fix(#6): narrow lastInsertId to string or null

SPEC.md

@@ -117,7 +117,7 @@ HTTP status: `200`.
 - `columns` (REQUIRED, array of strings) — the column names of the result, in the order produced by the SQL engine. Empty array for statements that produce no result set (INSERT, UPDATE, DELETE, DDL).
 - `rows` (REQUIRED, array of arrays) — each inner array has the same length as `columns`, with values in column order. Values use the same JSON / tagged-value encoding as section 5. Empty array if no rows.
 - `rowsAffected` (REQUIRED, integer) — the number of rows changed by the statement. `0` for SELECT.
-- `lastInsertId` (OPTIONAL, string, number, or null) — the identifier of the most recently inserted row when the server can determine it (typically the auto-increment id). `null` when not applicable or not available.
+- `lastInsertId` (OPTIONAL, string or null) — the identifier of the most recently inserted row when the server can determine it (typically the auto-increment id). Always a string, never a JSON number: when the identifier is an integer it is encoded as a decimal string, consistent with the `bigint` encoding in section 5, so 64-bit ids survive clients whose numbers are IEEE-754 doubles. Text identifiers (a UUID primary key, say) are carried as-is. `null` when not applicable or not available.
 
 The arrays-of-arrays shape (not arrays-of-objects) is normative. It keeps payloads compact, makes column order explicit, and supports duplicate column names from joins.
 

conformance/README.md

@@ -71,7 +71,7 @@ Conformance is self-asserted. The community can call out failures via issues.
 ## Optional / "nice to have"
 
 - Vendor error codes carry the `vendor:` prefix.
-- `lastInsertId` is populated for INSERT statements where the SQL engine reports it.
+- `lastInsertId` is populated for INSERT statements where the SQL engine reports it, as a JSON string (integer ids as decimal strings) or `null` — never a JSON number.
 - Rate-limited responses return `error.code` = `rate_limited` and HTTP 429.
 - Server enforces a maximum result row count and returns `payload_too_large` past it.
 

examples/cloudflare-worker-to-d1/src/index.ts

@@ -21,7 +21,7 @@ interface StatementResult {
   columns: string[];
   rows: unknown[][];
   rowsAffected: number;
-  lastInsertId?: string | number | null;
+  lastInsertId?: string | null;
 }
 
 const VERSION = "0.1";
@@ -107,11 +107,14 @@ async function runBatch(db: D1Database, batch: Statement[], atomic: boolean): Pr
 function projectD1Result(res: D1Result): StatementResult {
   const rows = (res.results ?? []) as Record<string, unknown>[];
   const columns = rows.length > 0 ? Object.keys(rows[0]) : [];
+  // SPEC 6.1: lastInsertId is a string or null, never a JSON number, so 64-bit
+  // rowids survive clients whose numbers are IEEE-754 doubles.
+  const lastRowId = res.meta?.last_row_id;
   return {
     columns,
     rows: rows.map((r) => columns.map((c) => encodeValue(r[c]))),
     rowsAffected: res.meta?.changes ?? 0,
-    lastInsertId: res.meta?.last_row_id ?? null,
+    lastInsertId: lastRowId === undefined || lastRowId === null ? null : String(lastRowId),
   };
 }