${cdata(article.title)}

{\n const controller = new AbortController();\n const timeout = setTimeout(() => controller.abort(), 12000);\n try {\n const res = await fetch(article.link, {\n headers: {\n \"User-Agent\": USER_AGENT,\n Accept: \"text/html,application/xhtml+xml\",\n \"Accept-Language\": \"de-DE,de;q=0.9\",\n },\n signal: controller.signal,\n });\n if (!res.ok) return article;\n\n const $ = cheerio.load(await res.text());\n return {\n ...article,\n image: $('meta[property=\"og:image\"]').attr(\"content\")?.trim(),\n description: article.description ||\n $('meta[property=\"og:description\"]').attr(\"content\")?.trim(),\n content: cleanContent($, article.link),\n };\n } catch {\n return article;\n } finally {\n clearTimeout(timeout);\n }\n}\n\nexport default async function (req: Request): Promise {\n try {\n const candidates = await discoverArticles();\n const articles = await runWithConcurrency(candidates, CONCURRENCY, enrichArticle);\n articles.sort((a, b) => (b.pubDate?.getTime() ?? 0) - (a.pubDate?.getTime() ?? 0));\n\n const now = new Date();\n const itemsXml = articles.map((article) => {\n const date = article.pubDate ?? now;\n const summary = `

${\n article.image ? `

` : \"\"\n }

${article.description ?? \"\"}

`;\n\n return [\n \" \",\n ` ${cdata(article.title)}`,\n ` ${cdata(summary)}`,\n article.content ? ` ${cdata(article.content)}` : \"\",\n ` ${escapeXml(article.link)}`,\n ` ${escapeXml(article.link)}#${FEED_REVISION}`,\n article.author ? ` ${cdata(article.author)}` : \"\",\n ...article.categories.map((c) => ` ${cdata(c)}`),\n ` ${date.toUTCString()}`,\n \" \",\n ].filter(Boolean).join(\"\\n\");\n }).join(\"\\n\");\n\n const xml = `\n\n \n ${cdata(FEED_TITLE)}\n ${cdata(FEED_DESCRIPTION)}\n ${SOURCE_URL}/\n Val.town RSS Scraper (cheerio)\n ${now.toUTCString()}\n \n de\n${itemsXml}\n \n`;\n\n return new Response(xml, {\n status: 200,\n headers: {\n \"Content-Type\": \"application/rss+xml; charset=utf-8\",\n \"Cache-Control\": \"public, max-age=300\",\n },\n });\n } catch (err) {\n return new Response(\n `Fehler beim Erzeugen des RSS-Feeds: ${err instanceof Error ? err.message : String(err)}`,\n { status: 500, headers: { \"Content-Type\": \"text/plain; charset=utf-8\" } },\n );\n }\n}\n","import * as cheerio from \"npm:cheerio@1.0.0\";\nimport { XMLParser } from \"npm:fast-xml-parser@4.4.1\";\n\nconst SOURCE_URL = \"https://esc-kompakt.de\";\nconst FEED_TITLE = \"ESC kompakt\";\nconst FEED_DESCRIPTION =\n \"Alles zum Eurovision Song Contest und zu den nationalen ESC-Vorentscheidungen – als Volltext-Feed mit Bildern und Videos.\";\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// Der eingebaute WordPress-Feed liefert 10 Artikel pro Seite.\nconst FEED_PAGES = 3;\nconst MAX_ITEMS = 30;\nconst CONCURRENCY = 6;\nconst MAX_CONTENT_CHARS = 300_000;\n\nconst FEED_REVISION = \"v1\";\n\n/**\n * Bausteine, die WordPress in den Artikeltext einstreut, aber nicht dazu\n * gehören: Teilen-Buttons, Like-Widgets, verwandte Beiträge, Newsletter-Box\n * und die Werbeplatzierungen der Seite.\n */\nconst JUNK_SELECTOR = [\n \"script\", \"style\", \"noscript\", \"form\",\n \".sharedaddy\", \".sd-block\", \".jp-relatedposts\", \"#jp-relatedposts\",\n \".wpcnt\", \".wpa\", \".adsbygoogle\", \"ins.adsbygoogle\",\n \".code-block\", \".wp-polls-loading\", \".wp-block-buttons\",\n \".addtoany_share_save_container\", \".post-views\", \".entry-meta\",\n // Werbeplätze von ESC kompakt (Banner zwischen den Absätzen)\n '[class*=\"kompa-entity-placement\"]', '[class*=\"gas_fallback\"]', '[class*=\"-ad_\"]',\n // Newsletter-Kasten am Artikelende\n \".wp-block-group.has-border-color\",\n // Unsichtbarer VG-Wort-Zählpixel – Reader nehmen ihn sonst als Vorschaubild\n 'img[src*=\"vgwort.de\"]', 'img[src*=\"vg0\"][src*=\"met.\"]',\n].join(\", \");\n\ninterface Article {\n title: string;\n link: string;\n description?: string;\n image?: string;\n author?: string;\n categories: string[];\n content?: string;\n pubDate?: Date;\n}\n\nfunction cdata(value: string): string {\n return `/g, \"]]]]>\")}]]>`;\n}\n\nfunction escapeXml(value: string): string {\n return value.replace(/&/g, \"&\").replace(//g, \">\");\n}\n\nfunction absolute(url: string, base = SOURCE_URL): string {\n try {\n if (url.startsWith(\"//\")) return `https:${url}`;\n return new URL(url, base).href;\n } catch {\n return \"\";\n }\n}\n\n/** Dateityp des Vorschaubilds für bestimmen. */\nfunction imageMime(url: string): string {\n const path = url.split(\"?\")[0].toLowerCase();\n if (path.endsWith(\".png\")) return \"image/png\";\n if (path.endsWith(\".gif\")) return \"image/gif\";\n if (path.endsWith(\".webp\")) return \"image/webp\";\n return \"image/jpeg\";\n}\n\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 await Promise.all(Array.from({ length: Math.min(limit, items.length) }, () => next()));\n return results;\n}\n\n/**\n * Liest die Artikelliste aus dem eingebauten WordPress-Feed. Der liefert\n * Titel, Link, Autor, Datum und Rubriken bereits sauber strukturiert – das\n * ist zuverlässiger, als die Startseite zu zerlegen.\n */\nasync function discoverArticles(): Promise {\n const parser = new XMLParser({ ignoreAttributes: false, cdataPropName: \"__cdata\" });\n const text = (v: unknown): string =>\n String(typeof v === \"string\" ? v : (v as { __cdata?: string })?.__cdata ?? \"\");\n\n const seen = new Set();\n const items: Article[] = [];\n\n for (let page = 1; page <= FEED_PAGES; page++) {\n const url = page === 1 ? `${SOURCE_URL}/feed/` : `${SOURCE_URL}/feed/?paged=${page}`;\n const res = await fetch(url, {\n headers: {\n \"User-Agent\": USER_AGENT,\n Accept: \"application/rss+xml,application/xml\",\n \"Cache-Control\": \"no-cache\",\n },\n });\n if (!res.ok) break;\n\n const parsed = parser.parse(await res.text());\n const entries = parsed?.rss?.channel?.item;\n const list = Array.isArray(entries) ? entries : entries ? [entries] : [];\n if (list.length === 0) break;\n\n for (const entry of list) {\n const link = String(entry.link ?? \"\").trim();\n if (!link || seen.has(link)) continue;\n seen.add(link);\n\n const cats = entry.category;\n const categories = (Array.isArray(cats) ? cats : cats ? [cats] : [])\n .map((c: unknown) => text(c)).filter(Boolean);\n\n const date = new Date(String(entry.pubDate ?? \"\"));\n items.push({\n title: text(entry.title).trim(),\n link,\n author: text(entry[\"dc:creator\"]).trim(),\n description: text(entry.description).replace(/<[^>]*>/g, \" \").replace(/\\s+/g, \" \").trim(),\n categories,\n pubDate: isNaN(date.getTime()) ? undefined : date,\n });\n }\n }\n\n return items.slice(0, MAX_ITEMS);\n}\n\n/**\n * Säubert den Artikeltext: Fremdbausteine raus, Attribute auf das Nötige\n * reduzieren, relative Adressen absolut machen. Struktur, Bilder, Listen\n * und Video-Einbettungen bleiben erhalten.\n */\nfunction cleanContent($: cheerio.CheerioAPI, pageUrl: string): string {\n const body = $(\".entry-content\").first();\n if (body.length === 0) return \"\";\n\n const fragment = cheerio.load(`

${body.html() ?? \"\"}

`, null, false);\n fragment(JUNK_SELECTOR).remove();\n\n /**\n * Instagram-Einbettungen bestehen aus einem Blockquote, das erst ein\n * Skript in einen Beitrag verwandelt. Im Reader bleibt davon nur\n * \"View this post on Instagram\" übrig. Instagram gibt den Inhalt ohne\n * Login nicht heraus – auch nicht über die Embed-Adresse. Deshalb wird\n * daraus ein klarer Link auf den Beitrag.\n */\n fragment(\"blockquote.instagram-media\").each((_, node) => {\n const $bq = fragment(node as never);\n const permalink = ($bq.attr(\"data-instgrm-permalink\") ??\n $bq.find('a[href*=\"instagram.com\"]').first().attr(\"href\") ?? \"\").split(\"?\")[0];\n if (!permalink) {\n $bq.remove();\n return;\n }\n const isReel = /\\/reel\\//.test(permalink);\n $bq.replaceWith(\n `

📷 ${isReel ? \"Instagram-Reel\" : \"Instagram-Beitrag\"} ansehen

`,\n );\n });\n\n fragment(\"*\").each((_, node) => {\n const el = node as unknown as { tagName?: string; attribs?: Record };\n const attrs = el.attribs ?? {};\n const kept: Record = {};\n\n switch (el.tagName) {\n case \"img\": {\n // Lazy-Loading: echte Adresse steht oft in data-src\n const src = attrs[\"data-src\"] || attrs[\"data-lazy-src\"] || attrs.src || \"\";\n const abs = absolute(src, pageUrl);\n if (abs) kept.src = abs;\n if (attrs.alt) kept.alt = attrs.alt;\n break;\n }\n case \"a\": {\n const abs = absolute(attrs.href ?? \"\", pageUrl);\n if (abs) kept.href = abs;\n break;\n }\n case \"iframe\": {\n const src = absolute(attrs[\"data-src\"] || attrs.src || \"\", pageUrl);\n // Nur echte Medien-Einbettungen behalten\n if (src && /youtube|youtu\\.be|vimeo|dailymotion|spotify|soundcloud/i.test(src)) {\n kept.src = src;\n kept.allowfullscreen = \"\";\n kept.frameborder = \"0\";\n }\n break;\n }\n default:\n break;\n }\n\n el.attribs = kept;\n });\n\n // Leere iframes (z.B. Like-Widgets) entfernen\n fragment(\"iframe:not([src])\").remove();\n\n const html = (fragment(\"div\").first().html() ?? \"\").trim();\n return html.length > MAX_CONTENT_CHARS ? html.slice(0, MAX_CONTENT_CHARS) : html;\n}\n\nasync function enrichArticle(article: Article): Promise

${\n article.image ? `

` : \"\"\n }

${article.description ?? \"\"}

`;\n\n return [\n \" \",\n ` ${cdata(article.title)}`,\n ` ${cdata(summary)}`,\n article.content ? ` ${cdata(article.content)}` : \"\",\n ` ${escapeXml(article.link)}`,\n ` ${escapeXml(article.link)}#${FEED_REVISION}`,\n article.author ? ` ${cdata(article.author)}` : \"\",\n ...article.categories.map((c) => ` ${cdata(c)}`),\n ` ${date.toUTCString()}`,\n \" \",\n ].filter(Boolean).join(\"\\n\");\n }).join(\"\\n\");\n\n const xml = `\n\n \n ${cdata(FEED_TITLE)}\n ${cdata(FEED_DESCRIPTION)}\n ${SOURCE_URL}/\n Val.town RSS Scraper (cheerio)\n ${now.toUTCString()}\n \n de\n${itemsXml}\n \n`;\n\n return new Response(xml, {\n status: 200,\n headers: {\n \"Content-Type\": \"application/rss+xml; charset=utf-8\",\n \"Cache-Control\": \"public, max-age=300\",\n },\n });\n } catch (err) {\n return new Response(\n `Fehler beim Erzeugen des RSS-Feeds: ${err instanceof Error ? err.message : String(err)}`,\n { status: 500, headers: { \"Content-Type\": \"text/plain; charset=utf-8\" } },\n );\n }\n}\n","createdAt",["D",1789289115217],"branchVersion","branchId","d47ce782-fe24-45e3-b869-1a2b738fd3d7","id","5cbcbe7c-c1eb-40e5-b729-b0526f1625fa","esc-kompakt",["D",1787812384677],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","","imageUrl","dbf2ae7a-e596-4450-9e80-f48182352e08","pinnedFileId","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","01a03d75-a2dc-7129-bcfe-464f7359f782","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_95":94,"_103":99},"org",{"_95":94,"_96":97,"_98":99,"_18":55,"_100":101,"_102":25},"branch",{"_13":84,"_32":85,"_43":42,"_86":44,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":45,"_93":94},"isOwned",20,"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",{"_43":44,"_32":45,"_38":173,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":-5,"_57":25,"_58":-5,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":174,"_79":175,"_81":176,"_83":25},"tree",[160,161,162],"files",[160,161,162],"counts",{"_150":61,"_151":133,"_152":61,"_153":154,"_155":156},"contents",[126,127,128],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles",true,"grants","orgs","customDomains",{},{"_43":143,"_32":144,"_30":145,"_132":-5,"_41":42,"_40":146,"_134":25,"_38":147,"_13":133,"_33":136,"_35":148,"_24":25,"_137":-5,"_27":149},{"_43":139,"_32":28,"_30":31,"_132":-5,"_41":42,"_40":14,"_134":25,"_38":140,"_13":141,"_33":34,"_35":37,"_24":25,"_137":-5,"_27":142},{"_43":129,"_32":130,"_30":131,"_132":-5,"_41":42,"_40":133,"_134":121,"_38":135,"_13":61,"_33":136,"_35":53,"_24":25,"_137":-5,"_27":138},"47690aa7-22a5-4bdc-b4b6-bf5c668e0476","main.ts","28a90a26-a1e1-11f1-9a97-1607ee4eb77e","parentId",1,"latest",["D",1787812384700],"script","label",[130],"76deb24f-c63d-40fe-bf68-66b8c88a390a",["D",1789289115217],3,[28],"7aca4e30-3f89-4d35-9cd5-7b0a558d3eae","debug.ts","4b08f9b6-af4f-11f1-8e6a-1607ee4eb77e",12,["D",1789289077898],"import * as cheerio from \"npm:cheerio@1.0.0\";\n\nconst UA =\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\nfor (const u of [\n \"https://www.instagram.com/p/DdDw259iAJI/embed/captioned/\",\n \"https://www.instagram.com/reel/DdE8DzGsJrY/embed/\",\n]) {\n try {\n const r = await fetch(u, { headers: { \"User-Agent\": UA, Accept: \"text/html\" } });\n const b = await r.text();\n console.log(\"===\", u, \"|\", r.status, \"|\", b.length, \"bytes\");\n const $ = cheerio.load(b);\n const imgs = $(\"img\").map((_, el) => ($(el).attr(\"src\") ?? \"\").slice(0, 100)).get();\n console.log(\"imgs:\", imgs.length, JSON.stringify(imgs.slice(0, 3), null, 1));\n console.log(\"caption:\", $(\".Caption, .CaptionText\").first().text().replace(/\\s+/g, \" \").slice(0, 200));\n console.log(\"login-wall?\", /login|Anmelden/i.test(b.slice(0, 3000)));\n console.log(\"shared_data?\", b.includes(\"contextJSON\") || b.includes(\"display_url\"));\n } catch (e) {\n console.log(\"===\", u, \"-> fehler\", String(e).slice(0, 100));\n }\n}\n",[144],"pulls","branches","forks","envVars",{"_157":61,"_159":61},"connections",{"_157":61,"_158":61},"count","unfilledCount","emptyCount",{"_43":143,"_32":144,"_30":145,"_132":-5,"_41":42,"_40":146,"_134":25,"_38":170,"_13":133,"_33":136,"_35":148,"_24":25,"_137":-5,"_164":171,"_27":172},{"_43":139,"_32":28,"_30":31,"_132":-5,"_41":42,"_40":14,"_134":25,"_38":167,"_13":141,"_33":34,"_35":37,"_24":25,"_137":-5,"_164":168,"_27":169},{"_43":129,"_32":130,"_30":131,"_132":-5,"_41":42,"_40":133,"_134":121,"_38":163,"_13":61,"_33":136,"_35":53,"_24":25,"_137":-5,"_164":165,"_27":166},["D",1787812384700],"children",[],[130],["D",1789289115217],[],[28],["D",1789289077898],[],[144],["D",1787812384677],{"_95":94,"_103":99},{"_95":94,"_96":97,"_98":99,"_18":55,"_100":101,"_102":25},{"_13":84,"_32":85,"_43":42,"_86":44,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":45,"_93":94},"ctx",{"_77":-5,"_123":210,"_211":212,"_213":214,"_215":25,"_216":25,"_217":25,"_218":121,"_219":-5,"_220":-7,"_221":-7,"_222":61,"_223":25,"_224":25,"_225":-5,"_226":25,"_227":25,"_228":25,"_229":-5},"ENV",{"_181":182,"_183":184,"_185":85,"_186":187,"_188":189,"_190":191,"_192":193,"_194":195,"_196":197,"_198":199,"_200":201,"_202":203,"_204":205,"_206":207,"_208":209},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Diff · v15