fix(route/bloomberg): switch to official RSS feeds as news sitemaps were removed

Bloomberg removed all per-site news sitemaps (/feeds/{site}/sitemap_news.xml
now returns 404), which broke every section of this route.

- Fetch the official RSS feeds (/feeds/{site}/news.rss) instead, via ofetch
  so PROXY_URI and the global request layer still apply
- Update section IDs to match the feeds Bloomberg currently serves, keeping
  bpol/bbiz as backward-compatible aliases of politics/business
- Drop sections Bloomberg no longer provides (green, pursuits, equality,
  citylab); add economics, industries and crypto
- Keep the full-text pipeline; pass the RSS summary through when the story
  API is blocked so degraded items still have a description

Fixes #22787

lib/routes/bloomberg/index.ts

@@ -2,29 +2,37 @@ import pMap from 'p-map';
 
 import type { Route } from '@/types';
 import { ViewType } from '@/types';
+import ofetch from '@/utils/ofetch';
+import { parseDate } from '@/utils/parse-date';
+import rssParser from '@/utils/rss-parser';
 
-import { parseArticle, parseNewsList, rootUrl } from './utils';
+import { parseArticle, rootUrl } from './utils';
 
 const siteTitleMapping = {
     '/': 'News',
-    bpol: 'Politics',
-    bbiz: 'Business',
+    politics: 'Politics',
+    business: 'Business',
     markets: 'Markets',
     technology: 'Technology',
-    green: 'Green',
     wealth: 'Wealth',
-    pursuits: 'Pursuits',
     bview: 'Opinion',
-    equality: 'Equality',
     businessweek: 'Businessweek',
-    citylab: 'CityLab',
+    economics: 'Economics',
+    industries: 'Industries',
+    crypto: 'Crypto',
+};
+
+// Bloomberg removed the per-site news sitemaps; map legacy site IDs to the equivalent RSS feeds
+const legacySiteMapping = {
+    bpol: 'politics',
+    bbiz: 'business',
 };
 
 export const route: Route = {
     path: '/:site?',
     categories: ['finance'],
     view: ViewType.Articles,
-    example: '/bloomberg/bbiz',
+    example: '/bloomberg/business',
     parameters: {
         site: {
             description: 'Site ID, can be found below',
@@ -44,29 +52,37 @@ export const route: Route = {
     description: `| Site ID      | Title        |
 | ------------ | ------------ |
 | /            | News         |
-| bpol         | Politics     |
-| bbiz         | Business     |
+| politics     | Politics     |
+| business     | Business     |
 | markets      | Markets      |
 | technology   | Technology   |
-| green        | Green        |
 | wealth       | Wealth       |
-| pursuits     | Pursuits     |
 | bview        | Opinion      |
-| equality     | Equality     |
 | businessweek | Businessweek |
-| citylab      | CityLab      |`,
+| economics    | Economics    |
+| industries   | Industries   |
+| crypto       | Crypto       |
+
+Legacy site IDs \`bpol\` and \`bbiz\` still work as aliases of \`politics\` and \`business\`.`,
     handler,
 };
 
 async function handler(ctx) {
     const site = ctx.req.param('site');
-    const currentUrl = site ? `${rootUrl}/${site}/sitemap_news.xml` : `${rootUrl}/sitemap_news.xml`;
+    const mappedSite = site ? (legacySiteMapping[site] ?? site) : undefined;
+    const currentUrl = mappedSite ? `${rootUrl}/${mappedSite}/news.rss` : `${rootUrl}/news.rss`;
 
-    const list = await parseNewsList(currentUrl, ctx);
+    const feed = await rssParser.parseString(await ofetch(currentUrl));
+    const list = feed.items.slice(0, ctx.req.query('limit') ? Number.parseInt(ctx.req.query('limit')) : 50).map((item) => ({
+        title: item.title,
+        link: item.link,
+        pubDate: item.pubDate ? parseDate(item.pubDate) : undefined,
+        description: item.content,
+    }));
     const items = await pMap(list, (item) => parseArticle(item), { concurrency: 1 });
     return {
-        title: `Bloomberg - ${siteTitleMapping[site ?? '/']}`,
-        link: currentUrl,
+        title: `Bloomberg - ${siteTitleMapping[mappedSite ?? '/'] ?? feed.title}`,
+        link: feed.link ?? currentUrl,
         item: items,
     };
 }

lib/routes/bloomberg/utils.ts

@@ -2,7 +2,6 @@ import { load } from 'cheerio';
 import { destr } from 'destr';
 
 import cache from '@/utils/cache';
-import got from '@/utils/got';
 import ofetch from '@/utils/ofetch';
 import { parseDate } from '@/utils/parse-date';
 
@@ -71,28 +70,6 @@ const redirectGot = (url) =>
         }),
     });
 
-const parseNewsList = async (url, ctx) => {
-    const resp = await got(url);
-    const $ = load(resp.data, {
-        xml: {
-            xmlMode: true,
-        },
-    });
-    const urls = $('urlset url');
-    return urls
-        .toArray()
-        .slice(0, ctx.req.query('limit') ? Number.parseInt(ctx.req.query('limit')) : 50)
-        .map((u) => {
-            u = $(u);
-            const item = {
-                title: u.find(String.raw`news\:title`).text(),
-                link: u.find('loc').text(),
-                pubDate: parseDate(u.find(String.raw`news\:publication_date`).text()),
-            };
-            return item;
-        });
-};
-
 const parseArticle = (item) =>
     cache.tryGet(item.link, async () => {
         const group = regex
@@ -119,6 +96,7 @@ const parseArticle = (item) =>
                                 title: item.title,
                                 link: item.link,
                                 pubDate: item.pubDate,
+                                description: item.description,
                             };
                         }
                     }
@@ -130,6 +108,7 @@ const parseArticle = (item) =>
                         title: item.title,
                         link: item.link,
                         pubDate: item.pubDate,
+                        description: item.description,
                     };
                 }
 
@@ -611,4 +590,4 @@ const documentToHtmlString = async (document) => {
     return str;
 };
 
-export { parseArticle, parseNewsList, rootUrl };
+export { parseArticle, rootUrl };