feat(sources): pin all four ECMA-376 parts in the manifest

Replace the single `ecma-376` placeholder with four part-specific
entries (`ecma-376-part1` through `ecma-376-part4`), each pinned with
URL, edition (5th), publication date, and sha256. Part 2 is the 2021
revision; Part 3 is from 2015; Parts 1 and 4 are 2016 - reflected in
each entry's version field.

The Part 4 zip URL is shared with the existing ecma-376-transitional
entry (the XSD zip is extracted from inside Part 4); both rows pin the
same outer-zip sha256.

scripts/sources-sync.ts now backfills spec_content.source_id by
part_number, mapping each row to the matching ecma-376-partN source.
The previous backfill targeted a single `ecma-376` source which no
longer exists.

Migration 0005 cleans up the legacy `ecma-376` placeholder row from
reference_sources, but only if no spec_content row references it
(safe for a developer who had already backfilled to the placeholder
id; idempotent).

44 / 0 tests still passing.

Author: caiopizzol

data/sources.json

@@ -2,13 +2,40 @@
   "$comment": "Source manifest. Human-edited; scripts/sources-sync.ts upserts these rows into reference_sources.",
   "sources": [
     {
-      "name": "ecma-376",
+      "name": "ecma-376-part1",
       "kind": "spec_pdf",
-      "edition": "unknown",
-      "version": null,
-      "url": "https://ecma-international.org/publications-and-standards/standards/ecma-376/",
+      "edition": "5th",
+      "version": "2016-12",
+      "url": "https://ecma-international.org/wp-content/uploads/ECMA-376-1_5th_edition_december_2016.zip",
+      "license_note": "Published by Ecma International. See the ECMA-376 publications page for the current download and licensing terms before redistribution.",
+      "sha256": "9d0bcad9cf06054785b03762fcfadbf6bab7e54a5f9d69434e34b7fd464d4129"
+    },
+    {
+      "name": "ecma-376-part2",
+      "kind": "spec_pdf",
+      "edition": "5th",
+      "version": "2021-12",
+      "url": "https://ecma-international.org/wp-content/uploads/ECMA-376-2_5th_edition_december_2021.zip",
+      "license_note": "Published by Ecma International. See the ECMA-376 publications page for the current download and licensing terms before redistribution.",
+      "sha256": "1d489dc491168ea1f9e9a59063acc8dd5f02b4ad1d21aa7ec19ba9a58d020c70"
+    },
+    {
+      "name": "ecma-376-part3",
+      "kind": "spec_pdf",
+      "edition": "5th",
+      "version": "2015-12",
+      "url": "https://ecma-international.org/wp-content/uploads/ECMA-376-3_5th_edition_december_2015.zip",
+      "license_note": "Published by Ecma International. See the ECMA-376 publications page for the current download and licensing terms before redistribution.",
+      "sha256": "42294159fbbbe9393ccadac95b859d7729cc68d908898bcbe31034dda059daa8"
+    },
+    {
+      "name": "ecma-376-part4",
+      "kind": "spec_pdf",
+      "edition": "5th",
+      "version": "2016-12",
+      "url": "https://ecma-international.org/wp-content/uploads/ECMA-376-4_5th_edition_december_2016.zip",
       "license_note": "Published by Ecma International. See the ECMA-376 publications page for the current download and licensing terms before redistribution.",
-      "sha256": null
+      "sha256": "bd25da1109f73762356596918bf5ff8b74a1331642dba5f1c1d1dfc6bed34ecd"
     },
     {
       "name": "ecma-376-transitional",

db/migrations/0005_drop_legacy_ecma376.sql

@@ -0,0 +1,17 @@
+-- Drop the legacy `ecma-376` placeholder source row.
+--
+-- An earlier version of data/sources.json had a single placeholder
+-- (`ecma-376`, edition=unknown, sha256=null) that stood in for the whole
+-- spec corpus before per-part entries existed. The manifest now pins the
+-- four ECMA-376 parts individually (`ecma-376-partN`), so the placeholder
+-- is obsolete.
+--
+-- This migration only deletes the row when nothing in spec_content
+-- references it, so a developer who already backfilled source_id to the
+-- legacy id stays safe. Idempotent.
+
+DELETE FROM reference_sources
+WHERE name = 'ecma-376'
+  AND NOT EXISTS (
+    SELECT 1 FROM spec_content WHERE spec_content.source_id = reference_sources.id
+  );

scripts/sources-sync.ts

@@ -68,16 +68,23 @@ async function main() {
 			);
 		}
 
-		const [ecma] = await sql<[{ id: number } | undefined]>`
-			SELECT id FROM reference_sources WHERE name = 'ecma-376' LIMIT 1
-		`;
-		if (ecma) {
+		// Backfill spec_content.source_id by part_number to the matching
+		// ecma-376-partN row. Idempotent: only touches rows where source_id IS NULL.
+		for (let part = 1; part <= 4; part++) {
+			const sourceName = `ecma-376-part${part}`;
+			const [src] = await sql<[{ id: number } | undefined]>`
+				SELECT id FROM reference_sources WHERE name = ${sourceName} LIMIT 1
+			`;
+			if (!src) continue;
 			const result = await sql`
-				UPDATE spec_content SET source_id = ${ecma.id} WHERE source_id IS NULL
+				UPDATE spec_content SET source_id = ${src.id}
+				WHERE part_number = ${part} AND source_id IS NULL
 			`;
-			console.log(`Backfilled ${result.count} spec_content row(s) -> source_id=${ecma.id}`);
-		} else {
-			console.warn("No ecma-376 source row found; skipped spec_content backfill.");
+			if (result.count > 0) {
+				console.log(
+					`Backfilled ${result.count} spec_content row(s) (part ${part}) -> source_id=${src.id}`,
+				);
+			}
 		}
 	} finally {
 		await db.close();