull_bar event in the run, grouped by UTC weekday, with the average bar range for each day.

Key decisions

Piece	Why
`flows.declare({ slug, as, version })`	Binds the definition's recorded events. `as` is what `ctx.event.flow` holds. `version` is the one you ran over history.
`scope` returns `{}` or `null`	Every event study exports `scope`. No dimensions, so a bull bar returns an empty object and any other event returns `null`.
`ctx.collect` in `onEvent`, `ctx.publish` in `onFinish`	Store one row per event, aggregate once all events are in.
Table by weekday	At most seven rows. A table with a row per event is too long to read.
`readout__summary` and `readout__events`	A written answer and a headline count, which lint expects beside a table.

Full script

meta({ name: "Bull bar count", kind: "study" });

flows.declare({ slug: "bull-bar", as: "bull", version: 2 });

results.declare({
  id: "readout__summary",
  kind: "prose",
  role: "answer",
  title: "Summary",
  description: "How often bull bars fired and on which weekday most often.",
});

results.declare({
  id: "readout__events",
  kind: "metric",
  title: "Bull bars",
  description: "Bull bar events in the run window.",
});

results.declare({
  id: "events_by_weekday",
  kind: "table",
  title: "Bull bars by weekday",
  description: "Event count and average bar range (high minus low) for each UTC weekday.",
});

export function scope(ctx) {
  if (ctx.event.flow !== "bull") return null;
  return {};
}

export function onEvent(ctx, scope) {
  ctx.collect("rows", {
    weekday: ctx.session.bucket(ctx.event.time, "dow"),
    range: ctx.event.payload.range,
  });
  ctx.mark(ctx.event.time, ctx.event.payload.close, { color: "#26a69a", shape: "circle" });
}

export function onFinish(ctx) {
  const rows = ctx.collected("rows");
  const counts = ctx.bin.byKey(rows, (r) => r.weekday, (r) => r.range, "count");
  const avgRange = new Map(ctx.bin.byKey(rows, (r) => r.weekday, (r) => r.range, "mean"));

  const table = counts.map(([weekday, events]) => ({
    weekday,
    events,
    avg_range: Number(avgRange.get(weekday).toFixed(2)),
  }));
  ctx.publish("events_by_weekday", table);

  ctx.publish("readout__events", { value: rows.length, format: "integer" });

  const busiest = table.reduce((best, r) => (r.events > best.events ? r : best), table[0]);
  ctx.publish("readout__summary", {
    body: rows.length
      ? `${rows.length} bull bars. ${busiest.weekday} had the most, with ${busiest.events}.`
      : "No bull bars in this run.",
    role: "answer",
  });
}

This study only counts, so Attach to chart does not offer it. Forward return measures what came next.

PreviousExamples NextForward return