fix(seo): normalize and safely encode wiki route slug to prevent double-encoding 404s

This commit is contained in:
Canina Automation committed 2026-10-26 16:32:52 +00:00
1 parent 721118787a
commit e34156b2d3
4 files changed
+291 -10

No files matched your search

+22 -10
View File
@@ -6,15 +6,23 @@ import sanitizeHtml from 'sanitize-html';
import { getSeoConfig, formatPageTitle } from "../../../lib/seo";
import { generateMedicalWebPageSchema, generateBreadcrumbSchema, safeJsonLd } from "../../../lib/schema";
import SafeImage from "../../../components/SafeImage";
import { normalizeRouteSlug, safeEncodeSlug } from "../../../lib/slug";
const API_URL = process.env.INTERNAL_API_URL || process.env.NEXT_PUBLIC_API_URL || 'http://localhost:4001/api';
const apiBase = API_URL.endsWith('/api') ? API_URL : `${API_URL.replace(/\/$/, '')}/api`;
function getApiBase(): string {
const apiUrl = process.env.INTERNAL_API_URL || process.env.NEXT_PUBLIC_API_URL || 'http://localhost:4001/api';
return apiUrl.endsWith('/api') ? apiUrl : `${apiUrl.replace(/\/$/, '')}/api`;
}
async function getWikiTerm(rawKey: string) {
const normalizedKey = normalizeRouteSlug(rawKey);
if (!normalizedKey) return null;
const encodedKey = safeEncodeSlug(normalizedKey);
const apiBase = getApiBase();
async function getWikiTerm(key: string) {
try {
// 1. Try scientific terms endpoint
const res = await fetch(`${apiBase}/wiki/${encodeURIComponent(key)}`, {
next: { revalidate: 3600, tags: ['wiki', `wiki-${key}`] }
const res = await fetch(`${apiBase}/wiki/${encodedKey}`, {
next: { revalidate: 3600, tags: ['wiki', `wiki-${encodedKey}`] }
});
if (res.ok) {
return await res.json();
@@ -25,8 +33,8 @@ async function getWikiTerm(key: string) {
try {
// 2. Fallback to ingredients endpoint if slug is an ingredient slug
const ingRes = await fetch(`${apiBase}/ingredients/${encodeURIComponent(key)}`, {
next: { revalidate: 3600, tags: ['ingredients', `ing-${key}`] }
const ingRes = await fetch(`${apiBase}/ingredients/${encodedKey}`, {
next: { revalidate: 3600, tags: ['ingredients', `ing-${encodedKey}`] }
});
if (ingRes.ok) {
const ing = await ingRes.json();
@@ -48,6 +56,7 @@ async function getWikiTerm(key: string) {
async function getRelatedProducts(termName: string) {
try {
const apiBase = getApiBase();
const res = await fetch(`${apiBase}/products?search=${encodeURIComponent(termName)}&limit=4`, {
next: { revalidate: 3600, tags: ['products'] },
});
@@ -62,6 +71,7 @@ async function getRelatedProducts(termName: string) {
export async function generateMetadata({ params }: { params: Promise<{ slug: string }> }): Promise<Metadata> {
const { slug } = await params;
const normalizedKey = normalizeRouteSlug(slug);
const term = await getWikiTerm(slug);
const config = await getSeoConfig();
@@ -77,7 +87,7 @@ export async function generateMetadata({ params }: { params: Promise<{ slug: str
term.metaDescription ||
(term.definition ? term.definition.replace(/<[^>]+>/g, '').substring(0, 155).trim() : '') ||
`بررسی خواص بالینی، فواید و مکمل‌های دارویی حاوی ${term.term} در دامپزشکی کنینا آلمان.`;
const canonicalUrl = `${config.canonicalBaseUrl.replace(/\/$/, '')}/wiki/${slug}`;
const canonicalUrl = `${config.canonicalBaseUrl.replace(/\/$/, '')}/wiki/${safeEncodeSlug(normalizedKey)}`;
return {
title,
@@ -102,6 +112,7 @@ export async function generateMetadata({ params }: { params: Promise<{ slug: str
export default async function WikiTermPage({ params }: { params: Promise<{ slug: string }> }) {
const { slug } = await params;
const normalizedKey = normalizeRouteSlug(slug);
const term = await getWikiTerm(slug);
if (!term) {
@@ -109,18 +120,19 @@ export default async function WikiTermPage({ params }: { params: Promise<{ slug:
}
const relatedProducts = await getRelatedProducts(term.term);
const encodedSlug = safeEncodeSlug(normalizedKey);
const medicalSchema = generateMedicalWebPageSchema({
title: `ترکیب درمانی ${term.term} - دانشنامه تخصصی کنینا`,
description: term.metaDescription || (term.definition ? term.definition.replace(/<[^>]+>/g, '').substring(0, 160) : ''),
url: `https://canina.ir/wiki/${slug}`,
url: `https://canina.ir/wiki/${encodedSlug}`,
aspects: [term.term, 'Veterinary Supplements', 'Pet Nutrition'],
});
const breadcrumbs = generateBreadcrumbSchema([
{ name: 'صفحه اصلی', url: 'https://canina.ir' },
{ name: 'دانشنامه ترکیبات', url: 'https://canina.ir/wiki' },
{ name: term.term, url: `https://canina.ir/wiki/${slug}` },
{ name: term.term, url: `https://canina.ir/wiki/${encodedSlug}` },
]);
return (
@@ -0,0 +1,127 @@
import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
import React from 'react';
import WikiTermPage, { generateMetadata } from './page';
vi.mock('../../../lib/seo', () => ({
getSeoConfig: vi.fn().mockResolvedValue({
brandNameFa: 'کنینا ایران',
canonicalBaseUrl: 'https://canina.ir',
ogImageUrl: '/assets/og.jpg',
defaultMetaDescription: 'توضیحات دانشنامه',
titleSeparator: '|',
titlePosition: 'suffix',
}),
formatPageTitle: vi.fn((title) => `${title} | کنینا ایران`),
}));
const mockNotFound = vi.fn();
vi.mock('next/navigation', () => ({
notFound: () => {
mockNotFound();
throw new Error('NEXT_NOT_FOUND');
},
}));
describe('WikiTermPage slug decoding and single-encoding', () => {
beforeEach(() => {
vi.clearAllMocks();
vi.stubEnv('INTERNAL_API_URL', 'http://backend:3000/api');
vi.stubEnv('NEXT_PUBLIC_SITE_URL', 'https://canina.ir');
});
afterEach(() => {
vi.unstubAllGlobals();
vi.unstubAllEnvs();
});
it('correctly fetches Persian term when slug is passed URL-encoded (آغوز)', async () => {
const fetchMock = vi.fn()
// First call: /wiki/%D8%A2%D8%BA%D9%88%D8%B2
.mockResolvedValueOnce({
ok: true,
json: async () => ({
key: 'آغوز',
term: 'آغوز (Colostrum)',
definition: 'اولین شیر پستانداران پس از زایمان.',
wikiId: 'colostrum',
}),
})
// Second call: /products?search=...
.mockResolvedValueOnce({
ok: true,
json: async () => ({ data: [] }),
});
vi.stubGlobal('fetch', fetchMock);
// Incoming params from Next router for /wiki/%D8%A2%D8%BA%D9%88%D8%B2
const jsx = await WikiTermPage({
params: Promise.resolve({ slug: '%D8%A2%D8%BA%D9%88%D8%B2' }),
});
expect(jsx).toBeDefined();
expect(fetchMock).toHaveBeenCalled();
const firstCallUrl = fetchMock.mock.calls[0][0];
// Must NOT be double-encoded %25D8%25A2...
expect(firstCallUrl).toBe('http://backend:3000/api/wiki/%D8%A2%D8%BA%D9%88%D8%B2');
expect(firstCallUrl).not.toContain('%25');
});
it('correctly generates metadata with single-encoded canonical for Persian slug', async () => {
const fetchMock = vi.fn().mockResolvedValueOnce({
ok: true,
json: async () => ({
key: 'آغوز',
term: 'آغوز (Colostrum)',
definition: 'اولین شیر پستانداران پس از زایمان.',
wikiId: 'colostrum',
}),
});
vi.stubGlobal('fetch', fetchMock);
const meta = await generateMetadata({
params: Promise.resolve({ slug: '%D8%A2%D8%BA%D9%88%D8%B2' }),
});
expect(meta.alternates?.canonical).toBe('https://canina.ir/wiki/%D8%A2%D8%BA%D9%88%D8%B2');
expect(meta.title).toBe('ترکیب دارویی و خواص آغوز (Colostrum) | کنینا ایران');
});
it('handles EPA/DHA without turning into malformed path or double encoding', async () => {
const fetchMock = vi.fn()
.mockResolvedValueOnce({
ok: true,
json: async () => ({
key: 'EPA/DHA',
term: 'EPA/DHA',
definition: 'اسیدهای چرب امگا ۳.',
wikiId: 'epa-dha',
}),
})
.mockResolvedValueOnce({
ok: true,
json: async () => ({ data: [] }),
});
vi.stubGlobal('fetch', fetchMock);
const jsx = await WikiTermPage({
params: Promise.resolve({ slug: 'EPA%2FDHA' }),
});
expect(jsx).toBeDefined();
expect(fetchMock.mock.calls[0][0]).toBe('http://backend:3000/api/wiki/EPA%2FDHA');
});
it('returns notFound() when neither scientific term nor ingredient exists', async () => {
const fetchMock = vi.fn()
// 1. wiki term 404
.mockResolvedValueOnce({ ok: false, status: 404 })
// 2. ingredient fallback 404
.mockResolvedValueOnce({ ok: false, status: 404 });
vi.stubGlobal('fetch', fetchMock);
await expect(
WikiTermPage({ params: Promise.resolve({ slug: 'non-existent' }) })
).rejects.toThrow('NEXT_NOT_FOUND');
expect(mockNotFound).toHaveBeenCalled();
});
});
+85
View File
@@ -0,0 +1,85 @@
import { describe, it, expect } from 'vitest';
import { normalizeRouteSlug, safeEncodeSlug } from './slug';
describe('normalizeRouteSlug', () => {
it('normalizes single-encoded Persian slug to plain decoded string', () => {
// آغوز encoded once
const input = '%D8%A2%D8%BA%D9%88%D8%B2';
expect(normalizeRouteSlug(input)).toBe('آغوز');
});
it('normalizes already-decoded Persian slug without changes', () => {
const input = 'آغوز';
expect(normalizeRouteSlug(input)).toBe('آغوز');
});
it('normalizes Persian spaced slug', () => {
// آب پنیر
expect(normalizeRouteSlug('%D8%A2%D8%A8%20%D9%BE%D9%86%DB%8C%D8%B1')).toBe('آب پنیر');
expect(normalizeRouteSlug('آب پنیر')).toBe('آب پنیر');
});
it('handles ASCII slugs safely', () => {
expect(normalizeRouteSlug('HACCP')).toBe('HACCP');
expect(normalizeRouteSlug('IFS')).toBe('IFS');
expect(normalizeRouteSlug('BARF')).toBe('BARF');
});
it('handles slash in slug such as EPA/DHA without turning into multiple path segments or breaking', () => {
expect(normalizeRouteSlug('EPA%2FDHA')).toBe('EPA/DHA');
expect(normalizeRouteSlug('EPA/DHA')).toBe('EPA/DHA');
});
it('handles double-encoded slug safely without double-decoding legit single %', () => {
// If incoming slug was double-encoded: %25D8%25A2%25D8%25BA%25D9%2588%25D8%25B2
expect(normalizeRouteSlug('%25D8%25A2%25D8%25BA%25D9%2588%25D8%25B2')).toBe('آغوز');
});
it('handles malformed percent encoding gracefully without throwing URIError', () => {
expect(normalizeRouteSlug('100%pure')).toBe('100%pure');
expect(normalizeRouteSlug('%E0%A4%A')).toBe('%E0%A4%A');
expect(normalizeRouteSlug('%')).toBe('%');
expect(normalizeRouteSlug('test%8')).toBe('test%8');
});
it('handles Ca:P ratio slug', () => {
// نسبت Ca:P
expect(normalizeRouteSlug('%D9%86%D8%B3%D8%A8%D8%AA%20Ca%3AP')).toBe('نسبت Ca:P');
});
});
describe('safeEncodeSlug', () => {
it('encodes slug exactly once for API and URL consumption', () => {
expect(safeEncodeSlug('آغوز')).toBe('%D8%A2%D8%BA%D9%88%D8%B2');
expect(safeEncodeSlug('HACCP')).toBe('HACCP');
expect(safeEncodeSlug('EPA/DHA')).toBe('EPA%2FDHA');
});
});
describe('Edge cases: legitimate single % not double-decoded', () => {
it('preserves slug with 100% discount or pure keyword without destroying %', () => {
// A slug that literally has "100%pure" (malformed percent)
expect(normalizeRouteSlug('100%pure')).toBe('100%pure');
// If someone literally passed 100%25pure (100%pure encoded)
expect(normalizeRouteSlug('100%25pure')).toBe('100%pure');
});
it('handles all sitemap sample slugs cleanly', () => {
const samples = [
'%D8%A2%D8%BA%D9%88%D8%B2',
'%D8%A2%D8%A8%20%D9%BE%D9%86%DB%8C%D8%B1',
'HACCP',
'IFS',
'BARF',
'EPA%2FDHA',
'%D9%86%D8%B3%D8%A8%D8%AA%20Ca%3AP',
'%D8%A8%DB%8C%D9%88%D8%AA%DB%8C%D9%86',
];
for (const sample of samples) {
const normalized = normalizeRouteSlug(sample);
expect(normalized).toBeTruthy();
const reencoded = safeEncodeSlug(normalized);
expect(normalizeRouteSlug(reencoded)).toBe(normalized);
}
});
});
+57
View File
@@ -0,0 +1,57 @@
/**
* Safe slug normalization and encoding for SEO routes.
*
* Handles:
* - Single-encoded Persian/UTF-8 (%D8%A2%D8%BA%D9%88%D8%B2 -> آغوز)
* - Double-encoded slugs (%25D8%25A2... -> آغوز)
* - Raw Unicode (آغوز -> آغوز)
* - Slugs with spaces or encoded spaces (آب پنیر / %D8%A2%D8%A8%20%D9%BE%D9%86%DB%8C%D8%B1)
* - Slashes like EPA/DHA (EPA%2FDHA -> EPA/DHA)
* - Malformed percent sequences without throwing URIError (e.g. 100%pure, %E0%A4%A)
* - Safe single-encoding for outgoing API requests and canonical URLs.
*/
function safeDecode(val: string): string {
try {
return decodeURIComponent(val);
} catch {
return val;
}
}
/**
* Normalizes an incoming route parameter slug to a clean, decoded UTF-8 string.
* Decodes iteratively (up to 2 passes) if double-encoded, but does not alter strings
* where percent is not a valid percent-encoding sequence.
*/
export function normalizeRouteSlug(rawSlug: string): string {
if (!rawSlug) return '';
let current = String(rawSlug).trim();
// If input contains '%', attempt safe decoding
if (current.includes('%')) {
const decodedOnce = safeDecode(current);
if (decodedOnce !== current) {
// Check if it was double-encoded (e.g. %25D8... -> %D8... -> Persian)
if (decodedOnce.includes('%')) {
const decodedTwice = safeDecode(decodedOnce);
// Only accept second decode if it produced a valid string that differs
if (decodedTwice !== decodedOnce) {
return decodedTwice;
}
}
return decodedOnce;
}
}
return current;
}
/**
* Encodes a normalized slug safely for outgoing HTTP query/path or canonical URL.
*/
export function safeEncodeSlug(slug: string): string {
if (!slug) return '';
return encodeURIComponent(slug);
}