[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_162":163,"_164":165},"routes/_app.x.$handle.$name",{"_99":98,"_32":44,"_108":89,"_27":53,"_13":14,"_109":-5,"_11":110,"_111":112,"_113":114,"_115":116,"_117":118,"_20":-5,"_119":-5,"_120":-5,"_121":48,"_122":123,"_124":25,"_125":-5,"_126":-5,"_127":128},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_42":43,"_32":44,"_37":45,"_46":-5,"_47":48,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":57,"_58":25,"_59":-5,"_60":-5,"_61":62,"_63":64,"_65":66,"_67":68,"_69":70,"_71":-5,"_72":73,"_74":73,"_75":76,"_77":76,"_78":76,"_79":73,"_80":73,"_81":82,"_83":84,"_85":86,"_87":48},"version",9,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":34,"_35":36,"_37":38,"_39":14,"_40":41,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",true,"deletedHere","path","main.ts","latestPath","fileId","2bf12eae-a13f-11f1-9fbe-1607ee4eb77e","name","fileType","script","content","import * as cheerio from \"npm:cheerio@1.0.0\";\nimport { Feed } from \"npm:feed@4.2.2\";\n\nconst SOURCE_URL = \"https://www.watson.ch\";\nconst USER_AGENT =\n \"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36\";\n\ninterface Article {\n title: string;\n link: string;\n}\n\n/**\n * Watson-Artikel-URLs folgen dem Muster\n * ///- (id = mind. 6 Ziffern)\n * z.B. /international/analyse/172909310-cia-direktor-in-moskau-...\n */\nconst ARTICLE_PATH_RE = /\\/\\d{6,}-[a-z0-9-]+\\/?$/i;\n\n/**\n * Ermittelt aus einem -Element den wahrscheinlichsten Artikel-Titel.\n * Watson rendert in den Teasern oft mehrere Text-Fragmente im selben Link\n * (Kommentarzahl, Rubrik-Badge, Autor, eigentlicher Titel). Da sich CSS-\n * Klassen jederzeit ändern können, nehmen wir robust das längste\n * eigenständige Text-Fragment als Titel.\n */\nfunction extractTitle(\n $: cheerio.CheerioAPI,\n el: cheerio.Element,\n): string | null {\n const $el = $(el);\n\n const titleAttr = $el.attr(\"title\")?.trim();\n if (titleAttr) return titleAttr;\n\n const texts: string[] = [];\n $el\n .find(\"*\")\n .addBack()\n .contents()\n .each((_, node) => {\n if (node.type === \"text\") {\n const t = $(node).text().replace(/\\s+/g, \" \").trim();\n if (t) texts.push(t);\n }\n });\n\n if (texts.length === 0) return null;\n\n texts.sort((a, b) => b.length - a.length);\n return texts[0];\n}\n\nasync function scrapeArticles(limit: number): Promise {\n const res = await fetch(SOURCE_URL, {\n headers: {\n \"User-Agent\": USER_AGENT,\n Accept: \"text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8\",\n \"Accept-Language\": \"de-CH,de;q=0.9,en;q=0.8\",\n },\n });\n\n if (!res.ok) {\n throw new Error(\n `Watson-Seite konnte nicht geladen werden: HTTP ${res.status}`,\n );\n }\n\n const html = await res.text();\n const $ = cheerio.load(html);\n\n const seen = new Set();\n const articles: Article[] = [];\n\n $(\"a[href]\").each((_, el) => {\n if (articles.length >= limit) return;\n\n const href = $(el).attr(\"href\");\n if (!href) return;\n\n let url: URL;\n try {\n url = new URL(href, SOURCE_URL);\n } catch {\n return;\n }\n\n if (url.hostname.replace(/^www\\./, \"\") !== \"watson.ch\") return;\n if (!ARTICLE_PATH_RE.test(url.pathname)) return;\n\n const cleanLink = `${url.origin}${url.pathname}`;\n if (seen.has(cleanLink)) return;\n\n const title = extractTitle($, el);\n if (!title || title.length < 8) return;\n\n seen.add(cleanLink);\n articles.push({ title, link: cleanLink });\n });\n\n return articles.slice(0, limit);\n}\n\nexport default async function (req: Request): Promise {\n try {\n const articles = await scrapeArticles(15);\n\n const feed = new Feed({\n title: \"watson.ch – Neueste Artikel\",\n description:\n \"Automatisch generierter RSS-Feed der neuesten Artikel von watson.ch\",\n id: SOURCE_URL,\n link: SOURCE_URL,\n language: \"de-CH\",\n updated: new Date(),\n generator: \"Val.town RSS Scraper (cheerio + feed)\",\n feedLinks: {\n rss: req.url,\n },\n });\n\n for (const article of articles) {\n feed.addItem({\n title: article.title,\n id: article.link,\n link: article.link,\n date: new Date(),\n });\n }\n\n return new Response(feed.rss2(), {\n status: 200,\n headers: {\n \"Content-Type\": \"application/rss+xml; charset=utf-8\",\n },\n });\n } catch (err) {\n return new Response(\n `Fehler beim Erzeugen des RSS-Feeds: ${\n err instanceof Error ? err.message : String(err)\n }`,\n { status: 500, headers: { \"Content-Type\": \"text/plain; charset=utf-8\" } },\n );\n }\n}","createdAt",["D",1787752617335],"branchVersion","branchId","590b2e9e-9bd9-4fc3-97be-a934c93ca018","id","373c21a1-89a9-44a3-b341-eee81cc855a9","watson-rss-new",["D",1787742811709],"deletedAt","legacyEnvVars",false,"privacy","public","httpPrivacy","description","","imageUrl","dbf2ae7a-e596-4450-9e80-f48182352e08","pinnedFileId","09254ade-a150-11f1-afb1-1607ee4eb77e","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","01a03d75-a2dc-7129-bcfe-464f7359f782","tursoName","373c21a189a944a3b341eee81cc855a9","tursoHostname","373c21a189a944a3b341eee81cc855a9-val-town.aws-us-east-2.turso.io","tursoMeasuredAt",["D",1789485645684],"pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_99":98,"_107":103},"org",{"_99":98,"_100":101,"_102":103,"_18":55,"_104":105,"_106":48},"branch",{"_13":88,"_32":89,"_42":41,"_90":43,"_91":-5,"_92":-5,"_93":-5,"_94":-5,"_95":-5,"_96":44,"_97":98},"isOwned",15,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","mikefrischknecht","handle","displayName","Mike Frischknecht","avatar","https://lh3.googleusercontent.com/a/ACg8ocJYhZ6TlQC9Px9BhWVC2KxGG5zXrx2pRqdDuxsgIfio5QgUBBEd=s96-c","tier","free","canRestrictVals","profileImageUrl","branchName","adoptOrgId",{"_42":43,"_32":44,"_37":157,"_46":-5,"_47":48,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":57,"_58":25,"_59":-5,"_60":-5,"_61":62,"_63":64,"_65":66,"_67":68,"_69":158,"_71":-5,"_72":73,"_74":73,"_75":76,"_77":76,"_78":76,"_79":73,"_80":73,"_81":159,"_83":160,"_85":161,"_87":48},"tree",[152],"files",[152],"counts",{"_141":62,"_142":143,"_144":62,"_145":146,"_147":148},"contents",[129],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_42":130,"_32":131,"_30":57,"_132":-5,"_40":41,"_39":133,"_134":48,"_37":135,"_13":136,"_33":137,"_35":138,"_24":48,"_139":-5,"_27":140},"c0b88c42-3cb9-43e6-b4b1-245317ca2fd0","watson.ts","parentId",8,"latest",["D",1787752607597],5,"http","import * as cheerio from \"npm:cheerio@1.0.0\";\nimport { Feed } from \"npm:feed@4.2.2\";\n\nconst SOURCE_URL = \"https://www.watson.ch\";\nconst USER_AGENT =\n \"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36\";\n\n// Nur Artikel der letzten X Tage aufnehmen (statt starrer Anzahl)\nconst DAYS_BACK = 3;\n// Sicherheitsobergrenze, falls an einem Tag aussergewöhnlich viel publiziert wird\nconst MAX_ITEMS = 250;\n// Wie viele Artikelseiten gleichzeitig für Bild/Text/Video abgerufen werden\nconst CONCURRENCY = 12;\n\n// Pfade, die keine redaktionellen News sind (Promotion/Sponsored Content, Kontaktseiten etc.)\nconst EXCLUDED_PATH_PREFIXES = [\n \"/promotion/\",\n \"/commercialization/\",\n \"/vermarktung/\",\n \"/sport/\",\n];\n\n// Watson-Artikel-URLs enden auf /- (id = mind. 6 Ziffern), z.B.\n// /international/analyse/172909310-cia-direktor-in-moskau-...\nconst ARTICLE_PATH_RE = /\\/(\\d{6,})-([a-z0-9-]+)\\/?$/i;\n\ninterface Article {\n title: string;\n link: string;\n lastmod?: string;\n description?: string;\n image?: string;\n video?: string;\n isVideo?: boolean;\n}\n\nfunction isExcludedPath(pathname: string): boolean {\n return EXCLUDED_PATH_PREFIXES.some((p) => pathname.startsWith(p));\n}\n\n/** Fallback-Titel direkt aus dem URL-Slug, falls die Artikelseite nicht abrufbar ist. */\nfunction titleFromSlug(pathname: string): string {\n const match = pathname.match(ARTICLE_PATH_RE);\n const slug = match ? match[2] : pathname;\n return slug\n .split(\"-\")\n .filter(Boolean)\n .map((w) => w.charAt(0).toUpperCase() + w.slice(1))\n .join(\" \");\n}\n\nfunction guessImageType(url: string): string {\n const ext = url.split(\"?\")[0].split(\".\").pop()?.toLowerCase();\n switch (ext) {\n case \"png\":\n return \"image/png\";\n case \"webp\":\n return \"image/webp\";\n case \"gif\":\n return \"image/gif\";\n default:\n return \"image/jpeg\";\n }\n}\n\nfunction guessVideoType(url: string): string {\n const ext = url.split(\"?\")[0].split(\".\").pop()?.toLowerCase();\n switch (ext) {\n case \"webm\":\n return \"video/webm\";\n case \"mov\":\n return \"video/quicktime\";\n default:\n return \"video/mp4\";\n }\n}\n\n/** Einfacher Concurrency-Pool, damit nicht alle Artikelseiten gleichzeitig abgerufen werden. */\nasync function runWithConcurrency(\n items: T[],\n limit: number,\n worker: (item: T) => Promise,\n): Promise {\n const results: R[] = new Array(items.length);\n let index = 0;\n\n async function next(): Promise {\n const current = index++;\n if (current >= items.length) return;\n results[current] = await worker(items[current]);\n return next();\n }\n\n const runners = Array.from(\n { length: Math.min(limit, items.length) },\n () => next(),\n );\n await Promise.all(runners);\n return results;\n}\n\n/** Liest die offizielle Watson-Sitemap eines Monats aus (enthält ALLE Artikel-URLs + Datum). */\nasync function fetchSitemapMonth(\n year: number,\n month: number,\n): Promise<{ link: string; lastmod?: string }[]> {\n const dateParam = `${year}${String(month).padStart(2, \"0\")}`;\n try {\n const res = await fetch(`${SOURCE_URL}/sitemap.xml?date=${dateParam}`, {\n headers: {\n \"User-Agent\": USER_AGENT,\n Accept: \"application/xml,text/xml,*/*\",\n },\n });\n if (!res.ok) return [];\n const xml = await res.text();\n const $ = cheerio.load(xml, { xmlMode: true });\n const results: { link: string; lastmod?: string }[] = [];\n $(\"url\").each((_, el) => {\n const loc = $(el).find(\"loc\").text().trim();\n const lastmod = $(el).find(\"lastmod\").text().trim();\n if (loc) results.push({ link: loc, lastmod: lastmod || undefined });\n });\n return results;\n } catch {\n return [];\n }\n}\n\n/** Fallback: Startseite scrapen, falls die Sitemap mal nicht erreichbar sein sollte. */\nasync function collectFromHomepage(): Promise {\n try {\n const res = await fetch(SOURCE_URL, {\n headers: {\n \"User-Agent\": USER_AGENT,\n Accept: \"text/html,application/xhtml+xml\",\n },\n });\n if (!res.ok) return [];\n const html = await res.text();\n const $ = cheerio.load(html);\n const seen = new Set();\n const items: Article[] = [];\n\n $(\"a[href]\").each((_, el) => {\n const href = $(el).attr(\"href\");\n if (!href) return;\n let url: URL;\n try {\n url = new URL(href, SOURCE_URL);\n } catch {\n return;\n }\n if (url.hostname.replace(/^www\\./, \"\") !== \"watson.ch\") return;\n if (isExcludedPath(url.pathname)) return;\n if (!ARTICLE_PATH_RE.test(url.pathname)) return;\n\n const cleanLink = `${url.origin}${url.pathname}`;\n if (seen.has(cleanLink)) return;\n seen.add(cleanLink);\n\n items.push({\n title: titleFromSlug(url.pathname),\n link: cleanLink,\n isVideo: url.pathname.includes(\"/videos/\"),\n });\n });\n\n return items;\n } catch {\n return [];\n }\n}\n\n/** Sammelt alle Artikel-Links der letzten DAYS_BACK Tage aus der/den Sitemap(s). */\nasync function discoverArticles(): Promise {\n const now = new Date();\n const cutoff = new Date(now.getTime() - DAYS_BACK * 24 * 60 * 60 * 1000);\n\n const thisMonth = { y: now.getUTCFullYear(), m: now.getUTCMonth() + 1 };\n const prevRef = new Date(\n Date.UTC(now.getUTCFullYear(), now.getUTCMonth() - 1, 1),\n );\n const prevMonth = {\n y: prevRef.getUTCFullYear(),\n m: prevRef.getUTCMonth() + 1,\n };\n\n const [current, previous] = await Promise.all([\n fetchSitemapMonth(thisMonth.y, thisMonth.m),\n // Vormonat nur nötig, falls wir gerade am Monatsanfang stehen (DAYS_BACK reicht sonst)\n now.getUTCDate() <= DAYS_BACK\n ? fetchSitemapMonth(prevMonth.y, prevMonth.m)\n : Promise.resolve([]),\n ]);\n\n const seen = new Set();\n const items: Article[] = [];\n\n for (const entry of [...current, ...previous]) {\n if (entry.lastmod) {\n const modDate = new Date(entry.lastmod);\n if (!isNaN(modDate.getTime()) && modDate < cutoff) continue;\n }\n\n let url: URL;\n try {\n url = new URL(entry.link);\n } catch {\n continue;\n }\n if (url.hostname.replace(/^www\\./, \"\") !== \"watson.ch\") continue;\n if (isExcludedPath(url.pathname)) continue;\n if (!ARTICLE_PATH_RE.test(url.pathname)) continue;\n\n const cleanLink = `${url.origin}${url.pathname}`;\n if (seen.has(cleanLink)) continue;\n seen.add(cleanLink);\n\n items.push({\n title: titleFromSlug(url.pathname),\n link: cleanLink,\n lastmod: entry.lastmod,\n isVideo: url.pathname.includes(\"/videos/\"),\n });\n }\n\n // Fallback, falls die Sitemap aus irgendeinem Grund leer zurückkommt\n if (items.length === 0) {\n return (await collectFromHomepage()).slice(0, MAX_ITEMS);\n }\n\n items.sort((a, b) => {\n const da = a.lastmod ? new Date(a.lastmod).getTime() : 0;\n const db = b.lastmod ? new Date(b.lastmod).getTime() : 0;\n return db - da;\n });\n\n return items.slice(0, MAX_ITEMS);\n}\n\n/**\n * Ruft eine Artikelseite ab und liest Titel/Beschreibung/Bild (Open Graph)\n * sowie – falls vorhanden – eine echte Video-Quelle aus (JSON-LD VideoObject,\n *

Diff · v9