ThothII/harness/.pi/extensions/gate/core/enrich.js

// enrich.js -- PURE merge/enrichment for the v2 gate payloads (WS2).
//
// Catalog lookups are INJECTED as `getColumns` params (for testability): a function
// (tableName) => { description, columns: [{ name, description, ... }] } | null.
// A catalog miss (null) yields empty descriptions, never an error (the hard-fail
// stays only in reviewer_schema_linking).
//
// No dependency on tht-gate.js: importable and testable in isolation.

"use strict";

// Memoize a raw catalog lookup so repeated table hits do a single I/O call. The
// gate wraps its `tht schema columns <t> --json` (try/catch -> null on miss) and
// passes it here; enrich itself stays pure.
function memoizeGetColumns(rawGetColumns) {
	const cache = new Map();
	return function getColumns(tableName) {
		if (cache.has(tableName)) return cache.get(tableName);
		const v = rawGetColumns(tableName);
		cache.set(tableName, v);
		return v;
	};
}

function columnDescription(cat, colName) {
	if (!cat || !Array.isArray(cat.columns)) return "";
	const col = cat.columns.find((c) => c.name === colName);
	return (col && col.description) || "";
}

// Contract A. Fills index (1-based), tables[].description and filters[].description
// via getColumns. For filters, `column: "tab.col"` resolves to the column's catalog
// description. Returns a NEW object; model-owned fields pass through unchanged.
function enrichCtePlanV2(data, getColumns) {
	const ctes = (data.ctes || []).map((cte, i) => {
		const out = { ...cte, index: i + 1 };
		if (Array.isArray(cte.tables)) {
			out.tables = cte.tables.map((t) => {
				const cat = getColumns(t.name);
				return { ...t, description: (cat && cat.description) || "" };
			});
		}
		if (Array.isArray(cte.filters)) {
			out.filters = cte.filters.map((f) => {
				const out2 = { ...f };
				if (typeof f.column === "string" && f.column.includes(".")) {
					const idx = f.column.indexOf(".");
					const tableName = f.column.slice(0, idx);
					const colName = f.column.slice(idx + 1);
					out2.description = columnDescription(getColumns(tableName), colName);
				} else {
					out2.description = "";
				}
				return out2;
			});
		}
		return out;
	});
	return { ...data, ctes };
}

// Contract B. Fuses the model's thin data with `tht cte info` output.
//   name/index/total/sql            <- cteInfo
//   purpose/rationale/depends_on    <- cteInfo.doc (fallback: thinData)
//   status/execution_ms/row_sample/warnings/sql_hash/preview <- cteInfo.last_test
//   note                            <- thinData
// preview = { columns: last_test.columns, rows: last_test.preview_rows }.
function buildCteResultV2(thinData, cteInfo) {
	const thin = thinData || {};
	const doc = cteInfo.doc || {};
	const lt = cteInfo.last_test || {};
	const pick = (docVal, thinVal) => (docVal !== undefined ? docVal : thinVal);
	return {
		schema_version: 2,
		name: cteInfo.name,
		index: cteInfo.index,
		total: cteInfo.total,
		purpose: pick(doc.purpose, thin.purpose),
		rationale: pick(doc.rationale, thin.rationale),
		depends_on: pick(doc.depends_on, thin.depends_on),
		sql: cteInfo.sql,
		status: lt.status,
		execution_ms: lt.execution_ms,
		row_sample: lt.row_sample,
		warnings: lt.warnings || [],
		sql_hash: lt.sql_hash,
		columns: (lt.columns || []).map((name) => ({ name, description: "" })),
		preview: { columns: lt.columns || [], rows: lt.preview_rows || [] },
		note: thin.note,
	};
}

// Contract C. Fills `phase` from phaseMeta and the `description` fields in
// tables[]/tables[].columns[]/sections[].items[] via getColumns. phaseMeta is
// { id, num, name }. Returns a NEW object.
function enrichPhaseSummaryV2(data, phaseMeta, getColumns) {
	const out = { ...data, phase: phaseMeta };
	if (Array.isArray(data.tables)) {
		out.tables = data.tables.map((t) => {
			const cat = getColumns(t.name);
			const enriched = { ...t, description: (cat && cat.description) || "" };
			if (Array.isArray(t.columns)) {
				enriched.columns = t.columns.map((c) => ({
					...c,
					description: columnDescription(cat, c.name),
				}));
			}
			return enriched;
		});
	}
	if (Array.isArray(data.sections)) {
		out.sections = data.sections.map((s) => {
			if (!Array.isArray(s.items)) return { ...s };
			return {
				...s,
				items: s.items.map((it) => {
					// An item's description resolves against its declared table+column.
					let desc = "";
					if (typeof it.table === "string" && typeof it.column === "string") {
						desc = columnDescription(getColumns(it.table), it.column);
					} else if (typeof it.table === "string") {
						const cat = getColumns(it.table);
						desc = (cat && cat.description) || "";
					}
					return { ...it, description: desc };
				}),
			};
		});
	}
	return out;
}

// Fills columns[].description of a built cte_result payload by matching each column
// name against the catalog of the plan's tables (`planTables`: string[]). First
// table whose catalog carries the column wins; unresolved -> empty description.
// Mutates a copy of payload.columns; returns the payload.
function enrichCteResultColumns(payload, planTables, getColumns) {
	if (!Array.isArray(payload.columns)) return payload;
	const tables = Array.isArray(planTables) ? planTables : [];
	const columns = payload.columns.map((col) => {
		let desc = "";
		for (const t of tables) {
			const d = columnDescription(getColumns(t), col.name);
			if (d) {
				desc = d;
				break;
			}
		}
		return { ...col, description: desc };
	});
	return { ...payload, columns };
}

// Contract C bis. Appends a deterministic "decisions of this phase" section built
// from the session ledger, so the model's recap can stay thin (Discipline 6 exactness
// comes from persisted state, not model prose). `decisions` is session-show's ledger
// dump; `emits` is the phase's substantive decision-type list from workflow.yaml.
// No matching decisions -> data returned unchanged. Pure: returns a NEW object.
function appendLedgerSection(data, decisions, emits) {
	const emitSet = new Set(emits || []);
	const rows = (decisions || []).filter((d) => d && emitSet.has(d.type));
	if (rows.length === 0) return data;
	const section = {
		title: "Decisioni registrate in questa fase (dal ledger)",
		items: rows.map((d) => ({
			label: `${d.type}: ${d.subject}`,
			value: d.detail || "",
			kind: "decision",
			rationale: d.rationale || "",
		})),
	};
	return { ...data, sections: [...(data.sections || []), section] };
}

module.exports = {
	memoizeGetColumns,
	enrichCtePlanV2,
	buildCteResultV2,
	enrichCteResultColumns,
	enrichPhaseSummaryV2,
	appendLedgerSection,
};