From 6ca2f66003e4d75f851db4b0ea191a3486046303 Mon Sep 17 00:00:00 2001 From: DIYgod Date: Mon, 7 Sep 2026 13:27:18 +0800 Subject: [PATCH] fix(worker): close shared cache and route compatibility gaps --- lib/routes/2048/index.tsx | 14 +- lib/routes/2048/utils.ts | 15 ++ lib/routes/aiea/index.ts | 11 +- lib/routes/airchina/index.ts | 20 +-- lib/routes/brooklynmuseum/exhibitions.ts | 6 +- lib/routes/eastmoney/report/index.tsx | 9 +- lib/routes/gov/zhengce/zhengceku.ts | 12 +- lib/routes/hellogithub/index.ts | 4 +- lib/routes/i-cable/news.tsx | 4 +- lib/routes/iiilab/index.ts | 22 +-- lib/routes/javbus/index.tsx | 14 +- lib/routes/jewishmuseum/exhibitions.ts | 4 +- lib/routes/newmuseum/exhibitions.ts | 6 +- lib/routes/oct0pu5/rss.ts | 16 +- lib/routes/polimi/news.ts | 15 +- lib/routes/queshu/book.ts | 12 +- lib/routes/queshu/sale.ts | 20 ++- lib/routes/readhub/daily.ts | 2 +- lib/routes/readhub/index.ts | 4 +- lib/routes/readhub/templates/description.tsx | 9 +- lib/routes/readhub/util.ts | 103 ++++++++---- lib/routes/smzdm/keyword.ts | 13 +- lib/routes/smzdm/utils.ts | 13 ++ lib/routes/stratechery/index.ts | 12 +- lib/routes/twitter/api/web-api/api.ts | 81 +++++---- lib/routes/weibo/oasis/user.ts | 20 ++- lib/routes/xiaoyuzhou/podcast.ts | 23 ++- lib/routes/xunhupay/index.ts | 22 +-- lib/shims/xxhash-wasm.ts | 80 --------- lib/utils/cache/http.test.ts | 14 ++ lib/utils/cache/http.ts | 11 +- lib/utils/common-config.test.ts | 153 +++++------------ lib/utils/common-config.ts | 89 ++++------ lib/utils/parse-date-in-timezone.test.ts | 32 ++++ lib/utils/parse-date-in-timezone.ts | 14 ++ tests/2048-utils.test.ts | 21 +++ tests/common-config-routes.test.ts | 141 ++++++++++++++++ tests/common-config.worker.test.ts | 40 +++++ tests/fixtures/readhub/timeline.json | 9 + tests/fixtures/readhub/topic.json | 22 +++ tests/readhub.test.ts | 163 +++++++++++++++++++ tests/readhub.worker.test.ts | 2 + tests/source-date-routes.test.ts | 104 ++++++++++++ tests/twitter-web-user.test.ts | 151 +++++++++++++++++ tests/twitter-web-user.worker.test.ts | 2 + tests/xiaoyuzhou-podcast.test.ts | 40 +++++ tests/xxhash-wasm.test.ts | 48 ++++++ tests/xxhash-wasm.worker.test.ts | 2 + tsdown-worker.config.ts | 3 +- vitest.workers.config.ts | 3 +- wrangler.toml | 2 +- 51 files changed, 1193 insertions(+), 459 deletions(-) create mode 100644 lib/routes/2048/utils.ts delete mode 100644 lib/shims/xxhash-wasm.ts create mode 100644 lib/utils/parse-date-in-timezone.test.ts create mode 100644 lib/utils/parse-date-in-timezone.ts create mode 100644 tests/2048-utils.test.ts create mode 100644 tests/common-config-routes.test.ts create mode 100644 tests/common-config.worker.test.ts create mode 100644 tests/fixtures/readhub/timeline.json create mode 100644 tests/fixtures/readhub/topic.json create mode 100644 tests/readhub.test.ts create mode 100644 tests/readhub.worker.test.ts create mode 100644 tests/source-date-routes.test.ts create mode 100644 tests/twitter-web-user.test.ts create mode 100644 tests/twitter-web-user.worker.test.ts create mode 100644 tests/xiaoyuzhou-podcast.test.ts create mode 100644 tests/xxhash-wasm.test.ts create mode 100644 tests/xxhash-wasm.worker.test.ts diff --git a/lib/routes/2048/index.tsx b/lib/routes/2048/index.tsx index b032a471a7..64759ad339 100644 --- a/lib/routes/2048/index.tsx +++ b/lib/routes/2048/index.tsx @@ -6,6 +6,8 @@ import ofetch from '@/utils/ofetch'; import { parseDate } from '@/utils/parse-date'; import timezone from '@/utils/timezone'; +import { getForumUrl } from './utils'; + export const route: Route = { path: '/:id?', categories: ['multimedia'], @@ -64,14 +66,10 @@ async function handler(ctx) { const id = ctx.req.param('id') ?? '3'; const rootUrl = 'https://hjd2048.com'; - // 获取地址发布页指向的 URL - const domainInfo = await cache.tryGet('2048:domainInfo', async () => { + // Resolve the address page before caching the forum URL. + const domainInfo = await cache.tryGet('2048:domainInfo:v2', async () => { const response = await ofetch('https://2048.info'); - const $ = load(response); - const onclickValue = $('.button').first().attr('onclick'); - const targetUrl = onclickValue?.match(/window\.open\('([^']+)'/)?.[1]; - - return { url: new URL(targetUrl!, 'https://2048.info').href }; + return { url: getForumUrl(response) }; }); // 获取重定向后的url const redirectResponse = await ofetch.raw(domainInfo.url); @@ -91,7 +89,7 @@ async function handler(ctx) { 86400, // fixed cookie duration: 24 hours false ); - const currentUrl = `${redirected.url}thread.php?fid-${id}.html`; + const currentUrl = new URL(`thread.php?fid-${id}.html`, redirected.url).href; const response = await ofetch.raw(currentUrl, { headers: { diff --git a/lib/routes/2048/utils.ts b/lib/routes/2048/utils.ts new file mode 100644 index 0000000000..94f652b48b --- /dev/null +++ b/lib/routes/2048/utils.ts @@ -0,0 +1,15 @@ +import { load } from 'cheerio'; + +export function getForumUrl(html: string) { + const $ = load(html); + const button = $('.button').first(); + const target = button.attr('href') || button.attr('onclick')?.match(/window\.open\(\s*(['"])(.*?)\1/)?.[2]; + if (!target) { + throw new Error('The 2048 address page did not contain a forum link.'); + } + const url = new URL(target, 'https://2048.info'); + if (!['https:', 'http:'].includes(url.protocol)) { + throw new Error('The 2048 address page contained an unsupported forum URL.'); + } + return url.href; +} diff --git a/lib/routes/aiea/index.ts b/lib/routes/aiea/index.ts index 630e676f97..26e1060c2f 100644 --- a/lib/routes/aiea/index.ts +++ b/lib/routes/aiea/index.ts @@ -49,15 +49,12 @@ async function handler(ctx) { return await buildData({ link, url: link, - title: '%title%', - params: { - title: 'AIEA Seminars', - }, + title: 'AIEA Seminars', item: { item: `.seminar-contents .seminar-partWrap:nth-child(${nth_child}) > .seminar-list`, - title: `$('.seminar-list-title > span').text()`, - link: `$('a[href^="/0504"]').attr('href')`, - description: `$('.seminar-list .txt > .title').text()`, + title: ($) => $('.seminar-list-title > span').text(), + link: ($) => $('a[href^="/0504"]').attr('href'), + description: ($) => $('.seminar-list .txt > .title').text(), }, }); } diff --git a/lib/routes/airchina/index.ts b/lib/routes/airchina/index.ts index f6bbf25da6..a126683559 100644 --- a/lib/routes/airchina/index.ts +++ b/lib/routes/airchina/index.ts @@ -4,6 +4,7 @@ import type { Route } from '@/types'; import cache from '@/utils/cache'; import buildData from '@/utils/common-config'; import got from '@/utils/got'; +import { parseDate } from '@/utils/parse-date'; const baseUrl = 'https://www.airchina.com.cn'; @@ -36,18 +37,17 @@ async function handler() { const data = await buildData({ link, url: link, - title: '%title%', - description: '%description%', - params: { - title: '国航服务公告', - description: '中国国际航空公司服务公告', - }, + title: '国航服务公告', + description: '中国国际航空公司服务公告', item: { item: '.serviceMsg li', - title: `$('a').text()`, - link: `$('a').attr('href')`, - pubDate: `parseDate($('span').text(), 'YYYY-MM-DD')`, - guid: Buffer.from(`$('a').attr('href')`).toString('base64'), + title: ($) => $('a').text(), + link: ($) => $('a').attr('href'), + pubDate: ($) => parseDate($('span').text(), 'YYYY-MM-DD'), + guid: ($) => { + const href = $('a').attr('href'); + return href ? Buffer.from(href).toString('base64') : undefined; + }, }, }); diff --git a/lib/routes/brooklynmuseum/exhibitions.ts b/lib/routes/brooklynmuseum/exhibitions.ts index 0c55e9efb1..64e0676919 100644 --- a/lib/routes/brooklynmuseum/exhibitions.ts +++ b/lib/routes/brooklynmuseum/exhibitions.ts @@ -38,9 +38,9 @@ async function handler(ctx) { title: 'Brooklyn Museum - Exhibitions', item: { item: '.exhibitions .image-card', - title: `$('h2 > a, h3 > a').text()`, - link: `$('h2 > a, h3 > a').attr('href')`, - description: `$('h6').text()`, + title: ($) => $('h2 > a, h3 > a').text(), + link: ($) => $('h2 > a, h3 > a').attr('href'), + description: ($) => $('h6').text(), }, }); } diff --git a/lib/routes/eastmoney/report/index.tsx b/lib/routes/eastmoney/report/index.tsx index a85e63967d..7ab2d84ef3 100644 --- a/lib/routes/eastmoney/report/index.tsx +++ b/lib/routes/eastmoney/report/index.tsx @@ -6,7 +6,7 @@ import type { Route } from '@/types'; import { ViewType } from '@/types'; import cache from '@/utils/cache'; import got from '@/utils/got'; -import { parseDate } from '@/utils/parse-date'; +import { parseDateInTimezone } from '@/utils/parse-date-in-timezone'; import { getEpsOrPeStr, getRatingChangeStr } from '../utils'; @@ -81,18 +81,18 @@ async function handler(ctx) { return { title: `[${item.orgSName}]${stockName}${item.title}`, link: `${baseUrl}/report/${linkType[category]}` + (category === 'stock' ? `/${item.infoCode}.html` : `.jshtml?encodeUrl=${item.encodeUrl}`), - pubDate: parseDate(item.publishDate), + pubDate: parseDateInTimezone(item.publishDate, 8), author: item.researcher, originItem: item, // temp use }; }); const items = await Promise.all( - list.map((item) => { + list.map(async (item) => { const tempOriginItem = item.originItem; delete item.originItem; // temp use - return cache.tryGet(item.link, async () => { + const cached = await cache.tryGet(item.link, async () => { try { const { data: response } = await got(item.link); const $ = load(response); @@ -158,6 +158,7 @@ async function handler(ctx) { return item; } }); + return { ...cached, pubDate: item.pubDate }; }) ); diff --git a/lib/routes/gov/zhengce/zhengceku.ts b/lib/routes/gov/zhengce/zhengceku.ts index feacba3be2..caf83404a9 100644 --- a/lib/routes/gov/zhengce/zhengceku.ts +++ b/lib/routes/gov/zhengce/zhengceku.ts @@ -1,5 +1,6 @@ import type { Route } from '@/types'; import buildData from '@/utils/common-config'; +import { parseDate } from '@/utils/parse-date'; export const route: Route = { path: '/zhengceku/:department', @@ -26,16 +27,13 @@ async function handler(ctx) { return await buildData({ link, url: link, - title: '%title%', + title: ($) => $('.channel_tab > .noline > a').text().trim() + ' - 政府文件库', description: '政府文件库, 当页的所有列表', - params: { - title: `$('.channel_tab > .noline > a').text().trim() + ' - 政府文件库'`, - }, item: { item: '.news_box > .list > ul > li:not(.line)', - title: `$('h4 > a').text()`, - link: `$('h4 > a').attr('href')`, - pubDate: `parseDate($('h4 > .date').text().trim())`, + title: ($) => $('h4 > a').text(), + link: ($) => $('h4 > a').attr('href'), + pubDate: ($) => parseDate($('h4 > .date').text().trim()), }, }); } diff --git a/lib/routes/hellogithub/index.ts b/lib/routes/hellogithub/index.ts index 52c625d50b..c0f17b90ec 100644 --- a/lib/routes/hellogithub/index.ts +++ b/lib/routes/hellogithub/index.ts @@ -2,7 +2,7 @@ import { load } from 'cheerio'; import type { Route } from '@/types'; import got from '@/utils/got'; -import { parseDate } from '@/utils/parse-date'; +import { parseDateInTimezone } from '@/utils/parse-date-in-timezone'; const sorts = { featured: '精选', @@ -64,7 +64,7 @@ async function handler(ctx) { title: `${item.name}: ${item.title}`, author: item.author, link: `${rootUrl}/repository/${item.item_id}`, - pubDate: parseDate(item.updated_at), + pubDate: parseDateInTimezone(item.updated_at, 8), name: `${item.author}/${item.name}`, description: item.summary, language: item.primary_lang, diff --git a/lib/routes/i-cable/news.tsx b/lib/routes/i-cable/news.tsx index 7190694b2a..64eea03fc8 100644 --- a/lib/routes/i-cable/news.tsx +++ b/lib/routes/i-cable/news.tsx @@ -6,6 +6,7 @@ import InvalidParameterError from '@/errors/types/invalid-parameter'; import type { Route } from '@/types'; import cache from '@/utils/cache'; import got from '@/utils/got'; +import { parseDateInTimezone } from '@/utils/parse-date-in-timezone'; export const route: Route = { path: '/news/:category?', @@ -67,7 +68,8 @@ async function handler(ctx) { return { title: item.title.rendered, link: item.link, - pubDate: item.date_gmt, + // WordPress date_gmt is UTC even when its string omits the Z suffix. + pubDate: item.date_gmt ? parseDateInTimezone(item.date_gmt, 0) : undefined, description, category: item._embedded['wp:term'][0].map((term) => term.name) ?? [], }; diff --git a/lib/routes/iiilab/index.ts b/lib/routes/iiilab/index.ts index 0c773ea294..3b5512f885 100644 --- a/lib/routes/iiilab/index.ts +++ b/lib/routes/iiilab/index.ts @@ -1,5 +1,6 @@ import type { Route } from '@/types'; import buildData from '@/utils/common-config'; +import { parseDate } from '@/utils/parse-date'; const baseUrl = 'https://www.iiilab.com/'; @@ -24,19 +25,18 @@ async function handler() { return await buildData({ link, url: link, - title: '%title%', - description: '%description%', - params: { - title: '发现', - description: '人人都是自媒体-发现', - }, + title: '发现', + description: '人人都是自媒体-发现', item: { item: '.aw-common-list > div', - title: `$('a').first().text()`, - link: `$('a').first().attr('href')`, - description: `$('.markitup-box').text()`, - pubDate: `parseDate($('.text-color-999').first().text(), 'YYYY-MM-DD HH:mm')`, - guid: Buffer.from(`$('a').attr('href')`).toString('base64'), + title: ($) => $('a').first().text(), + link: ($) => $('a').first().attr('href'), + description: ($) => $('.markitup-box').text(), + pubDate: ($) => parseDate($('.text-color-999').first().text(), 'YYYY-MM-DD HH:mm'), + guid: ($) => { + const href = $('a').attr('href'); + return href ? Buffer.from(href).toString('base64') : undefined; + }, }, }); } diff --git a/lib/routes/javbus/index.tsx b/lib/routes/javbus/index.tsx index add5223136..c037c4e187 100644 --- a/lib/routes/javbus/index.tsx +++ b/lib/routes/javbus/index.tsx @@ -9,7 +9,7 @@ import { ViewType } from '@/types'; import cache from '@/utils/cache'; import { getSubPath } from '@/utils/common-utils'; import got from '@/utils/got'; -import { parseDate } from '@/utils/parse-date'; +import { parseDateInTimezone } from '@/utils/parse-date-in-timezone'; const toSize = (raw) => { const matches = raw.match(/(\d+(\.\d+)?)(\D\w*)/); @@ -109,14 +109,15 @@ async function handler(ctx) { return { link: $item.attr('href'), guid: $item.find('date').first().text(), - pubDate: parseDate($item.find('date').last().text()), + // This is a calendar release date, represented at UTC midnight. + pubDate: parseDateInTimezone($item.find('date').last().text(), 0), title: '', }; }); items = await Promise.all( - items.map((item) => - cache.tryGet(item.link!, async () => { + items.map(async (item) => { + const cached = await cache.tryGet(item.link!, async () => { const detailResponse = await got({ method: 'get', url: item.link, @@ -206,8 +207,9 @@ async function handler(ctx) { }); return item; - }) - ) + }); + return { ...cached, pubDate: item.pubDate }; + }) ); const title = $('head title').text(); diff --git a/lib/routes/jewishmuseum/exhibitions.ts b/lib/routes/jewishmuseum/exhibitions.ts index e92b79931e..b06a7408d4 100644 --- a/lib/routes/jewishmuseum/exhibitions.ts +++ b/lib/routes/jewishmuseum/exhibitions.ts @@ -28,8 +28,8 @@ async function handler() { title: 'Jewish Museums - Exhibitions', item: { item: '#current article.exhibition, #upcoming article, #past article.exhibition', - title: `$('h3').text()`, - link: `$('h3').parent().attr('href')`, + title: ($) => $('h3').text(), + link: ($) => $('h3').parent().attr('href'), }, }); } diff --git a/lib/routes/newmuseum/exhibitions.ts b/lib/routes/newmuseum/exhibitions.ts index 975dac84a5..769f0c1485 100644 --- a/lib/routes/newmuseum/exhibitions.ts +++ b/lib/routes/newmuseum/exhibitions.ts @@ -38,9 +38,9 @@ async function handler(ctx) { title: 'New Museum - Exhibitions', item: { item: '.exh', - title: `$('.exh .title').text()`, - link: `$('.exh > a').attr('href')`, - description: `$('.exh .body-reveal').text()`, + title: ($) => $('.exh .title').text(), + link: ($) => $('.exh > a').attr('href'), + description: ($) => $('.exh .body-reveal').text(), }, }); } diff --git a/lib/routes/oct0pu5/rss.ts b/lib/routes/oct0pu5/rss.ts index 48bea0bd77..7d4d3cbec5 100644 --- a/lib/routes/oct0pu5/rss.ts +++ b/lib/routes/oct0pu5/rss.ts @@ -31,18 +31,14 @@ async function handler() { return await buildData({ link, url: link, - title: '%title%', - description: '%description%', - params: { - title: '博客', - description: 'Oct0pu5的博客', - }, + title: '博客', + description: 'Oct0pu5的博客', item: { item: '.recent-posts > .recent-post-item', - title: `$('.recent-post-info > a').text()`, - link: `$('.recent-post-info > a').attr('href')`, - description: `$('.recent-post-info > .content').text()`, - pubDate: `Date.parse($('div.recent-post-info > div.article-meta-wrap > span.post-meta-date > time').text().trim())`, + title: ($) => $('.recent-post-info > a').text(), + link: ($) => $('.recent-post-info > a').attr('href'), + description: ($) => $('.recent-post-info > .content').text(), + pubDate: ($) => Date.parse($('div.recent-post-info > div.article-meta-wrap > span.post-meta-date > time').text().trim()), }, }); } diff --git a/lib/routes/polimi/news.ts b/lib/routes/polimi/news.ts index b27909c824..568fee717e 100644 --- a/lib/routes/polimi/news.ts +++ b/lib/routes/polimi/news.ts @@ -2,6 +2,7 @@ import type { Context } from 'hono'; import type { Route } from '@/types'; import buildData from '@/utils/common-config'; +import { parseDate } from '@/utils/parse-date'; export const route: Route = { path: '/news/:language?', @@ -21,15 +22,15 @@ async function handler(ctx: Context) { link, url: link, title: 'Polimi News', - params: { - homeLink: 'https://www.polimi.it', - }, item: { item: '.card--editorial-photo', - title: `$('.card-title').text()`, - link: `'%homeLink%' + $('.card-footer a').attr('href')`, - description: `$('.news-bodytext').html()`, - pubDate: `parseDate($('time').attr('datetime'))`, + title: ($) => $('.card-title').text(), + link: ($) => 'https://www.polimi.it' + $('.card-footer a').attr('href'), + description: ($) => $('.news-bodytext').html(), + pubDate: ($) => { + const datetime = $('time').attr('datetime'); + return datetime ? parseDate(datetime) : undefined; + }, }, }); } diff --git a/lib/routes/queshu/book.ts b/lib/routes/queshu/book.ts index 36dc5137b3..c44ddc47b4 100644 --- a/lib/routes/queshu/book.ts +++ b/lib/routes/queshu/book.ts @@ -23,17 +23,13 @@ async function handler(ctx: Context) { return await buildData({ link, url: link, - title: '%title%', - description: `$('#detail_intro .detail_body').text().trim()`, + title: ($) => $('#book_left > #h1').text().trim() + ' - 单品活动信息 - 缺书网', + description: ($) => $('#detail_intro .detail_body').text().trim(), allowEmpty: true, - params: { - title: `$('#book_left > #h1').text().trim() + ' - 单品活动信息 - 缺书网'`, - host, - }, item: { item: '.stacked.right_state > a', - title: `$('.right_item .bodycol_258').text()`, - link: `'%host%' + $('.right_item').parent().attr('href')`, + title: ($) => $('.right_item .bodycol_258').text(), + link: ($) => host + $('.right_item').parent().attr('href'), }, }); } diff --git a/lib/routes/queshu/sale.ts b/lib/routes/queshu/sale.ts index 75644d6915..5678f2010d 100644 --- a/lib/routes/queshu/sale.ts +++ b/lib/routes/queshu/sale.ts @@ -17,16 +17,20 @@ async function handler() { return await buildData({ link, url: link, - title: '%title%', - params: { - title: '图书促销 - 缺书网', - host, - }, + title: '图书促销 - 缺书网', item: { item: '#tb_sale tr', - title: `$('.news_sale_detail .sale_btn').text() + '：' + $('.news_sale_title a').text()`, - link: `'%host%' + $('.news_sale_title a').attr('href')`, - description: `$('.news_sale_detail .sale_btn').text() + '：' + $('.news_sale_title a').text() + '
' + $('.news_sale_detail .sale_time_end').first().text() + '
' + '发布时间：' + $('.news_sale_detail .sale_time_end.inline_right').text()`, + title: ($) => $('.news_sale_detail .sale_btn').text() + '：' + $('.news_sale_title a').text(), + link: ($) => host + $('.news_sale_title a').attr('href'), + description: ($) => + $('.news_sale_detail .sale_btn').text() + + '：' + + $('.news_sale_title a').text() + + '
' + + $('.news_sale_detail .sale_time_end').first().text() + + '
' + + '发布时间：' + + $('.news_sale_detail .sale_time_end.inline_right').text(), }, }); } diff --git a/lib/routes/readhub/daily.ts b/lib/routes/readhub/daily.ts index 00e0f8806c..4ef4966e1c 100644 --- a/lib/routes/readhub/daily.ts +++ b/lib/routes/readhub/daily.ts @@ -59,7 +59,7 @@ async function handler(ctx) { return { item: items, title: `${author} - ${route.name}`, - link: currentUrl, + link: infoUrl, description: $('meta[name="description"]').prop('content'), language: 'zh' as const satisfies Language, image, diff --git a/lib/routes/readhub/index.ts b/lib/routes/readhub/index.ts index 55836a9665..9df017ab3f 100644 --- a/lib/routes/readhub/index.ts +++ b/lib/routes/readhub/index.ts @@ -56,9 +56,9 @@ async function handler(ctx) { rootUrl, }), author: item.siteNameDisplay, - category: [...(item.entityList.map((c) => c.name) ?? []), ...(item.tagList.map((c) => c.name) ?? [])], + category: [...(item.entityList?.map((c) => c.name) ?? []), ...(item.tagList?.map((c) => c.name) ?? [])], guid: item.uid, - pubDate: parseDate(item.publishDate), + pubDate: item.publishDate ? parseDate(item.publishDate) : undefined, })); items = await processItems(items); diff --git a/lib/routes/readhub/templates/description.tsx b/lib/routes/readhub/templates/description.tsx index 3c716a0451..cddb48dc2e 100644 --- a/lib/routes/readhub/templates/description.tsx +++ b/lib/routes/readhub/templates/description.tsx @@ -1,6 +1,9 @@ import dayjs from 'dayjs'; +import utc from 'dayjs/plugin/utc.js'; import { renderToString } from 'hono/jsx/dom/server'; +dayjs.extend(utc); + type NewsItem = { url?: string; title?: string; @@ -28,7 +31,7 @@ export const renderDescription = ({ description, news, timeline, rootUrl }: Desc renderToString( <> {description ?

{description}

: null} - {news ? ( + {news?.length ? ( <>

媒体报道

@@ -47,7 +50,7 @@ export const renderDescription = ({ description, news, timeline, rootUrl }: Desc
) : null} - {timeline ? ( + {timeline?.topics?.length ? ( <>

事件追踪

@@ -55,7 +58,7 @@ export const renderDescription = ({ description, news, timeline, rootUrl }: Desc {timeline.topics?.map((topic, index) => (
- {topic.publishDate ? dayjs(topic.publishDate).format('YYYY-MM-DD HH:mm:ss') : ''} + {topic.publishDate ? dayjs(topic.publishDate).utcOffset(8).format('YYYY-MM-DD HH:mm:ss') : ''} {topic.title} diff --git a/lib/routes/readhub/util.ts b/lib/routes/readhub/util.ts index 5f0d4a1836..a7b5fd3dd3 100644 --- a/lib/routes/readhub/util.ts +++ b/lib/routes/readhub/util.ts @@ -1,3 +1,6 @@ +import { z } from 'zod'; + +import type { DataItem } from '@/types'; import cache from '@/utils/cache'; import got from '@/utils/got'; import { parseDate } from '@/utils/parse-date'; @@ -8,45 +11,77 @@ const domain = 'readhub.cn'; const rootUrl = `https://${domain}`; const apiRootUrl = `https://api.${domain}`; const apiTopicUrl = new URL('topic/list', apiRootUrl).href; +const apiDetailUrl = new URL('topic/detail', apiRootUrl).href; +const apiTimelineUrl = new URL('topic/timeline/list', apiRootUrl).href; -/** - * Process items asynchronously. - * - * @param {Array} items - The array of items to process. - * @returns {Promise>} Returns a Promise that resolves to an array of processed items. - */ -const processItems = async (items) => - await Promise.all( - items.map((item) => - cache.tryGet(item.link, async () => { - try { - if (!item.link.startsWith(rootUrl)) { - throw new Error(`"${item.link}" is an external URL`); - } +const namedEntry = z.object({ name: z.string() }); +const newsSchema = z.object({ url: z.url(), title: z.string(), siteNameDisplay: z.string().nullish() }); +const topicSchema = z.object({ + uid: z.string().min(1), + title: z.string().min(1), + summary: z.string().min(1), + url: z.url().nullish(), + publishDate: z.iso.datetime({ offset: true }), + siteNameDisplay: z.string().nullish(), + newsAggList: z.array(newsSchema), + entityList: z.array(namedEntry).nullish(), + tagList: z.array(namedEntry).nullish(), +}); +const timelineEntrySchema = z.object({ uid: z.string().min(1), title: z.string(), publishDate: z.iso.datetime({ offset: true }).nullish() }); +const timelineDataSchema = z.object({ items: z.array(timelineEntrySchema) }); +const topicDataSchema = z.object({ items: z.array(topicSchema) }); +const timelineSchema = z.object({ data: timelineDataSchema }); +const detailSchema = z.object({ data: topicDataSchema }); - const { data: detailResponse } = await got(item.link); +type ReadhubItem = DataItem & { link: string; guid: string }; - const data = JSON.parse(detailResponse.match(/\{\\"topic\\":(.*?)\}\]\\n"\]\)<\/script>/)[1].replaceAll(String.raw`\"`, '"')); +async function fetchTopic(uid: string): Promise { + const { data: detailResponse } = await got(apiDetailUrl, { searchParams: { uid } }); + const detail = detailSchema.safeParse(detailResponse); + if (!detail.success) { + throw new Error('Readhub topic detail response is invalid or missing publication dates'); + } + const matches = detail.data.data.items.filter((topic) => topic.uid === uid); + if (matches.length !== 1) { + throw new Error('Readhub topic detail response does not contain the requested topic'); + } + const topic = matches[0]; - item.title = data.title; - item.link = data.url ?? new URL(`topic/${data.uid}`, rootUrl).href; - item.description = renderDescription({ - description: data.summary, - news: data.newsAggList, - timeline: data.timeline, - rootUrl, - }); - item.author = data.siteNameDisplay; - item.category = [...(data.entityList.map((c) => c.name) ?? []), ...(data.tagList.map((c) => c.name) ?? [])]; - item.guid = `readhub-${data.uid}`; - item.pubDate = parseDate(data.publishDate.replaceAll(/\s/g, '')); - } catch { - item.guid = `readhub-${item.guid}`; - } + const { data: timelineResponse } = await got(apiTimelineUrl, { searchParams: { topic_uid: uid, size: 10 } }); + const timeline = timelineSchema.safeParse(timelineResponse); + if (!timeline.success) { + throw new Error('Readhub topic timeline response is invalid'); + } - return item; - }) - ) + return { + title: topic.title, + link: topic.url ?? new URL(`topic/${topic.uid}`, rootUrl).href, + description: renderDescription({ + description: topic.summary, + news: topic.newsAggList.map((news) => ({ ...news, siteNameDisplay: news.siteNameDisplay ?? undefined })), + timeline: { topics: timeline.data.data.items.map((entry) => ({ ...entry, publishDate: entry.publishDate ?? undefined })) }, + rootUrl, + }), + author: topic.siteNameDisplay ?? undefined, + category: [...(topic.entityList ?? []), ...(topic.tagList ?? [])].map((entry) => entry.name), + guid: `readhub-${topic.uid}`, + pubDate: parseDate(topic.publishDate), + }; +} + +const processItems = (items: ReadhubItem[]): Promise => + Promise.all( + items.map((item) => { + const url = new URL(item.link); + const topicPath = /^\/topic\/([^/]+)\/?$/.exec(url.pathname); + if (url.origin !== rootUrl || !topicPath) { + return { ...item, guid: `readhub-${item.guid}` }; + } + + // Keep failed enrichment out of the cache, including old summary-only entries. + const uid = decodeURIComponent(topicPath[1]); + return cache.tryGet(`readhub:topic:v2:${uid}`, () => fetchTopic(uid)); + }) ); export { apiRootUrl, apiTopicUrl, processItems, rootUrl }; diff --git a/lib/routes/smzdm/keyword.ts b/lib/routes/smzdm/keyword.ts index acf3a09380..69e007b762 100644 --- a/lib/routes/smzdm/keyword.ts +++ b/lib/routes/smzdm/keyword.ts @@ -5,10 +5,8 @@ import ConfigNotFoundError from '@/errors/types/config-not-found'; import type { Route } from '@/types'; import { ViewType } from '@/types'; import got from '@/utils/got'; -import { parseDate } from '@/utils/parse-date'; -import timezone from '@/utils/timezone'; -import { getHeaders } from './utils'; +import { getHeaders, parseSearchDate } from './utils'; export const route: Route = { path: '/keyword/:keyword', @@ -63,14 +61,15 @@ async function handler(ctx) { return { title: `${keyword} - 什么值得买`, link: `https://search.smzdm.com/?c=home&s=${encodeURIComponent(keyword)}&order=time`, - item: - list && - list.toArray().map((item) => { + item: list + .toArray() + .filter((item) => $(item).find('.feed-block-title a').first().attr('href')) + .map((item) => { const $item = $(item); return { title: `${$item.find('.feed-block-title a').eq(0).text().trim()} - ${$item.find('.feed-block-title a').eq(1).text().trim()}`, description: `${$item.find('.feed-block-descripe').contents().eq(2).text().trim()}
${$item.find('.feed-block-extras span').text().trim()}
`, - pubDate: timezone(parseDate($item.find('.feed-block-extras').contents().eq(0).text().trim(), ['MM-DD HH:mm', 'HH:mm']), 8), + pubDate: parseSearchDate($item.find('.feed-block-extras').contents().eq(0).text().trim()), link: $item.find('.feed-block-title a').attr('href'), }; }), diff --git a/lib/routes/smzdm/utils.ts b/lib/routes/smzdm/utils.ts index 005e4fd343..3df7a5b6b7 100644 --- a/lib/routes/smzdm/utils.ts +++ b/lib/routes/smzdm/utils.ts @@ -1,4 +1,17 @@ import { config } from '@/config'; +import { parseDateInTimezone } from '@/utils/parse-date-in-timezone'; + +export const parseSearchDate = (value: string, now = new Date()): Date | undefined => { + const today = new Date(now.getTime() + 8 * 60 * 60 * 1000).toISOString().slice(0, 10); + if (/^\d{2}:\d{2}$/.test(value)) { + return parseDateInTimezone(`${today}T${value}`, 8); + } + if (/^\d{2}-\d{2} \d{2}:\d{2}$/.test(value)) { + // Search results omit the year; a later month/day belongs to last year. + const year = Number(today.slice(0, 4)) - Number(value.slice(0, 5) > today.slice(5)); + return parseDateInTimezone(`${year}-${value}`, 8); + } +}; export const getHeaders = () => ({ accept: 'application/json, text/javascript, */*; q=0.01', diff --git a/lib/routes/stratechery/index.ts b/lib/routes/stratechery/index.ts index 21cdfba085..76498d46d6 100644 --- a/lib/routes/stratechery/index.ts +++ b/lib/routes/stratechery/index.ts @@ -1,5 +1,6 @@ import type { Route } from '@/types'; import buildData from '@/utils/common-config'; +import { parseDate } from '@/utils/parse-date'; export const route: Route = { path: '/', @@ -21,10 +22,13 @@ async function handler() { description: 'Stratechery provides analysis of the strategy and business side of technology and media, and the impact of technology on society. ', item: { item: 'article', - title: `$('article > header > h1 > a').text()`, - link: `$('article > header > h1 > a').attr('href')`, - pubDate: `parseDate($('article .entry-date').attr('datetime'))`, - description: `$('article > .entry-content').html().replace(/%/g, '%')`, + title: ($) => $('article > header > h1 > a').text(), + link: ($) => $('article > header > h1 > a').attr('href'), + pubDate: ($) => { + const datetime = $('article .entry-date').attr('datetime'); + return datetime ? parseDate(datetime) : undefined; + }, + description: ($) => $('article > .entry-content').html()!.replaceAll('%', '%'), }, }); } diff --git a/lib/routes/twitter/api/web-api/api.ts b/lib/routes/twitter/api/web-api/api.ts index 7bfe5c75b0..280db0c7e8 100644 --- a/lib/routes/twitter/api/web-api/api.ts +++ b/lib/routes/twitter/api/web-api/api.ts @@ -8,8 +8,30 @@ import { baseUrl, gqlFeatures, gqlMap, initGqlMap } from './constants'; import type { ApiParams } from './utils'; import { gatherLegacyFromData, paginationTweets, twitterGot } from './utils'; -const getUserData = (id) => - cache.tryGet(`twitter-userdata-${id}`, () => { +const getUserResult = (response: any) => { + const errors = response?.errors; + if (errors !== undefined && errors !== null && (!Array.isArray(errors) || errors.length > 0)) { + const codes = Array.isArray(errors) ? [...new Set(errors.map((error) => error?.code).filter((code) => Number.isSafeInteger(code)))] : []; + throw new Error(`Twitter API user lookup failed${codes.length ? ` (codes: ${codes.join(', ')})` : ''}`); + } + + const user = response?.data?.user; + if (user === null || user?.result === null) { + throw new InvalidParameterError("This account doesn't exist"); + } + const result = user?.result; + if (result?.__typename === 'UserUnavailable') { + throw new InvalidParameterError('Twitter user is unavailable'); + } + if (!result || typeof result.rest_id !== 'string' || result.rest_id.length === 0) { + throw new Error('Twitter API returned an incomplete user response'); + } + return result; +}; + +const getUserData = async (id: string) => { + // Other Twitter API adapters cache different schemas under the legacy key. + const data = await cache.tryGet(`twitter:web:userdata:v2:${id}`, async () => { const params = { variables: id.startsWith('+') ? JSON.stringify({ @@ -26,31 +48,28 @@ const getUserData = (id) => }), }; - if (config.twitter.thirdPartyApi) { - const endpoint = id.startsWith('+') ? gqlMap.UserByRestId : gqlMap.UserByScreenName; + const response = config.twitter.thirdPartyApi + ? await ofetch(`${config.twitter.thirdPartyApi}${id.startsWith('+') ? gqlMap.UserByRestId : gqlMap.UserByScreenName}`, { + method: 'GET', + params, + headers: { + 'accept-encoding': 'gzip', + }, + }) + : await twitterGot(`${baseUrl}${id.startsWith('+') ? gqlMap.UserByRestId : gqlMap.UserByScreenName}`, params, { + allowNoAuth: !id.startsWith('+'), + }); - return ofetch(`${config.twitter.thirdPartyApi}${endpoint}`, { - method: 'GET', - params, - headers: { - 'accept-encoding': 'gzip', - }, - }); - } - - return twitterGot(`${baseUrl}${id.startsWith('+') ? gqlMap.UserByRestId : gqlMap.UserByScreenName}`, params, { - allowNoAuth: !id.startsWith('+'), - }); + // HTTP 200 can still contain GraphQL errors; never cache those responses. + getUserResult(response); + return response; }); + return getUserResult(data); +}; const cacheTryGet = async (_id, params, operationName, func) => { - const userData: any = await getUserData(_id); - const id = userData.data?.user?.result?.rest_id; - if (id === undefined) { - cache.set(`twitter-userdata-${_id}`, '', config.cache.contentExpire); - throw new InvalidParameterError('User not found'); - } - return cache.tryGet(getTwitterUserCacheKey(id, operationName, params), () => func(id, params), config.cache.routeExpire, false); + const user = await getUserData(_id); + return cache.tryGet(getTwitterUserCacheKey(user.rest_id, operationName, params), () => func(user.rest_id, params), config.cache.routeExpire, false); }; const getUserTweets = (id: string, params?: ApiParams) => @@ -164,19 +183,11 @@ const getList = async (id: string, params?: ApiParams) => ); const getUser = async (id: string) => { - const userData: any = await getUserData(id); - - if (!userData.data.user) { - throw new InvalidParameterError("This account doesn't exist"); - } - if (userData.data.user.result.__typename === 'UserUnavailable') { - throw new InvalidParameterError(userData.data.user.result.message || 'User is unavailable'); - } - + const user = await getUserData(id); return { - profile_image_url: userData.data?.user?.result?.avatar?.image_url, - description: userData.data?.user?.result?.profile_bio?.description, - ...userData.data?.user?.result?.core, + profile_image_url: user.avatar?.image_url, + description: user.profile_bio?.description, + ...user.core, }; }; diff --git a/lib/routes/weibo/oasis/user.ts b/lib/routes/weibo/oasis/user.ts index 078c3346b9..47258066a3 100644 --- a/lib/routes/weibo/oasis/user.ts +++ b/lib/routes/weibo/oasis/user.ts @@ -34,13 +34,23 @@ async function handler(ctx) { await buildData({ link, url: link, - title: `$('.name-main').text().trim() + ' - 用户 - 绿洲'`, - description: `$('.desc').text().trim()`, + title: ($) => $('.name-main').text().trim() + ' - 用户 - 绿洲', + description: ($) => $('.desc').text().trim(), item: { item: '.container .status-item', - title: `$('.status-item-title').contents().filter((_, node) => node.type === 'text').text()`, - description: `$('.status-item-title').contents().filter((_, node) => node.type === 'text').text() + '
' + $('.status-img').html()`, - link: `'https://oasis.weibo.cn/v1/h5/share?sid=' + $('.status-item-title').parent().data('id')`, + title: ($) => + $('.status-item-title') + .contents() + .filter((_, node) => node.type === 'text') + .text(), + description: ($) => + $('.status-item-title') + .contents() + .filter((_, node) => node.type === 'text') + .text() + + '
' + + $('.status-img').html(), + link: ($) => 'https://oasis.weibo.cn/v1/h5/share?sid=' + $('.status-item-title').parent().data('id'), }, }) ); diff --git a/lib/routes/xiaoyuzhou/podcast.ts b/lib/routes/xiaoyuzhou/podcast.ts index 48e86d83fb..4300f3383c 100644 --- a/lib/routes/xiaoyuzhou/podcast.ts +++ b/lib/routes/xiaoyuzhou/podcast.ts @@ -44,14 +44,16 @@ async function handler(ctx) { response = await ofetch(link); $ = load(response); - const nextDataElement = $('#__NEXT_DATA__').get(0); - page_data = JSON.parse(nextDataElement.children[0].data); - - // If no episodes found, we should try episode URL - if (!page_data.props.pageProps.podcast?.episodes) { - throw new Error('No episodes found in podcast data'); + page_data = JSON.parse($('#__NEXT_DATA__').text()); + } catch (error) { + // An episode ID may return 404 at the podcast URL. Preserve access and + // transport errors instead of hiding them behind a second failed request. + if (!(error instanceof Error) || !('status' in error) || error.status !== 404) { + throw error; } - } catch { + } + + if (!page_data?.props?.pageProps?.podcast?.episodes) { // Try as episode instead link = `https://www.xiaoyuzhoufm.com/episode/${id}`; response = await ofetch(link); @@ -65,11 +67,14 @@ async function handler(ctx) { response = await ofetch(link); $ = load(response); - const nextDataElement = $('#__NEXT_DATA__').get(0); - page_data = JSON.parse(nextDataElement.children[0].data); + page_data = JSON.parse($('#__NEXT_DATA__').text()); } } + if (!page_data?.props?.pageProps?.podcast?.episodes) { + throw new Error('Xiaoyuzhou did not return podcast data for this podcast or episode ID.'); + } + let episodes = page_data.props.pageProps.podcast.episodes.map((item) => ({ title: item.title, enclosure_url: item.enclosure.url, diff --git a/lib/routes/xunhupay/index.ts b/lib/routes/xunhupay/index.ts index a9ad383df9..a02c679fae 100644 --- a/lib/routes/xunhupay/index.ts +++ b/lib/routes/xunhupay/index.ts @@ -1,5 +1,6 @@ import type { Route } from '@/types'; import buildData from '@/utils/common-config'; +import { parseDate } from '@/utils/parse-date'; const baseUrl = 'https://www.xunhupay.com'; @@ -32,19 +33,18 @@ async function handler() { return await buildData({ link, url: link, - title: '%title%', - description: '%description%', - params: { - title: '博客', - description: '虎皮椒-博客', - }, + title: '博客', + description: '虎皮椒-博客', item: { item: '.blog-post > article', - title: `$('h5').text()`, - link: `$('a').attr('href')`, - description: `$('.content').text()`, - pubDate: `parseDate($('.date').text(), 'YYYY-MM-DD')`, - guid: Buffer.from(`$('a').attr('href')`).toString('base64'), + title: ($) => $('h5').text(), + link: ($) => $('a').attr('href'), + description: ($) => $('.content').text(), + pubDate: ($) => parseDate($('.date').text(), 'YYYY-MM-DD'), + guid: ($) => { + const href = $('a').attr('href'); + return href ? Buffer.from(href).toString('base64') : undefined; + }, }, }); } diff --git a/lib/shims/xxhash-wasm.ts b/lib/shims/xxhash-wasm.ts deleted file mode 100644 index e9aefdb49f..0000000000 --- a/lib/shims/xxhash-wasm.ts +++ /dev/null @@ -1,80 +0,0 @@ -// oxlint-disable unicorn/prefer-math-trunc unicorn-js/no-this-outside-of-class -// xxhash-wasm shim for Cloudflare Workers -// Uses Web Crypto API instead of WebAssembly - -type XXHash = { - update(input: string | Uint8Array): XXHash; - digest(): T; -}; - -type XXHashAPI = { - h32(input: string, seed?: number): number; - h32ToString(input: string, seed?: number): string; - h32Raw(inputBuffer: Uint8Array, seed?: number): number; - create32(seed?: number): XXHash; - h64(input: string, seed?: bigint): bigint; - h64ToString(input: string, seed?: bigint): string; - h64Raw(inputBuffer: Uint8Array, seed?: bigint): bigint; - create64(seed?: bigint): XXHash; -}; - -const encoder = new TextEncoder(); - -// Simple sync hash for h32 methods (fallback) -const simpleHash32 = (input: Uint8Array, seed = 0): number => { - let hash = seed; - for (const byte of input) { - hash = Math.trunc((hash << 5) - hash + byte); - } - return hash >>> 0; -}; - -function xxhash(): XXHashAPI { - return { - h32: (input: string, seed?: number): number => simpleHash32(encoder.encode(input), seed), - h32ToString: (input: string, seed?: number): string => simpleHash32(encoder.encode(input), seed).toString(16).padStart(8, '0'), - h32Raw: (inputBuffer: Uint8Array, seed?: number): number => simpleHash32(inputBuffer, seed), - create32: (seed?: number): XXHash => { - const chunks: Uint8Array[] = []; - return { - update(input: string | Uint8Array) { - chunks.push(input instanceof Uint8Array ? input : encoder.encode(input)); - return this; - }, - digest() { - const totalLength = chunks.reduce((sum, arr) => sum + arr.length, 0); - const combined = new Uint8Array(totalLength); - let offset = 0; - for (const chunk of chunks) { - combined.set(chunk, offset); - offset += chunk.length; - } - return simpleHash32(combined, seed); - }, - }; - }, - // h64 methods use async SHA-256 but return sync - this is a limitation - // In practice, only h64ToString is used and it's called with await xxhash() - h64: (_input: string, _seed?: bigint): bigint => { - throw new Error('h64 is not supported in Worker shim, use h64ToString instead'); - }, - h64ToString: (input: string, _seed?: bigint): string => { - // This needs to be sync to match the API, but we use a simple hash - // The actual usage in cache.ts awaits xxhash() first, so this works - let hash = 0n; - const data = encoder.encode(input); - for (const byte of data) { - hash = ((hash << 5n) - hash + BigInt(byte)) & 0xff_ff_ff_ff_ff_ff_ff_ffn; - } - return hash.toString(16).padStart(16, '0'); - }, - h64Raw: (_inputBuffer: Uint8Array, _seed?: bigint): bigint => { - throw new Error('h64Raw is not supported in Worker shim'); - }, - create64: (_seed?: bigint): XXHash => { - throw new Error('create64 is not supported in Worker shim'); - }, - }; -} - -export default xxhash; diff --git a/lib/utils/cache/http.test.ts b/lib/utils/cache/http.test.ts index ea7637f20a..acea7ea59c 100644 --- a/lib/utils/cache/http.test.ts +++ b/lib/utils/cache/http.test.ts @@ -117,4 +117,18 @@ describe('http cache module', () => { expect(cache.status.available).toBe(true); expect(urls[0]).not.toContain('?refresh=1'); }); + + it('reports Cloudflare routing errors instead of treating them as cache misses', async () => { + setHttpCacheEnv(); + const { errorSpy } = await spyOnLogger(); + vi.stubGlobal( + 'fetch', + vi.fn(() => new Response('error code: 1042', { status: 404 })) + ); + const cache = (await import('@/utils/cache/http')).default; + cache.init(); + + await expect(cache.get('missing')).resolves.toBeNull(); + expect(errorSpy).toHaveBeenCalledWith('HTTP cache get returned HTTP 404 without a cache miss response.'); + }); }); diff --git a/lib/utils/cache/http.ts b/lib/utils/cache/http.ts index 5b876cf7e9..c1bbc9737f 100644 --- a/lib/utils/cache/http.ts +++ b/lib/utils/cache/http.ts @@ -106,7 +106,16 @@ export default { } if (response.status === 404) { - return ''; + try { + const data = await response.json(); + if (data && typeof data === 'object' && 'hit' in data && data.hit === false) { + return ''; + } + } catch { + // Cloudflare routing errors can also return 404, without reaching the cache service. + } + logger.error('HTTP cache get returned HTTP 404 without a cache miss response.'); + return null; } if (!response.ok) { diff --git a/lib/utils/common-config.test.ts b/lib/utils/common-config.test.ts index 6cf3b60358..3a4a5d08dd 100644 --- a/lib/utils/common-config.test.ts +++ b/lib/utils/common-config.test.ts @@ -2,128 +2,61 @@ import iconv from 'iconv-lite'; import { http, HttpResponse } from 'msw'; import { describe, expect, it } from 'vitest'; -import configUtils, { getProp, replaceParams, transElemText } from '@/utils/common-config'; +import buildData from '@/utils/common-config'; +import { parseDate } from '@/utils/parse-date'; +import timezone from '@/utils/timezone'; -describe('index', () => { - it('transElemText', () => { - const $ = () => 'RSSHub'; - expect(transElemText($, '$()')).toBe('RSSHub'); - }); +const url = 'http://rsshub.test/buildData'; - it('replaceParams', () => { - const $ = () => 'RSSHub'; - const data = { - params: { - title: 'RSSHub', - }, - title: '%title%', - }; - expect(replaceParams(data, data.title, $)).toBe('RSSHub'); - }); - - it('getProp', () => { - const $ = () => 'RSSHub'; - const data = { - title: 'RSSHub', - }; - expect(getProp(data, ['title'], $)).toBe('RSSHub'); - expect(getProp(data, 'title', $)).toBe('RSSHub'); - }); - - it('all', () => { - const $ = () => 'RSSHub'; - const data = { - params: { - title: '$()', - }, - title: '%title%', - }; - expect(getProp(data, ['title'], $)).toBe('RSSHub'); - }); - - it('buildData', async () => { - const data = await configUtils({ - link: 'http://rsshub.test/buildData', - url: 'http://rsshub.test/buildData', - title: '%title%', - params: { - title: 'buildData', - }, +describe('common-config', () => { + it('extracts page and item values with callbacks and preserves literal text', async () => { + const suffix = 'buildData'; + const literal = `$('script').text() %one% + %two%`; + const data = await buildData({ + link: url, + url, + title: ($) => $('.content li a').first().text() + ' - ' + suffix, + description: literal, item: { item: '.content li', - title: `$('a').text() + ' - %title%'`, - link: `$('a').attr('href')`, - description: `$('.description').html()`, - pubDate: `timezone(parseDate($('.date').text(), 'YYYY-MM-DD'), 0)`, + title: ($) => $('a').text() + ' - ' + suffix, + link: ($) => $('a').attr('href'), + description: ($) => $('.description').html(), + pubDate: ($) => timezone(parseDate($('.date').text(), 'YYYY-MM-DD'), 0), }, }); - - expect(data).toMatchObject({ - link: 'http://rsshub.test/buildData', - title: 'buildData', + expect(data).toEqual({ + link: url, + title: '1 - buildData', + description: literal, + allowEmpty: false, item: [ - { - description: 'RSSHub1', - guid: undefined, - link: '/1', - pubDate: new Date('2025-01-01T00:00:00Z'), - title: '1 - buildData', - }, - { - description: 'RSSHub2', - guid: undefined, - link: '/2', - pubDate: new Date('2025-01-02T00:00:00Z'), - title: '2 - buildData', - }, + { title: '1 - buildData', link: '/1', description: 'RSSHub1', pubDate: new Date('2025-01-01T00:00:00Z'), guid: undefined }, + { title: '2 - buildData', link: '/2', description: 'RSSHub2', pubDate: new Date('2025-01-02T00:00:00Z'), guid: undefined }, ], }); }); -}); -describe('charset', () => { - const html = `
-
    -
  • - 1 -
    中文RSSHub1
    -
    2025-01-01
    -
    • -
-
`; - - it('parses charset from content-type', async () => { - const { default: server } = await import('@/setup.test'); - server.use( - http.get( - 'http://rsshub.test/buildData-gbk', - () => - new HttpResponse(iconv.encode(html, 'gbk'), { - headers: { - 'content-type': 'text/html; charset=gbk', - }, - }) - ) - ); - - const data = await configUtils({ - link: 'http://rsshub.test/buildData-gbk', - url: 'http://rsshub.test/buildData-gbk', - title: '%title%', - params: { - title: 'buildData', - }, - item: { - item: '.content li', - title: `$('a').text() + ' - %title%'`, - link: `$('a').attr('href')`, - description: `$('.description').html()`, - pubDate: `timezone(parseDate($('.date').text(), 'YYYY-MM-DD'), 0)`, - }, + it('preserves constants, numeric dates, missing fields and allowEmpty', async () => { + const date = 1_735_689_600_000; + const data = await buildData({ + url, + title: 'Feed', + allowEmpty: true, + item: { item: '.content li', title: 'Item', description: undefined, pubDate: date }, }); + expect(data.allowEmpty).toBe(true); + expect(data.item).toEqual(Array.from({ length: 2 }, () => ({ title: 'Item', description: undefined, pubDate: date, link: undefined, guid: undefined }))); + const empty = await buildData({ url, title: 'Empty', item: { item: '.missing', title: 'Item' } }); + expect(empty.item).toEqual([]); + expect(empty.allowEmpty).toBe(false); + }); - expect(data.title).toBe('buildData'); - expect(data.item[0].title).toBe('1 - buildData'); - expect(data.item[0].description).toBe('中文RSSHub1'); + it('decodes non-UTF-8 responses before calling selectors', async () => { + const { default: server } = await import('@/setup.test'); + server.use(http.get('http://rsshub.test/buildData-gbk', () => new HttpResponse(iconv.encode('

中文标题

', 'gbk'), { headers: { 'content-type': 'text/html; charset=gbk' } }))); + const data = await buildData({ url: 'http://rsshub.test/buildData-gbk', title: ($) => $('h1').text(), item: { item: 'li', title: ($) => $('a').text(), link: ($) => $('a').attr('href') } }); + expect(data.title).toBe('中文标题'); + expect(data.item[0]).toMatchObject({ title: '中文条目', link: '/1' }); }); }); diff --git a/lib/utils/common-config.ts b/lib/utils/common-config.ts index 3de6f7cc68..46481d61de 100644 --- a/lib/utils/common-config.ts +++ b/lib/utils/common-config.ts @@ -1,52 +1,29 @@ -import { load } from 'cheerio'; +import { type Cheerio, load } from 'cheerio'; +import type { AnyNode } from 'domhandler'; import iconv from 'iconv-lite'; +import type { Data, DataItem } from '@/types'; import ofetch from '@/utils/ofetch'; -import { parseDate as _parseDate } from '@/utils/parse-date'; -import _timezone from '@/utils/timezone'; -function transElemText($, prop) { - const regex = /\$\(.*\)/; - let result = prop; - // eslint-disable-next-line @typescript-eslint/no-unused-vars - const parseDate = _parseDate; - // eslint-disable-next-line @typescript-eslint/no-unused-vars - const timezone = _timezone; - if (regex.test(result)) { - // eslint-disable-next-line no-eval - result = eval(result); - } - return result; +type Selector = (selector: string) => Cheerio; +type Value = T | ((select: Selector) => T); +type ItemFields = Pick; + +type BuildDataConfig = Pick & { + url: string; + title: Value; + description?: Value; + item: { item: string } & { [Key in keyof ItemFields]: Value }; +}; + +function resolveValue(value: Value, select: Selector): T { + return typeof value === 'function' ? (value as (select: Selector) => T)(select) : value; } -function replaceParams(data, prop, $) { - const regex = /%(.*)%/g; - let result = prop; - let group = regex.exec(prop); - while (group) { - // FIXME Multi vars - result = result.replace(group[0], () => transElemText($, data.params[group![1]])); - group = regex.exec(prop); - } - return result; -} - -function getProp(data, prop, $) { - let result = data; - if (Array.isArray(prop)) { - for (const e of prop) { - result = transElemText($, result[e]); - } - } else { - result = transElemText($, result[prop]); - } - return replaceParams(data, result, $); -} - -export default async function buildData(data) { +export default async function buildData(data: BuildDataConfig) { const response = await ofetch.raw(data.url); const contentType = response.headers.get('content-type') || ''; - // 若没有指定编码，则默认utf-8 + // Default to UTF-8 when no encoding is specified. let charset = 'utf-8'; for (const attr of contentType.split(';')) { if (attr.includes('charset=')) { @@ -55,24 +32,22 @@ export default async function buildData(data) { } const responseData = charset === 'utf-8' ? response._data : iconv.decode(Buffer.from(await ofetch(data.url, { responseType: 'arrayBuffer' })), charset); const $ = load(responseData); - const $item = $(data.item.item); - // 这里应该是可以通过参数注入一些代码的，不过应该无伤大雅 return { link: data.link, - title: getProp(data, 'title', $), - description: getProp(data, 'description', $), + title: resolveValue(data.title, $), + description: resolveValue(data.description, $), allowEmpty: data.allowEmpty || false, - item: $item.toArray().map((e) => { - const $elem = (selector) => $(e).find(selector); - return { - title: getProp(data, ['item', 'title'], $elem), - description: getProp(data, ['item', 'description'], $elem), - pubDate: getProp(data, ['item', 'pubDate'], $elem), - link: getProp(data, ['item', 'link'], $elem), - guid: getProp(data, ['item', 'guid'], $elem), - }; - }), + item: $(data.item.item) + .toArray() + .map((element) => { + const select: Selector = (selector) => $(element).find(selector); + return { + title: resolveValue(data.item.title, select), + description: resolveValue(data.item.description, select), + pubDate: resolveValue(data.item.pubDate, select), + link: resolveValue(data.item.link, select), + guid: resolveValue(data.item.guid, select), + }; + }), }; } - -export { getProp, replaceParams, transElemText }; diff --git a/lib/utils/parse-date-in-timezone.test.ts b/lib/utils/parse-date-in-timezone.test.ts new file mode 100644 index 0000000000..fec997814f --- /dev/null +++ b/lib/utils/parse-date-in-timezone.test.ts @@ -0,0 +1,32 @@ +import { describe, expect, it } from 'vitest'; + +import { parseDateInTimezone } from './parse-date-in-timezone'; + +describe('parseDateInTimezone', () => { + it.each([ + ['2026-08-28T08:07:22', 8, '2026-08-28T00:07:22.000Z'], + ['2026-09-07 00:00:00.000', 8, '2026-09-06T16:00:00.000Z'], + ['2026-09-06T15:41:43', 0, '2026-09-06T15:41:43.000Z'], + ['2026-09-18', 0, '2026-09-18T00:00:00.000Z'], + ['2026-09-07T00:00:00', -3.5, '2026-09-07T03:30:00.000Z'], + ['2026-09-07T08:00:00+08:00', 8, '2026-09-07T00:00:00.000Z'], + ['2026-09-07T08:00:00+0800', 0, '2026-09-07T00:00:00.000Z'], + ['2026-09-07T08:00:00-05:00', 8, '2026-09-07T13:00:00.000Z'], + ['2026-09-07T00:00:00Z', 8, '2026-09-07T00:00:00.000Z'], + ['Mon, 07 Sep 2026 08:00:00 GMT+8', 8, '2026-09-07T00:00:00.000Z'], + ])('parses %s in offset %s independently of the host timezone', (input, offset, expected) => { + expect(parseDateInTimezone(input, offset).toISOString()).toBe(expected); + }); + + it('preserves absolute Date and millisecond timestamp inputs without mutating them', () => { + const date = new Date('2026-09-07T00:00:00Z'); + expect(parseDateInTimezone(date, 8).getTime()).toBe(date.getTime()); + expect(parseDateInTimezone(date.getTime(), -5).getTime()).toBe(date.getTime()); + expect(date.toISOString()).toBe('2026-09-07T00:00:00.000Z'); + }); + + it('does not invent a date for an empty or unparseable string', () => { + expect(Number.isNaN(parseDateInTimezone('', 8).getTime())).toBe(true); + expect(Number.isNaN(parseDateInTimezone('unknown', 8).getTime())).toBe(true); + }); +}); diff --git a/lib/utils/parse-date-in-timezone.ts b/lib/utils/parse-date-in-timezone.ts new file mode 100644 index 0000000000..4f77ba81ba --- /dev/null +++ b/lib/utils/parse-date-in-timezone.ts @@ -0,0 +1,14 @@ +import dayjs from 'dayjs'; +import utc from 'dayjs/plugin/utc'; + +dayjs.extend(utc); + +// Interpret a source's wall clock without using the host's timezone. Dates, +// timestamps and ISO strings with an explicit offset already identify an instant. +export const parseDateInTimezone = (date: string | number | Date, offset: number): Date => { + const parsed = dayjs.utc(date); + if (typeof date !== 'string' || /(?:Z|[+-]\d{2}:?\d{2}|(?:GMT|UTC)(?:[+-]\d{1,4})?)$/i.test(date.trim())) { + return parsed.toDate(); + } + return parsed.subtract(offset, 'hour').toDate(); +}; diff --git a/tests/2048-utils.test.ts b/tests/2048-utils.test.ts new file mode 100644 index 0000000000..7891e2a5f8 --- /dev/null +++ b/tests/2048-utils.test.ts @@ -0,0 +1,21 @@ +import { describe, expect, it } from 'vitest'; + +import { getForumUrl } from '../lib/routes/2048/utils'; + +describe('2048 forum address discovery', () => { + it('follows the current address page anchor', () => { + expect(getForumUrl('Forum')).toBe('https://2048.info/bbs.php'); + }); + + it('accepts the previous onclick format', () => { + expect(getForumUrl('')).toBe('https://forum.example/'); + }); + + it('rejects missing links before they can be cached as /undefined', () => { + expect(() => getForumUrl('Unavailable')).toThrow('did not contain a forum link'); + }); + + it('rejects script URLs', () => { + expect(() => getForumUrl('Forum')).toThrow('unsupported forum URL'); + }); +}); diff --git a/tests/common-config-routes.test.ts b/tests/common-config-routes.test.ts new file mode 100644 index 0000000000..0d0c0cd548 --- /dev/null +++ b/tests/common-config-routes.test.ts @@ -0,0 +1,141 @@ +import type { Context } from 'hono'; +import { beforeEach, describe, expect, it, vi } from 'vitest'; + +import type { Data, Route } from '../lib/types'; + +const mocks = vi.hoisted(() => ({ raw: vi.fn(), get: vi.fn() })); +vi.mock('../lib/utils/ofetch', () => ({ default: Object.assign(vi.fn(), { raw: mocks.raw }) })); +vi.mock('../lib/utils/got', () => ({ default: mocks.get })); +vi.mock('../lib/utils/cache', () => ({ default: { tryGet: (_key: string, load: () => unknown) => load() } })); +vi.mock('../lib/routes/weibo/utils', () => ({ default: { sinaimgTvax: (data: Data) => data } })); + +const fixtures = [ + { + name: 'government policy library', + module: () => import('../lib/routes/gov/zhengce/zhengceku'), + params: { department: 'bmwj' }, + html: '
部门文件
', + title: '部门文件 - 政府文件库', + item: { title: '政策正文', link: '/policy', pubDate: new Date(2025, 0, 1) }, + }, + { + name: 'Polimi', + module: () => import('../lib/routes/polimi/news'), + params: { language: 'en' }, + html: '', + title: 'Polimi News', + item: { title: 'News', link: 'https://www.polimi.it/story', description: 'Body', pubDate: new Date('2025-01-01T00:00:00Z') }, + }, + { + name: 'Jewish Museum', + module: () => import('../lib/routes/jewishmuseum/exhibitions'), + html: '
', + title: 'Jewish Museums - Exhibitions', + item: { title: 'Exhibition', link: '/show' }, + }, + { + name: 'Brooklyn Museum', + module: () => import('../lib/routes/brooklynmuseum/exhibitions'), + html: '
', + title: 'Brooklyn Museum - Exhibitions', + item: { title: 'Exhibition', link: '/show', description: 'Description' }, + }, + { + name: 'New Museum', + module: () => import('../lib/routes/newmuseum/exhibitions'), + html: '
Exhibition
Description
', + title: 'New Museum - Exhibitions', + item: { title: 'Exhibition', link: '/show', description: 'Description' }, + }, + { + name: 'Stratechery', + module: () => import('../lib/routes/stratechery/index'), + html: '', + title: 'Stratechery by Ben Thompson', + item: { title: 'Post', link: '/post', description: '

100% growth

', pubDate: new Date('2025-01-01T00:00:00Z') }, + }, + { + name: 'Queshu book', + module: () => import('../lib/routes/queshu/book'), + params: { bookid: '123' }, + html: '

Book

Description
Offer
', + title: 'Book - 单品活动信息 - 缺书网', + item: { title: 'Offer', link: 'http://www.queshu.com/sale' }, + }, + { + name: 'Queshu sale', + module: () => import('../lib/routes/queshu/sale'), + html: '
BookOfferEnds todayYesterday
', + title: '图书促销 - 缺书网', + item: { title: 'Offer：Book', link: 'http://www.queshu.com/book', description: 'Offer：Book
Ends today
发布时间：Yesterday' }, + }, + { + name: 'AIEA', + module: () => import('../lib/routes/aiea/index'), + params: { period: 'upcoming' }, + html: '
Seminar
Details
Description
', + title: 'AIEA Seminars', + item: { title: 'Seminar', link: '/0504/1', description: 'Description' }, + }, + { + name: 'Oasis', + module: () => import('../lib/routes/weibo/oasis/user'), + params: { userid: '123' }, + html: '
Author
Profile
PostIgnored child
', + title: 'Author - 用户 - 绿洲', + item: { title: 'Post', link: 'https://oasis.weibo.cn/v1/h5/share?sid=456', description: 'Post
' }, + }, + { + name: 'Oct0pu5', + module: () => import('../lib/routes/oct0pu5/rss'), + html: '
Post
Description
', + title: '博客', + item: { title: 'Post', link: '/post', description: 'Description', pubDate: 1_735_689_600_000 }, + }, + { + name: 'Air China', + module: () => import('../lib/routes/airchina/index'), + html: '', + title: '国航服务公告', + item: { title: 'First', link: '/1', pubDate: new Date(2025, 0, 1), description: 'Article detail' }, + uniqueGuid: true, + }, + { + name: 'Xunhupay', + module: () => import('../lib/routes/xunhupay/index'), + html: '
First
Description
2025-01-01
Second
2025-01-02
', + title: '博客', + item: { title: 'First', link: '/1', pubDate: new Date(2025, 0, 1), description: 'Description' }, + uniqueGuid: true, + }, + { + name: 'iiilab', + module: () => import('../lib/routes/iiilab/index'), + html: '
First
Description
2025-01-01 00:00
Second
2025-01-02 00:00
', + title: '发现', + item: { title: 'First', link: '/1', pubDate: new Date(2025, 0, 1), description: 'Description' }, + uniqueGuid: true, + }, +]; + +beforeEach(() => { + vi.clearAllMocks(); + mocks.get.mockResolvedValue({ data: '
Article detail
' }); +}); + +describe('common-config route callbacks', () => { + it.each(fixtures)('$name extracts its HTML fixture without evaluating selector strings', async (fixture) => { + mocks.raw.mockResolvedValue({ headers: new Headers({ 'content-type': 'text/html' }), _data: fixture.html }); + const { route } = await fixture.module(); + const params: Record = fixture.params ?? {}; + const ctx = { req: { param: (key?: string) => (key ? params[key] : params), query: () => {} } } as unknown as Context; + const data = (await (route.handler as Exclude)(ctx)) as Data; + expect(mocks.raw).toHaveBeenCalledTimes(1); + expect(data.title).toBe(fixture.title); + expect(data.item).toHaveLength(fixture.uniqueGuid ? 2 : 1); + expect(data.item![0]).toMatchObject(fixture.item); + if (fixture.uniqueGuid) { + expect(data.item!.map((item) => item.guid)).toEqual(['LzE=', 'LzI=']); + } + }); +}); diff --git a/tests/common-config.worker.test.ts b/tests/common-config.worker.test.ts new file mode 100644 index 0000000000..6702b6884d --- /dev/null +++ b/tests/common-config.worker.test.ts @@ -0,0 +1,40 @@ +import { afterEach, expect, it, vi } from 'vitest'; + +import { route } from '../lib/routes/gov/zhengce/zhengceku'; +import type { Data, Route } from '../lib/types'; +import buildData from '../lib/utils/common-config'; + +// These unused network/devtools exports are tree-shaken in the production bundle. +vi.mock('undici', () => ({})); +vi.mock('node-network-devtools', () => ({ register: vi.fn() })); +vi.mock('../lib/config', () => ({ config: { requestRetry: 0, loggerLevel: 'error' } })); + +afterEach(() => vi.unstubAllGlobals()); + +it('extracts the policy library in workerd with page/item callbacks and dates', async () => { + const fetchMock = vi.fn(() => + Promise.resolve( + new Response( + '
部门文件
', + { headers: { 'content-type': 'text/html' } } + ) + ) + ); + vi.stubGlobal('fetch', fetchMock); + const data = (await (route.handler as Exclude)({ req: { param: () => 'bmwj' } } as never)) as Data; + expect(data.title).toBe('部门文件 - 政府文件库'); + expect(data.item).toEqual([{ title: '政策', link: '/policy', pubDate: new Date('2025-01-01T00:00:00Z'), description: undefined, guid: undefined }]); + expect(fetchMock).toHaveBeenCalledTimes(1); +}); + +it('treats JavaScript-looking text and percent signs as literal data in workerd', async () => { + vi.stubGlobal( + 'fetch', + vi.fn(() => Promise.resolve(new Response('
  • 100% growth %remaining%
    • ', { headers: { 'content-type': 'text/html' } }))) + ); + const title = `$('a').text(); globalThis.shouldNotRun = true`; + const result = await buildData({ url: 'https://example.test/', title, item: { item: 'li', title: ($) => $('a').text() } }); + expect(result.title).toBe(title); + expect(result.item[0].title).toBe('100% growth %remaining%'); + expect(Reflect.has(globalThis, 'shouldNotRun')).toBe(false); +}); diff --git a/tests/fixtures/readhub/timeline.json b/tests/fixtures/readhub/timeline.json new file mode 100644 index 0000000000..81837b24db --- /dev/null +++ b/tests/fixtures/readhub/timeline.json @@ -0,0 +1,9 @@ +{ + "data": { + "self": { "title": "相关事件", "topicTitle": "Topic" }, + "items": [ + { "uid": "earlier-1", "title": "Earlier event", "publishDate": "2026-09-05T02:03:04.000Z" }, + { "uid": "earlier-2", "title": "Another event", "publishDate": "2026-09-04T15:04:05+08:00" } + ] + } +} diff --git a/tests/fixtures/readhub/topic.json b/tests/fixtures/readhub/topic.json new file mode 100644 index 0000000000..1b24cd3c33 --- /dev/null +++ b/tests/fixtures/readhub/topic.json @@ -0,0 +1,22 @@ +{ + "data": { + "items": [ + { + "uid": "topic-42", + "title": "Topic \"quoted\" & 中文", + "summary": "First paragraph.\nA quoted \"statement\", path C:\\topic, and .", + "createdAt": "2026-09-06T08:00:00.000Z", + "publishDate": "2026-09-06T09:40:34.084Z", + "siteCount": 2, + "siteNameDisplay": "Source A", + "newsAggList": [ + { "uid": "news-1", "url": "https://example.com/report-1", "title": "Report A", "siteNameDisplay": "Source A" }, + { "uid": "news-2", "url": "https://example.com/report-2", "title": "Report B", "siteNameDisplay": "Source B" } + ], + "entityList": [{ "name": "Entity" }], + "tagList": [{ "name": "Technology" }], + "itemId": 42 + } + ] + } +} diff --git a/tests/readhub.test.ts b/tests/readhub.test.ts new file mode 100644 index 0000000000..0c9bbb1539 --- /dev/null +++ b/tests/readhub.test.ts @@ -0,0 +1,163 @@ +import { beforeEach, describe, expect, it, vi } from 'vitest'; + +import { route as daily } from '../lib/routes/readhub/daily'; +import { processItems } from '../lib/routes/readhub/util'; +import cache from '../lib/utils/cache'; +import got from '../lib/utils/got'; +import timelineFixture from './fixtures/readhub/timeline.json'; +import topicFixture from './fixtures/readhub/topic.json'; + +vi.mock('../lib/utils/cache', () => ({ default: { tryGet: vi.fn() } })); +vi.mock('../lib/utils/got', () => ({ default: vi.fn() })); + +// Response shapes from public Readhub detail/timeline APIs on 2026-09-07. +// Values are synthetic; escaped text and explicit offsets exercise preservation. +const detailUrl = 'https://api.readhub.cn/topic/detail'; +const timelineUrl = 'https://api.readhub.cn/topic/timeline/list'; +const link = 'https://readhub.cn/topic/topic-42'; +const input = { title: 'Daily summary', link, guid: 'topic-42', description: '

    Incomplete summary

    ' }; +const cached = new Map(); + +beforeEach(() => { + vi.clearAllMocks(); + cached.clear(); + vi.mocked(cache.tryGet).mockImplementation(async (key, fetchValue) => { + if (cached.has(key)) { + return cached.get(key); + } + const value = await fetchValue(); + cached.set(key, value); + return value; + }); + vi.mocked(got).mockImplementation((url) => { + if (url === detailUrl) { + return Promise.resolve({ data: structuredClone(topicFixture) }); + } + if (url === timelineUrl) { + return Promise.resolve({ data: structuredClone(timelineFixture) }); + } + throw new Error('Unexpected upstream request'); + }); +}); + +describe('Readhub API enrichment', () => { + it('preserves the full summary, media, timeline, tags, title, GUID, and upstream publication date', async () => { + const [item] = await processItems([input]); + + expect(item).toMatchObject({ + title: topicFixture.data.items[0].title, + link, + author: 'Source A', + category: ['Entity', 'Technology'], + guid: 'readhub-topic-42', + pubDate: new Date('2026-09-06T09:40:34.084Z'), + }); + expect(item.description).toContain('First paragraph.\nA quoted "statement", path C:\\topic, and <markup>.'); + expect(item.description).toContain('

    媒体报道

    '); + expect(item.description).toContain('Report A'); + expect(item.description).toContain('Source B'); + expect(item.description).toContain('

    事件追踪

    '); + expect(item.description).toContain('Earlier event'); + expect(item.description).toContain('2026-09-05 10:03:04'); + expect(item.description).toContain('2026-09-04 15:04:05'); + expect(got).toHaveBeenNthCalledWith(1, detailUrl, { searchParams: { uid: 'topic-42' } }); + expect(got).toHaveBeenNthCalledWith(2, timelineUrl, { searchParams: { topic_uid: 'topic-42', size: 10 } }); + expect(input.guid).toBe('topic-42'); + }); + + it('does not reuse or extend old cached summary-only items', async () => { + cached.set(link, { ...input, guid: 'readhub-topic-42' }); + const [first] = await processItems([input]); + const [second] = await processItems([{ ...input, title: 'Another route title' }]); + + expect(first).toEqual(second); + expect(first.pubDate).toBeInstanceOf(Date); + expect(cache.tryGet).toHaveBeenCalledWith('readhub:topic:v2:topic-42', expect.any(Function)); + expect(got).toHaveBeenCalledTimes(2); + }); + + it('selects the requested UID instead of another item in the response', async () => { + const response = structuredClone(topicFixture); + response.data.items.unshift({ ...response.data.items[0], uid: 'unrelated', title: 'Unrelated' }); + vi.mocked(got).mockResolvedValueOnce({ data: response }); + expect((await processItems([input]))[0].title).toBe(topicFixture.data.items[0].title); + }); + + it.each(['missing', 'duplicate'])('rejects a %s requested topic without caching the summary', async (kind) => { + const response = structuredClone(topicFixture); + if (kind === 'missing') { + response.data.items[0].uid = 'unrelated'; + } else { + response.data.items.push(response.data.items[0]); + } + vi.mocked(got).mockResolvedValueOnce({ data: response }); + await expect(processItems([input])).rejects.toThrow('does not contain the requested topic'); + expect(cached.size).toBe(0); + }); + + it.each([ + { data: { items: [{ ...topicFixture.data.items[0], publishDate: undefined }] } }, + { data: { items: [{ ...topicFixture.data.items[0], publishDate: 'not a date' }] } }, + { data: { items: [{ ...topicFixture.data.items[0], summary: '' }] } }, + { data: { items: [{ ...topicFixture.data.items[0], newsAggList: undefined }] } }, + 'Unexpected upstream page', + ])('rejects invalid detail data instead of returning an incomplete item', async (response) => { + vi.mocked(got).mockResolvedValueOnce({ data: response }); + await expect(processItems([input])).rejects.toThrow('Readhub topic detail response is invalid'); + expect(cached.size).toBe(0); + }); + + it('retries after a failed enrichment rather than caching the fallback', async () => { + vi.mocked(got).mockRejectedValueOnce(new Error('Upstream unavailable')); + await expect(processItems([input])).rejects.toThrow('Upstream unavailable'); + expect(cached.size).toBe(0); + expect((await processItems([input]))[0].pubDate).toEqual(new Date('2026-09-06T09:40:34.084Z')); + }); + + it('rejects a malformed timeline instead of dropping related events', async () => { + vi.mocked(got) + .mockResolvedValueOnce({ data: topicFixture }) + .mockResolvedValueOnce({ data: { data: {} } }); + await expect(processItems([input])).rejects.toThrow('Readhub topic timeline response is invalid'); + expect(cached.size).toBe(0); + }); + + it('accepts an explicitly empty timeline and optional tags without inventing related items', async () => { + const response = { data: { items: [{ ...topicFixture.data.items[0], entityList: undefined, tagList: null }] } }; + vi.mocked(got) + .mockResolvedValueOnce({ data: response }) + .mockResolvedValueOnce({ data: { data: { items: [] } } }); + const [item] = await processItems([input]); + expect(item.category).toEqual([]); + expect(item.description).toContain('媒体报道'); + expect(item.description).not.toContain('事件追踪'); + expect(item.pubDate).toEqual(new Date('2026-09-06T09:40:34.084Z')); + }); + + it.each(['https://example.com/article', 'https://readhub.cn.example.com/topic/topic-42'])('preserves external category items without fetching their pages: %s', async (externalLink) => { + const external = { ...input, link: externalLink, pubDate: new Date('2026-09-01T00:00:00Z') }; + expect(await processItems([external])).toEqual([{ ...external, guid: 'readhub-topic-42' }]); + expect(got).not.toHaveBeenCalled(); + expect(cache.tryGet).not.toHaveBeenCalled(); + }); + + it('enriches daily summaries that contain no date and keeps the public feed URL', async () => { + const upstreamMock = vi.mocked(got).getMockImplementation()!; + vi.mocked(got).mockImplementation((url, options) => { + if (url === 'https://api.readhub.cn/daily') { + return Promise.resolve({ data: { data: { items: [{ uid: 'topic-42', title: input.title, summary: 'Daily summary' }] } } }); + } + if (url === 'https://readhub.cn/daily') { + return Promise.resolve({ data: '' }); + } + return upstreamMock(url, options); + }); + const feed = await daily.handler({ req: { query: vi.fn() } } as never); + + expect(feed.link).toBe('https://readhub.cn/daily'); + expect(feed.item).toHaveLength(1); + expect(feed.item![0].pubDate).toEqual(new Date('2026-09-06T09:40:34.084Z')); + expect(feed.item![0].description).toContain('Report A'); + expect(feed.item![0].description).toContain('Earlier event'); + }); +}); diff --git a/tests/readhub.worker.test.ts b/tests/readhub.worker.test.ts new file mode 100644 index 0000000000..88139a4393 --- /dev/null +++ b/tests/readhub.worker.test.ts @@ -0,0 +1,2 @@ +// Run the same enrichment and failure cases in workerd. +import './readhub.test'; diff --git a/tests/source-date-routes.test.ts b/tests/source-date-routes.test.ts new file mode 100644 index 0000000000..d26519b659 --- /dev/null +++ b/tests/source-date-routes.test.ts @@ -0,0 +1,104 @@ +import type { Context } from 'hono'; +import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'; + +import { route as eastmoney } from '../lib/routes/eastmoney/report'; +import { route as hellogithub } from '../lib/routes/hellogithub'; +import { route as iCable } from '../lib/routes/i-cable/news'; +import { route as javbus } from '../lib/routes/javbus'; +import { route as smzdm } from '../lib/routes/smzdm/keyword'; +import { parseSearchDate } from '../lib/routes/smzdm/utils'; +import type { Data, Route } from '../lib/types'; + +const mocks = vi.hoisted(() => ({ request: vi.fn(), tryGet: vi.fn() })); +vi.mock('../lib/utils/got', () => ({ default: mocks.request })); +vi.mock('../lib/utils/cache', () => ({ default: { tryGet: mocks.tryGet } })); +vi.mock('../lib/config', () => ({ config: { smzdm: { cookie: 'test-only-cookie' }, feature: { allow_user_supply_unsafe_domain: false }, cache: { routeExpire: 900 } } })); + +const invoke = async (route: Route, path: string, params: Record = {}) => + (await route.handler({ req: { path, query: vi.fn(), param: (name?: string) => (name ? params[name] : params) } } as unknown as Context)) as Data; + +beforeEach(() => { + vi.resetAllMocks(); + mocks.request.mockRejectedValue(new Error('Unexpected upstream request')); + mocks.tryGet.mockImplementation(async (_key, callback) => await callback()); + vi.stubGlobal('fetch', vi.fn().mockRejectedValue(new Error('External network access is disabled'))); +}); + +afterEach(() => { + vi.useRealTimers(); + vi.unstubAllGlobals(); +}); + +describe('explicit source dates', () => { + it('interprets HelloGitHub wall clock dates as UTC+8 and preserves explicit offsets', async () => { + mocks.request.mockResolvedValueOnce({ + data: { data: ['2026-08-28T08:07:22', '2026-08-28T00:07:22Z', '2026-08-28T08:07:22+08:00'].map((updated_at, index) => ({ item_id: String(index), name: 'Project', title: 'Summary', author: 'Author', updated_at })) }, + }); + const result = await invoke(hellogithub, '/hellogithub/home'); + expect(result.item.map((item) => (item.pubDate as Date).toISOString())).toEqual(Array.from({ length: 3 }, () => '2026-08-28T00:07:22.000Z')); + }); + + it('uses the WordPress GMT field as UTC and leaves absent dates absent', async () => { + mocks.tryGet.mockResolvedValueOnce({ data: [{ id: 1, name: 'News', link: 'https://example.com/news' }] }); + mocks.request.mockResolvedValueOnce({ + data: ['2026-09-06T15:41:43', '2026-09-06T23:41:43+08:00', undefined].map((date_gmt) => ({ + title: { rendered: 'Title' }, + link: 'https://example.com/article', + date_gmt, + content: { rendered: '

    Article

    ' }, + _embedded: { 'wp:term': [[]] }, + })), + }); + const result = await invoke(iCable, '/i-cable/news'); + expect(result.item.slice(0, 2).map((item) => (item.pubDate as Date).toISOString())).toEqual(Array.from({ length: 2 }, () => '2026-09-06T15:41:43.000Z')); + expect(result.item[2].pubDate).toBeUndefined(); + }); + + it('corrects Eastmoney list dates even when full article entries were cached with the old host timezone', async () => { + const cached = { title: 'Cached report', link: 'https://example.com/report.pdf', description: '

    Cached body

    ', pubDate: '2026-09-07T00:00:00.000Z' }; + mocks.request.mockResolvedValueOnce({ + data: ``, + }); + mocks.tryGet.mockResolvedValueOnce(cached); + const result = await invoke(eastmoney, '/eastmoney/report/industry', { category: 'industry' }); + expect(result.item[0]).toMatchObject({ description: cached.description, link: cached.link, pubDate: new Date('2026-09-06T16:00:00Z') }); + expect(cached.pubDate).toBe('2026-09-07T00:00:00.000Z'); + expect(mocks.request).toHaveBeenCalledTimes(1); + }); + + it('preserves the Javbus calendar release date at UTC midnight on cache hits', async () => { + const cached = { title: 'Cached release', link: 'https://www.javbus.com/release', description: '

    Cached body

    ', pubDate: '2026-09-17T16:00:00.000Z' }; + mocks.request.mockResolvedValueOnce({ data: 'Releasesrelease-id2026-09-18' }); + mocks.tryGet.mockResolvedValueOnce(cached); + const result = await invoke(javbus, '/javbus/star/example'); + expect(result.item[0]).toMatchObject({ description: cached.description, pubDate: new Date('2026-09-18T00:00:00Z') }); + expect(cached.pubDate).toBe('2026-09-17T16:00:00.000Z'); + expect(mocks.request).toHaveBeenCalledTimes(1); + }); + + it('skips SMZDM layout rows without article links and resolves time-only dates in the source calendar day', async () => { + vi.useFakeTimers({ toFake: ['Date'] }); + vi.setSystemTime(new Date('2026-09-06T18:10:00Z')); + mocks.request.mockResolvedValueOnce({ + data: ` +
    00:40
    +
    DealPrice
    +
    00:40Source
    `, + }); + const result = await invoke(smzdm, '/smzdm/keyword/example', { keyword: 'example' }); + expect(result.item).toHaveLength(1); + expect(result.item[0]).toMatchObject({ link: 'https://example.com/deal', title: 'Deal - Price', pubDate: new Date('2026-09-06T16:40:00Z') }); + }); + + it('resolves SMZDM month/day values at the source year boundary', () => { + const now = new Date('2025-12-31T18:10:00Z'); + expect(parseSearchDate('00:40', now)?.toISOString()).toBe('2025-12-31T16:40:00.000Z'); + expect(parseSearchDate('12-31 23:40', now)?.toISOString()).toBe('2025-12-31T15:40:00.000Z'); + expect(parseSearchDate('01-01 01:40', now)?.toISOString()).toBe('2025-12-31T17:40:00.000Z'); + }); + + it('leaves unknown SMZDM dates absent instead of inventing the current time', () => { + expect(parseSearchDate('')).toBeUndefined(); + expect(parseSearchDate('unknown')).toBeUndefined(); + }); +}); diff --git a/tests/twitter-web-user.test.ts b/tests/twitter-web-user.test.ts new file mode 100644 index 0000000000..85288f838e --- /dev/null +++ b/tests/twitter-web-user.test.ts @@ -0,0 +1,151 @@ +import { beforeEach, describe, expect, it, vi } from 'vitest'; + +import { config } from '../lib/config'; +import InvalidParameterError from '../lib/errors/types/invalid-parameter'; +import api from '../lib/routes/twitter/api/web-api/api'; +import { gatherLegacyFromData, paginationTweets, twitterGot } from '../lib/routes/twitter/api/web-api/utils'; +import cache from '../lib/utils/cache'; +import ofetch from '../lib/utils/ofetch'; + +vi.mock('../lib/config', () => ({ config: { twitter: {}, cache: { routeExpire: 300 } } })); +vi.mock('../lib/utils/cache', () => ({ default: { tryGet: vi.fn(), set: vi.fn() } })); +vi.mock('../lib/utils/ofetch', () => ({ default: vi.fn() })); +vi.mock('../lib/routes/twitter/utils', () => ({ getTwitterUserCacheKey: vi.fn(() => 'tweet-cache') })); +vi.mock('../lib/routes/twitter/api/web-api/constants', () => ({ + baseUrl: 'https://x.com/i/api', + gqlMap: { UserByScreenName: '/graphql/user-by-name', UserByRestId: '/graphql/user-by-id' }, + gqlFeatures: { UserByScreenName: {}, UserByRestId: {} }, + initGqlMap: vi.fn(), +})); +vi.mock('../lib/routes/twitter/api/web-api/utils', () => ({ twitterGot: vi.fn(), paginationTweets: vi.fn(), gatherLegacyFromData: vi.fn() })); + +const cached = new Map(); +const user = { + data: { + user: { + result: { + __typename: 'User', + rest_id: '123', + core: { name: 'Example', screen_name: 'example' }, + avatar: { image_url: 'https://example.com/avatar.png' }, + profile_bio: { description: 'Example bio' }, + }, + }, + }, +}; + +beforeEach(() => { + vi.clearAllMocks(); + cached.clear(); + config.twitter.thirdPartyApi = undefined; + vi.mocked(cache.tryGet).mockImplementation(async (key, fetchValue) => { + if (cached.has(key)) { + return cached.get(key); + } + const value = await fetchValue(); + cached.set(key, value); + return value; + }); + vi.mocked(twitterGot).mockResolvedValue(structuredClone(user)); + vi.mocked(ofetch).mockResolvedValue(structuredClone(user)); +}); + +describe('Twitter web user lookup', () => { + it('preserves valid user metadata and caches only the verified web response', async () => { + cached.set('twitter-userdata-example', { errors: [{ code: 88, message: 'Legacy cached failure' }] }); + expect(await api.getUser('example')).toEqual({ + name: 'Example', + screen_name: 'example', + profile_image_url: 'https://example.com/avatar.png', + description: 'Example bio', + }); + await api.getUser('example'); + expect(twitterGot).toHaveBeenCalledTimes(1); + expect(cache.tryGet).toHaveBeenCalledWith('twitter:web:userdata:v2:example', expect.any(Function)); + expect(cached.get('twitter-userdata-example')).toEqual({ errors: [{ code: 88, message: 'Legacy cached failure' }] }); + expect(cache.set).not.toHaveBeenCalled(); + }); + + it.each([ + [{ errors: [{ code: 88, message: 'Private upstream details' }] }, 'Twitter API user lookup failed (codes: 88)'], + [{ ...user, errors: [{ code: 89 }] }, 'Twitter API user lookup failed (codes: 89)'], + [{ errors: [{ message: 'Private upstream details' }] }, 'Twitter API user lookup failed'], + [{ errors: { message: 'Malformed errors' } }, 'Twitter API user lookup failed'], + [{ data: {} }, 'Twitter API returned an incomplete user response'], + [{ data: { user: {} } }, 'Twitter API returned an incomplete user response'], + [{ data: { user: { result: { __typename: 'User' } } } }, 'Twitter API returned an incomplete user response'], + [undefined, 'Twitter API returned an incomplete user response'], + ['Unexpected upstream response', 'Twitter API returned an incomplete user response'], + ])('rejects an upstream failure without caching or reporting a missing account', async (response, message) => { + vi.mocked(twitterGot).mockResolvedValueOnce(response); + let result: unknown; + try { + await api.getUser('example'); + } catch (error) { + result = error; + } + expect(result).toBeInstanceOf(Error); + expect(result).not.toBeInstanceOf(InvalidParameterError); + expect((result as Error).message).toBe(message); + expect(cached.size).toBe(0); + expect(cache.set).not.toHaveBeenCalled(); + }); + + it.each([ + [{ data: { user: null } }, "This account doesn't exist"], + [{ data: { user: { result: null } } }, "This account doesn't exist"], + [{ data: { user: { result: { __typename: 'UserUnavailable', message: 'Upstream details' } } } }, 'Twitter user is unavailable'], + ])('distinguishes an explicitly missing or unavailable user', async (response, message) => { + vi.mocked(twitterGot).mockResolvedValueOnce(response); + await expect(api.getUser('example')).rejects.toThrow(new InvalidParameterError(message)); + expect(cached.size).toBe(0); + }); + + it('retries normally after a GraphQL error and accepts an empty errors array', async () => { + vi.mocked(twitterGot) + .mockResolvedValueOnce({ errors: [{ code: 88 }] }) + .mockResolvedValueOnce({ ...user, errors: [] }); + await expect(api.getUser('example')).rejects.toThrow('Twitter API user lookup failed'); + expect((await api.getUser('example')).name).toBe('Example'); + expect(twitterGot).toHaveBeenCalledTimes(2); + }); + + it('validates cached data before dereferencing it', async () => { + cached.set('twitter:web:userdata:v2:example', { data: {} }); + await expect(api.getUser('example')).rejects.toThrow('Twitter API returned an incomplete user response'); + expect(twitterGot).not.toHaveBeenCalled(); + expect(cache.set).not.toHaveBeenCalled(); + }); + + it('preserves numeric-ID lookup parameters and auth requirements', async () => { + await api.getUser('+123'); + expect(twitterGot).toHaveBeenCalledWith('https://x.com/i/api/graphql/user-by-id', expect.objectContaining({ variables: JSON.stringify({ userId: '123', withSafetyModeUserFields: true }) }), { + allowNoAuth: false, + }); + }); + + it('validates third-party GraphQL errors without calling the direct API', async () => { + config.twitter.thirdPartyApi = 'https://api.example.com'; + vi.mocked(ofetch).mockResolvedValueOnce({ errors: [{ code: 88 }] }); + await expect(api.getUser('example')).rejects.toThrow('Twitter API user lookup failed (codes: 88)'); + expect(cached.size).toBe(0); + expect(twitterGot).not.toHaveBeenCalled(); + expect(ofetch).toHaveBeenCalledWith('https://api.example.com/graphql/user-by-name', expect.objectContaining({ method: 'GET' })); + }); + + it('does not request tweets or write an empty legacy cache entry when user lookup fails', async () => { + vi.mocked(twitterGot).mockResolvedValueOnce({ errors: [{ code: 88 }] }); + await expect(api.getUserTweets('example')).rejects.toThrow('Twitter API user lookup failed'); + expect(paginationTweets).not.toHaveBeenCalled(); + expect(cache.set).not.toHaveBeenCalled(); + expect(cached.size).toBe(0); + }); + + it('passes a verified user ID to the existing tweet cache and mapper', async () => { + vi.mocked(paginationTweets).mockResolvedValueOnce({ tweets: 'upstream result' }); + vi.mocked(gatherLegacyFromData).mockReturnValueOnce([{ id_str: 'tweet-1' }]); + expect(await api.getUserTweets('example')).toEqual([{ id_str: 'tweet-1' }]); + expect(paginationTweets).toHaveBeenCalledWith('UserTweets', '123', expect.objectContaining({ count: 20 })); + expect(cache.tryGet).toHaveBeenCalledWith('tweet-cache', expect.any(Function), 300, false); + }); +}); diff --git a/tests/twitter-web-user.worker.test.ts b/tests/twitter-web-user.worker.test.ts new file mode 100644 index 0000000000..6e41fa8500 --- /dev/null +++ b/tests/twitter-web-user.worker.test.ts @@ -0,0 +1,2 @@ +// Run user-response validation and cache behavior in workerd as well. +import './twitter-web-user.test'; diff --git a/tests/xiaoyuzhou-podcast.test.ts b/tests/xiaoyuzhou-podcast.test.ts new file mode 100644 index 0000000000..b657300362 --- /dev/null +++ b/tests/xiaoyuzhou-podcast.test.ts @@ -0,0 +1,40 @@ +import { beforeEach, describe, expect, it, vi } from 'vitest'; + +import { route } from '../lib/routes/xiaoyuzhou/podcast'; +import ofetch from '../lib/utils/ofetch'; + +vi.mock('../lib/utils/ofetch', () => ({ default: vi.fn() })); +vi.mock('../lib/utils/cache', () => ({ default: { tryGet: (_key: string, fetchValue: () => unknown) => fetchValue() } })); + +const context = { req: { param: () => 'source-id' } } as any; +const podcastHtml = ``; + +beforeEach(() => vi.resetAllMocks()); + +describe('Xiaoyuzhou podcast and episode resolution', () => { + it('preserves a podcast access error without requesting the same ID as an episode', async () => { + const blocked = Object.assign(new Error('Forbidden'), { status: 403 }); + vi.mocked(ofetch).mockRejectedValue(blocked); + + await expect(route.handler!(context)).rejects.toBe(blocked); + expect(ofetch).toHaveBeenCalledTimes(1); + }); + + it('resolves episode IDs when the podcast URL returns 404', async () => { + vi.mocked(ofetch) + .mockRejectedValueOnce(Object.assign(new Error('Not Found'), { status: 404 })) + .mockResolvedValueOnce('Podcast') + .mockResolvedValueOnce(podcastHtml); + + await expect(route.handler!(context)).resolves.toMatchObject({ title: 'Podcast', link: 'https://www.xiaoyuzhoufm.com/podcast/podcast-id' }); + expect(ofetch).toHaveBeenNthCalledWith(2, 'https://www.xiaoyuzhoufm.com/episode/source-id'); + expect(ofetch).toHaveBeenCalledTimes(3); + }); + + it('does not hide malformed page data behind an unrelated episode request', async () => { + vi.mocked(ofetch).mockResolvedValue('Unexpected response'); + + await expect(route.handler!(context)).rejects.toBeInstanceOf(SyntaxError); + expect(ofetch).toHaveBeenCalledTimes(1); + }); +}); diff --git a/tests/xxhash-wasm.test.ts b/tests/xxhash-wasm.test.ts new file mode 100644 index 0000000000..c57e50f2e4 --- /dev/null +++ b/tests/xxhash-wasm.test.ts @@ -0,0 +1,48 @@ +import { describe, expect, it } from 'vitest'; +import xxhash from 'xxhash-wasm'; + +// Empty/a/abc are standard XXH vectors; the remaining vectors were generated +// with xxhash-wasm 1.1.0's Node entry. This suite also runs via its workerd entry. +const vectors = [ + { input: '', h32: '02cc5d05', h64: 'ef46db3751d8e999', seed32: 'bd209070', seed64: '51e24c0e9077a48c' }, + { input: 'a', h32: '550d7456', h64: 'd24ec4f1a98c6e5b', seed32: '0d5a8e75', seed64: '63a379f532ba23d4' }, + { input: 'abc', h32: '32d153ff', h64: '44bc2cf5ad770999', seed32: '11364062', seed64: '1fc03ef74cebaa7d' }, + { input: '0123456789abcdef'.repeat(5), h32: '3fe8e7af', h64: '0d4cc7d5057880df', seed32: 'e0c25edb', seed64: '3d2aa4692d5507dc' }, + { input: 'RSSHub 中文 🐧', h32: '57c604bf', h64: '24862e5b32f4f03f', seed32: '129c225f', seed64: '163f1ecdcce47f0a' }, + { input: '/gov/zhengce/zhengceku/bmwj:json', h32: 'e008b760', h64: '390f38bd8df8e23e', seed32: 'abc5ecf2', seed64: 'b1183e2c2e5ba95d' }, + { input: '/gov/zhengce/zhengceku/bmwj:json:47', h32: 'f144b479', h64: '24d1ed2f1309ed35', seed32: 'f6461457', seed64: '2675f5df8c4442da' }, +]; + +const hash = await xxhash(); + +describe('official xxhash Node and Worker parity', () => { + it.each(vectors)('matches known 32/64-bit and seeded vectors for $input', (vector) => { + const bytes = new TextEncoder().encode(vector.input); + expect(hash.h32ToString(vector.input)).toBe(vector.h32); + expect(hash.h64ToString(vector.input)).toBe(vector.h64); + expect(hash.h32(vector.input)).toBe(Number.parseInt(vector.h32, 16)); + expect(hash.h64(vector.input)).toBe(BigInt('0x' + vector.h64)); + expect(hash.h32Raw(bytes)).toBe(Number.parseInt(vector.h32, 16)); + expect(hash.h64Raw(bytes)).toBe(BigInt('0x' + vector.h64)); + expect(hash.h32ToString(vector.input, 0x12_34_56_78)).toBe(vector.seed32); + expect(hash.h64ToString(vector.input, 0x01_23_45_67_89_ab_cd_efn)).toBe(vector.seed64); + }); + + it('supports streaming mixed strings and byte subarrays across block boundaries', () => { + const prefix = '0123456789abcdef'.repeat(5); + const bytes = new TextEncoder().encode('skipRSSHub 中文 🐧skip'); + const suffix = bytes.subarray(4, -4); + const h32 = hash.create32(0x12_34_56_78); + const h64 = hash.create64(0x01_23_45_67_89_ab_cd_efn); + for (const chunk of [prefix.slice(0, 7), prefix.slice(7), suffix]) { + h32.update(chunk); + h64.update(chunk); + } + expect(h32.digest()).toBe(hash.h32(prefix + 'RSSHub 中文 🐧', 0x12_34_56_78)); + expect(h64.digest()).toBe(hash.h64(prefix + 'RSSHub 中文 🐧', 0x01_23_45_67_89_ab_cd_efn)); + }); + + it('does not retain the polynomial shim collision', () => { + expect(hash.h64ToString('Aa')).not.toBe(hash.h64ToString('BB')); + }); +}); diff --git a/tests/xxhash-wasm.worker.test.ts b/tests/xxhash-wasm.worker.test.ts new file mode 100644 index 0000000000..1a987605aa --- /dev/null +++ b/tests/xxhash-wasm.worker.test.ts @@ -0,0 +1,2 @@ +// Run the same reference vectors against the package's static-WASM workerd export. +import './xxhash-wasm.test'; diff --git a/tsdown-worker.config.ts b/tsdown-worker.config.ts index 6ca7321b2c..0b7fbb40f6 100644 --- a/tsdown-worker.config.ts +++ b/tsdown-worker.config.ts @@ -84,7 +84,6 @@ export default defineConfig({ 'dotenv/config': path.resolve('./lib/shims/dotenv-config.ts'), '@sentry/node': path.resolve('./lib/shims/sentry-node.ts'), '@honeybadger-io/js': path.resolve('./lib/shims/honeybadger.ts'), - 'xxhash-wasm': path.resolve('./lib/shims/xxhash-wasm.ts'), typescript: path.resolve('./lib/shims/typescript.ts'), // Routes file with Worker-specific build (match relative import from lib/) '../assets/build/routes.js': path.resolve('./assets/build/routes-worker.js'), @@ -94,6 +93,8 @@ export default defineConfig({ deps: { onlyBundle: false, neverBundle: [ + // Let Wrangler select the package's static-WASM workerd export. + 'xxhash-wasm', // Exclude non-code files that might be accidentally imported /\/_README$/, /\.node$/, diff --git a/vitest.workers.config.ts b/vitest.workers.config.ts index a4578e0de0..a2c9ab7da1 100644 --- a/vitest.workers.config.ts +++ b/vitest.workers.config.ts @@ -58,6 +58,7 @@ export default defineConfig({ compatibilityDate: '2026-08-22', compatibilityFlags: ['nodejs_compat'], kvNamespaces: ['CACHE'], + modulesRules: [{ type: 'CompiledWasm', include: ['**/*.wasm'] }], }, }), workerAliasPlugin(), @@ -69,6 +70,6 @@ export default defineConfig({ }, }, test: { - include: ['lib/**/*.worker.test.ts', 'tests/**/*.worker.test.ts', 'lib/utils/parse-script-data.test.ts'], + include: ['lib/**/*.worker.test.ts', 'tests/**/*.worker.test.ts', 'lib/utils/parse-script-data.test.ts', 'lib/utils/parse-date-in-timezone.test.ts', 'tests/source-date-routes.test.ts'], }, }); diff --git a/wrangler.toml b/wrangler.toml index b989c01860..7db2ae2535 100644 --- a/wrangler.toml +++ b/wrangler.toml @@ -1,7 +1,7 @@ name = "rsshub" main = "dist-worker/worker.mjs" compatibility_date = "2026-09-01" -compatibility_flags = ["nodejs_compat"] +compatibility_flags = ["nodejs_compat", "global_fetch_strictly_public"] find_additional_modules = true base_dir = "dist-worker/modules"