fix(core): backfill locale casing on ec_* tables (emdash-cms#1572)

Addresses review feedback on emdash-cms#1572: removing localeCode's lowercasing
transform fixed new locale-filtered queries against the canonical
casing, but pre-existing rows already stored under the lowercased form
(e.g. zh-tw) were still missed by an exact locale = 'zh-TW' filter.

Adds migration 044, which canonicalizes the locale column on every
ec_* table against getI18nConfig().locales, case-insensitively. No-op
when i18n isn't configured. Not reversible (down is a no-op) -- the
original incorrect casing isn't recoverable and re-lowercasing would
reintroduce the bug.

Updates the "partially applied" migration-runner test to include the
new trailing migration.

Author: marcusbellamyshaw-cell

.changeset/localecode-preserve-casing.md

@@ -2,4 +2,4 @@
 "emdash": patch
 ---
 
-Fixes the content and search APIs returning zero results when filtering by a locale with an uppercase region or script subtag (e.g. `?locale=zh-TW`, `pt-BR`, `zh-Hant`). The `localeCode` validator lowercased the value, but config locales, the stored `locale` column, and the public query path all preserve the original casing, so the filter matched nothing. Validation stays case-insensitive; the value is now preserved verbatim. The taxonomy and menu endpoints' shared `?locale=` filter now applies the same BCP-47 validation, so a malformed locale is rejected with a clear error instead of silently matching nothing. Closes #1551.
+Fixes the content and search APIs returning zero results when filtering by a locale with an uppercase region or script subtag (e.g. `?locale=zh-TW`, `pt-BR`, `zh-Hant`). The `localeCode` validator lowercased the value, but config locales, the stored `locale` column, and the public query path all preserve the original casing, so the filter matched nothing. Validation stays case-insensitive; the value is now preserved verbatim. The taxonomy and menu endpoints' shared `?locale=` filter now applies the same BCP-47 validation, so a malformed locale is rejected with a clear error instead of silently matching nothing. A migration backfills existing content rows that were previously stored under the lowercased casing so upgraded sites match immediately, not just new content. Closes #1551.

packages/core/src/database/migrations/044_canonicalize_locale_casing.ts

@@ -0,0 +1,43 @@
+import type { Kysely } from "kysely";
+import { sql } from "kysely";
+
+import { getI18nConfig } from "../../i18n/config.js";
+import { listTablesLike } from "../dialect-helpers.js";
+
+/**
+ * Backfill migration for #1572.
+ *
+ * `contentCreateBody`'s `locale` field used to lowercase every explicit
+ * value (`localeCode`'s `.transform()`), so a site with a configured locale
+ * like `zh-TW` could end up with rows stored as `zh-tw`. The transform was
+ * removed so canonical-cased queries (`?locale=zh-TW`) now match, but that
+ * only fixes future writes -- existing rows already stored under the
+ * lowercased form are still missed by an exact-match `locale = 'zh-TW'`
+ * filter.
+ *
+ * For every `ec_*` content table, canonicalize any row whose `locale`
+ * case-insensitively matches a configured locale but isn't stored in that
+ * locale's canonical casing.
+ */
+export async function up(db: Kysely<unknown>): Promise<void> {
+	const locales = getI18nConfig()?.locales ?? [];
+	if (locales.length === 0) return;
+
+	const tableNames = await listTablesLike(db, "ec_%");
+
+	for (const tableName of tableNames) {
+		const table = sql.ref(tableName);
+		for (const locale of locales) {
+			await sql`
+				UPDATE ${table}
+				SET locale = ${locale}
+				WHERE lower(locale) = lower(${locale}) AND locale != ${locale}
+			`.execute(db);
+		}
+	}
+}
+
+export async function down(_db: Kysely<unknown>): Promise<void> {
+	// Not reversible: the original (incorrectly lowercased) casing is not
+	// recoverable, and re-lowercasing would reintroduce the bug this fixes.
+}

packages/core/src/database/migrations/runner.ts

@@ -45,6 +45,7 @@ import * as m040 from "./040_byline_i18n.js";
 import * as m041 from "./041_content_locale_list_index.js";
 import * as m042 from "./042_byline_fields.js";
 import * as m043 from "./043_content_references.js";
+import * as m044 from "./044_canonicalize_locale_casing.js";
 
 const MIGRATIONS: Readonly<Record<string, Migration>> = Object.freeze({
 	"001_initial": m001,
@@ -89,6 +90,7 @@ const MIGRATIONS: Readonly<Record<string, Migration>> = Object.freeze({
 	"041_content_locale_list_index": m041,
 	"042_byline_fields": m042,
 	"043_content_references": m043,
+	"044_canonicalize_locale_casing": m044,
 });
 
 /** Total number of registered migrations. Exported for use in tests. */

packages/core/tests/integration/database/migrations.test.ts

@@ -126,6 +126,7 @@ describe("Database Migrations (Integration)", () => {
 			"041_content_locale_list_index",
 			"042_byline_fields",
 			"043_content_references",
+			"044_canonicalize_locale_casing",
 		];
 
 		await db.deleteFrom("_emdash_migrations").where("name", "in", trailing).execute();

packages/core/tests/unit/database/migrations/044_canonicalize_locale_casing.test.ts

@@ -0,0 +1,95 @@
+import type { Kysely } from "kysely";
+import { sql } from "kysely";
+import { afterEach, beforeEach, describe, expect, it } from "vitest";
+
+import { up } from "../../../../src/database/migrations/044_canonicalize_locale_casing.js";
+import { ContentRepository } from "../../../../src/database/repositories/content.js";
+import type { Database } from "../../../../src/database/types.js";
+import { getI18nConfig, setI18nConfig } from "../../../../src/i18n/config.js";
+import { createPostFixture } from "../../../utils/fixtures.js";
+import { setupTestDatabaseWithCollections, teardownTestDatabase } from "../../../utils/test-db.js";
+
+/**
+ * Regression coverage for #1572.
+ *
+ * `contentCreateBody` used to lowercase every explicit `locale` value, so a
+ * site with a configured locale like `zh-TW` could have existing rows stored
+ * as `zh-tw`. Removing that transform fixed new queries against the
+ * canonical casing, but pre-existing rows would still be missed by an exact
+ * `locale = 'zh-TW'` filter without a data backfill.
+ */
+describe("migration 044: canonicalize locale casing (#1572)", () => {
+	let db: Kysely<Database>;
+	const previousI18nConfig = getI18nConfig();
+
+	beforeEach(async () => {
+		db = await setupTestDatabaseWithCollections();
+	});
+
+	afterEach(async () => {
+		setI18nConfig(previousI18nConfig);
+		await teardownTestDatabase(db);
+	});
+
+	it("rewrites rows stored in the wrong case to the configured locale's canonical casing", async () => {
+		setI18nConfig({ defaultLocale: "en", locales: ["en", "zh-TW"] });
+
+		const repo = new ContentRepository(db);
+		await repo.create(
+			createPostFixture({ slug: "guide", status: "published", data: { title: "Guide" } }),
+		);
+		// Simulate a pre-fix row: written under the lowercased locale.
+		await sql`UPDATE ec_post SET locale = 'zh-tw' WHERE slug = 'guide'`.execute(db);
+
+		await up(db);
+
+		const row = await db
+			.selectFrom("ec_post")
+			.select("locale")
+			.where("slug", "=", "guide")
+			.executeTakeFirstOrThrow();
+		expect(row.locale).toBe("zh-TW");
+	});
+
+	it("leaves rows already in canonical casing untouched", async () => {
+		setI18nConfig({ defaultLocale: "en", locales: ["en", "zh-TW"] });
+
+		const repo = new ContentRepository(db);
+		await repo.create(
+			createPostFixture({
+				slug: "already-correct",
+				status: "published",
+				data: { title: "Correct" },
+				locale: "zh-TW",
+			}),
+		);
+
+		await up(db);
+
+		const row = await db
+			.selectFrom("ec_post")
+			.select("locale")
+			.where("slug", "=", "already-correct")
+			.executeTakeFirstOrThrow();
+		expect(row.locale).toBe("zh-TW");
+	});
+
+	it("does nothing when i18n is not configured", async () => {
+		setI18nConfig(null);
+
+		const repo = new ContentRepository(db);
+		await repo.create(
+			createPostFixture({ slug: "no-i18n", status: "published", data: { title: "No i18n" } }),
+		);
+		await sql`UPDATE ec_post SET locale = 'zh-tw' WHERE slug = 'no-i18n'`.execute(db);
+
+		await up(db);
+
+		const row = await db
+			.selectFrom("ec_post")
+			.select("locale")
+			.where("slug", "=", "no-i18n")
+			.executeTakeFirstOrThrow();
+		expect(row.locale).toBe("zh-tw");
+	});
+});