@staticn0va/wigolo 0.1.0 → 0.1.1

package/dist/extraction/v1/recipe.js

@@ -0,0 +1,136 @@
+import { extractJsonLd } from "../jsonld.js";
+const MIN_CONTENT_THRESHOLD = 100;
+async function extractRecipe(html, _url) {
+  if (!html) return null;
+  let blocks;
+  try {
+    blocks = extractJsonLd(html);
+  } catch {
+    return null;
+  }
+  const recipe = blocks.find((block) => typeIncludes(block["@type"], "recipe"));
+  if (!recipe) return null;
+  const name = stringField(recipe["name"]);
+  if (!name) return null;
+  const lines = [`# ${name}`];
+  const description = stringField(recipe["description"]);
+  if (description) {
+    lines.push("", description);
+  }
+  const metaLines = [];
+  const totalTime = stringField(recipe["totalTime"]);
+  if (totalTime) metaLines.push(`**Total time:** ${totalTime}`);
+  const prepTime = stringField(recipe["prepTime"]);
+  if (prepTime) metaLines.push(`**Prep time:** ${prepTime}`);
+  const cookTime = stringField(recipe["cookTime"]);
+  if (cookTime) metaLines.push(`**Cook time:** ${cookTime}`);
+  const recipeYield = stringField(recipe["recipeYield"]);
+  if (recipeYield) metaLines.push(`**Yield:** ${recipeYield}`);
+  const cuisine = stringField(recipe["recipeCuisine"]);
+  if (cuisine) metaLines.push(`**Cuisine:** ${cuisine}`);
+  if (metaLines.length > 0) {
+    lines.push("", ...metaLines);
+  }
+  const ingredients = stringArray(recipe["recipeIngredient"]);
+  if (ingredients.length > 0) {
+    lines.push("", "## Ingredients");
+    for (const ingredient of ingredients) {
+      lines.push(`- ${ingredient}`);
+    }
+  }
+  const instructions = readInstructions(recipe["recipeInstructions"]);
+  if (instructions.length > 0) {
+    lines.push("", "## Instructions");
+    instructions.forEach((step, idx) => {
+      lines.push(`${idx + 1}. ${step}`);
+    });
+  }
+  const markdown = lines.join("\n").trim();
+  if (markdown.length < MIN_CONTENT_THRESHOLD) return null;
+  const author = readAuthor(recipe["author"]);
+  const datePublished = stringField(recipe["datePublished"]);
+  return {
+    title: name,
+    markdown,
+    metadata: {
+      ...description ? { description } : {},
+      ...author ? { author } : {},
+      ...datePublished ? { date: datePublished } : {}
+    },
+    links: [],
+    images: [],
+    extractor: "site-specific"
+  };
+}
+function typeIncludes(raw, want) {
+  const target = want.toLowerCase();
+  if (typeof raw === "string") return normalizeType(raw) === target;
+  if (Array.isArray(raw)) {
+    return raw.some((entry) => typeof entry === "string" && normalizeType(entry) === target);
+  }
+  return false;
+}
+function normalizeType(raw) {
+  const tail = raw.split(/[/#:]/).pop() ?? raw;
+  return tail.toLowerCase();
+}
+function stringField(value) {
+  if (typeof value !== "string") return void 0;
+  const trimmed = value.trim();
+  return trimmed.length > 0 ? trimmed : void 0;
+}
+function stringArray(value) {
+  if (!Array.isArray(value)) return [];
+  const out = [];
+  for (const entry of value) {
+    if (typeof entry === "string") {
+      const trimmed = entry.trim();
+      if (trimmed) out.push(trimmed);
+    }
+  }
+  return out;
+}
+function readInstructions(value) {
+  if (!Array.isArray(value)) {
+    if (typeof value === "string") {
+      const trimmed = value.trim();
+      return trimmed ? [trimmed] : [];
+    }
+    return [];
+  }
+  const out = [];
+  for (const entry of value) {
+    if (typeof entry === "string") {
+      const trimmed = entry.trim();
+      if (trimmed) out.push(trimmed);
+      continue;
+    }
+    if (entry && typeof entry === "object") {
+      const text = entry["text"];
+      if (typeof text === "string") {
+        const trimmed = text.trim();
+        if (trimmed) out.push(trimmed);
+      }
+    }
+  }
+  return out;
+}
+function readAuthor(value) {
+  if (typeof value === "string") return stringField(value);
+  if (Array.isArray(value)) {
+    for (const entry of value) {
+      const name = readAuthor(entry);
+      if (name) return name;
+    }
+    return void 0;
+  }
+  if (value && typeof value === "object") {
+    const name = value["name"];
+    if (typeof name === "string") return stringField(name);
+  }
+  return void 0;
+}
+export {
+  extractRecipe
+};
+//# sourceMappingURL=recipe.js.map

package/dist/extraction/v1/recipe.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../src/extraction/v1/recipe.ts"],"sourcesContent":["import { extractJsonLd } from '../jsonld.js';\nimport type { ExtractionResult } from '../../types.js';\n\nconst MIN_CONTENT_THRESHOLD = 100;\n\nexport async function extractRecipe(html: string, _url: string): Promise<ExtractionResult | null> {\n  if (!html) return null;\n\n  let blocks: Record<string, unknown>[];\n  try {\n    blocks = extractJsonLd(html);\n  } catch {\n    return null;\n  }\n\n  const recipe = blocks.find((block) => typeIncludes(block['@type'], 'recipe'));\n  if (!recipe) return null;\n\n  const name = stringField(recipe['name']);\n  if (!name) return null;\n\n  const lines: string[] = [`# ${name}`];\n\n  const description = stringField(recipe['description']);\n  if (description) {\n    lines.push('', description);\n  }\n\n  const metaLines: string[] = [];\n  const totalTime = stringField(recipe['totalTime']);\n  if (totalTime) metaLines.push(`**Total time:** ${totalTime}`);\n  const prepTime = stringField(recipe['prepTime']);\n  if (prepTime) metaLines.push(`**Prep time:** ${prepTime}`);\n  const cookTime = stringField(recipe['cookTime']);\n  if (cookTime) metaLines.push(`**Cook time:** ${cookTime}`);\n  const recipeYield = stringField(recipe['recipeYield']);\n  if (recipeYield) metaLines.push(`**Yield:** ${recipeYield}`);\n  const cuisine = stringField(recipe['recipeCuisine']);\n  if (cuisine) metaLines.push(`**Cuisine:** ${cuisine}`);\n  if (metaLines.length > 0) {\n    lines.push('', ...metaLines);\n  }\n\n  const ingredients = stringArray(recipe['recipeIngredient']);\n  if (ingredients.length > 0) {\n    lines.push('', '## Ingredients');\n    for (const ingredient of ingredients) {\n      lines.push(`- ${ingredient}`);\n    }\n  }\n\n  const instructions = readInstructions(recipe['recipeInstructions']);\n  if (instructions.length > 0) {\n    lines.push('', '## Instructions');\n    instructions.forEach((step, idx) => {\n      lines.push(`${idx + 1}. ${step}`);\n    });\n  }\n\n  const markdown = lines.join('\\n').trim();\n  if (markdown.length < MIN_CONTENT_THRESHOLD) return null;\n\n  const author = readAuthor(recipe['author']);\n  const datePublished = stringField(recipe['datePublished']);\n\n  return {\n    title: name,\n    markdown,\n    metadata: {\n      ...(description ? { description } : {}),\n      ...(author ? { author } : {}),\n      ...(datePublished ? { date: datePublished } : {}),\n    },\n    links: [],\n    images: [],\n    extractor: 'site-specific',\n  };\n}\n\nfunction typeIncludes(raw: unknown, want: string): boolean {\n  const target = want.toLowerCase();\n  if (typeof raw === 'string') return normalizeType(raw) === target;\n  if (Array.isArray(raw)) {\n    return raw.some((entry) => typeof entry === 'string' && normalizeType(entry) === target);\n  }\n  return false;\n}\n\nfunction normalizeType(raw: string): string {\n  const tail = raw.split(/[/#:]/).pop() ?? raw;\n  return tail.toLowerCase();\n}\n\nfunction stringField(value: unknown): string | undefined {\n  if (typeof value !== 'string') return undefined;\n  const trimmed = value.trim();\n  return trimmed.length > 0 ? trimmed : undefined;\n}\n\nfunction stringArray(value: unknown): string[] {\n  if (!Array.isArray(value)) return [];\n  const out: string[] = [];\n  for (const entry of value) {\n    if (typeof entry === 'string') {\n      const trimmed = entry.trim();\n      if (trimmed) out.push(trimmed);\n    }\n  }\n  return out;\n}\n\nfunction readInstructions(value: unknown): string[] {\n  if (!Array.isArray(value)) {\n    if (typeof value === 'string') {\n      const trimmed = value.trim();\n      return trimmed ? [trimmed] : [];\n    }\n    return [];\n  }\n  const out: string[] = [];\n  for (const entry of value) {\n    if (typeof entry === 'string') {\n      const trimmed = entry.trim();\n      if (trimmed) out.push(trimmed);\n      continue;\n    }\n    if (entry && typeof entry === 'object') {\n      const text = (entry as Record<string, unknown>)['text'];\n      if (typeof text === 'string') {\n        const trimmed = text.trim();\n        if (trimmed) out.push(trimmed);\n      }\n    }\n  }\n  return out;\n}\n\nfunction readAuthor(value: unknown): string | undefined {\n  if (typeof value === 'string') return stringField(value);\n  if (Array.isArray(value)) {\n    for (const entry of value) {\n      const name = readAuthor(entry);\n      if (name) return name;\n    }\n    return undefined;\n  }\n  if (value && typeof value === 'object') {\n    const name = (value as Record<string, unknown>)['name'];\n    if (typeof name === 'string') return stringField(name);\n  }\n  return undefined;\n}\n"],"mappings":"AAAA,SAAS,qBAAqB;AAG9B,MAAM,wBAAwB;AAE9B,eAAsB,cAAc,MAAc,MAAgD;AAChG,MAAI,CAAC,KAAM,QAAO;AAElB,MAAI;AACJ,MAAI;AACF,aAAS,cAAc,IAAI;AAAA,EAC7B,QAAQ;AACN,WAAO;AAAA,EACT;AAEA,QAAM,SAAS,OAAO,KAAK,CAAC,UAAU,aAAa,MAAM,OAAO,GAAG,QAAQ,CAAC;AAC5E,MAAI,CAAC,OAAQ,QAAO;AAEpB,QAAM,OAAO,YAAY,OAAO,MAAM,CAAC;AACvC,MAAI,CAAC,KAAM,QAAO;AAElB,QAAM,QAAkB,CAAC,KAAK,IAAI,EAAE;AAEpC,QAAM,cAAc,YAAY,OAAO,aAAa,CAAC;AACrD,MAAI,aAAa;AACf,UAAM,KAAK,IAAI,WAAW;AAAA,EAC5B;AAEA,QAAM,YAAsB,CAAC;AAC7B,QAAM,YAAY,YAAY,OAAO,WAAW,CAAC;AACjD,MAAI,UAAW,WAAU,KAAK,mBAAmB,SAAS,EAAE;AAC5D,QAAM,WAAW,YAAY,OAAO,UAAU,CAAC;AAC/C,MAAI,SAAU,WAAU,KAAK,kBAAkB,QAAQ,EAAE;AACzD,QAAM,WAAW,YAAY,OAAO,UAAU,CAAC;AAC/C,MAAI,SAAU,WAAU,KAAK,kBAAkB,QAAQ,EAAE;AACzD,QAAM,cAAc,YAAY,OAAO,aAAa,CAAC;AACrD,MAAI,YAAa,WAAU,KAAK,cAAc,WAAW,EAAE;AAC3D,QAAM,UAAU,YAAY,OAAO,eAAe,CAAC;AACnD,MAAI,QAAS,WAAU,KAAK,gBAAgB,OAAO,EAAE;AACrD,MAAI,UAAU,SAAS,GAAG;AACxB,UAAM,KAAK,IAAI,GAAG,SAAS;AAAA,EAC7B;AAEA,QAAM,cAAc,YAAY,OAAO,kBAAkB,CAAC;AAC1D,MAAI,YAAY,SAAS,GAAG;AAC1B,UAAM,KAAK,IAAI,gBAAgB;AAC/B,eAAW,cAAc,aAAa;AACpC,YAAM,KAAK,KAAK,UAAU,EAAE;AAAA,IAC9B;AAAA,EACF;AAEA,QAAM,eAAe,iBAAiB,OAAO,oBAAoB,CAAC;AAClE,MAAI,aAAa,SAAS,GAAG;AAC3B,UAAM,KAAK,IAAI,iBAAiB;AAChC,iBAAa,QAAQ,CAAC,MAAM,QAAQ;AAClC,YAAM,KAAK,GAAG,MAAM,CAAC,KAAK,IAAI,EAAE;AAAA,IAClC,CAAC;AAAA,EACH;AAEA,QAAM,WAAW,MAAM,KAAK,IAAI,EAAE,KAAK;AACvC,MAAI,SAAS,SAAS,sBAAuB,QAAO;AAEpD,QAAM,SAAS,WAAW,OAAO,QAAQ,CAAC;AAC1C,QAAM,gBAAgB,YAAY,OAAO,eAAe,CAAC;AAEzD,SAAO;AAAA,IACL,OAAO;AAAA,IACP;AAAA,IACA,UAAU;AAAA,MACR,GAAI,cAAc,EAAE,YAAY,IAAI,CAAC;AAAA,MACrC,GAAI,SAAS,EAAE,OAAO,IAAI,CAAC;AAAA,MAC3B,GAAI,gBAAgB,EAAE,MAAM,cAAc,IAAI,CAAC;AAAA,IACjD;AAAA,IACA,OAAO,CAAC;AAAA,IACR,QAAQ,CAAC;AAAA,IACT,WAAW;AAAA,EACb;AACF;AAEA,SAAS,aAAa,KAAc,MAAuB;AACzD,QAAM,SAAS,KAAK,YAAY;AAChC,MAAI,OAAO,QAAQ,SAAU,QAAO,cAAc,GAAG,MAAM;AAC3D,MAAI,MAAM,QAAQ,GAAG,GAAG;AACtB,WAAO,IAAI,KAAK,CAAC,UAAU,OAAO,UAAU,YAAY,cAAc,KAAK,MAAM,MAAM;AAAA,EACzF;AACA,SAAO;AACT;AAEA,SAAS,cAAc,KAAqB;AAC1C,QAAM,OAAO,IAAI,MAAM,OAAO,EAAE,IAAI,KAAK;AACzC,SAAO,KAAK,YAAY;AAC1B;AAEA,SAAS,YAAY,OAAoC;AACvD,MAAI,OAAO,UAAU,SAAU,QAAO;AACtC,QAAM,UAAU,MAAM,KAAK;AAC3B,SAAO,QAAQ,SAAS,IAAI,UAAU;AACxC;AAEA,SAAS,YAAY,OAA0B;AAC7C,MAAI,CAAC,MAAM,QAAQ,KAAK,EAAG,QAAO,CAAC;AACnC,QAAM,MAAgB,CAAC;AACvB,aAAW,SAAS,OAAO;AACzB,QAAI,OAAO,UAAU,UAAU;AAC7B,YAAM,UAAU,MAAM,KAAK;AAC3B,UAAI,QAAS,KAAI,KAAK,OAAO;AAAA,IAC/B;AAAA,EACF;AACA,SAAO;AACT;AAEA,SAAS,iBAAiB,OAA0B;AAClD,MAAI,CAAC,MAAM,QAAQ,KAAK,GAAG;AACzB,QAAI,OAAO,UAAU,UAAU;AAC7B,YAAM,UAAU,MAAM,KAAK;AAC3B,aAAO,UAAU,CAAC,OAAO,IAAI,CAAC;AAAA,IAChC;AACA,WAAO,CAAC;AAAA,EACV;AACA,QAAM,MAAgB,CAAC;AACvB,aAAW,SAAS,OAAO;AACzB,QAAI,OAAO,UAAU,UAAU;AAC7B,YAAM,UAAU,MAAM,KAAK;AAC3B,UAAI,QAAS,KAAI,KAAK,OAAO;AAC7B;AAAA,IACF;AACA,QAAI,SAAS,OAAO,UAAU,UAAU;AACtC,YAAM,OAAQ,MAAkC,MAAM;AACtD,UAAI,OAAO,SAAS,UAAU;AAC5B,cAAM,UAAU,KAAK,KAAK;AAC1B,YAAI,QAAS,KAAI,KAAK,OAAO;AAAA,MAC/B;AAAA,IACF;AAAA,EACF;AACA,SAAO;AACT;AAEA,SAAS,WAAW,OAAoC;AACtD,MAAI,OAAO,UAAU,SAAU,QAAO,YAAY,KAAK;AACvD,MAAI,MAAM,QAAQ,KAAK,GAAG;AACxB,eAAW,SAAS,OAAO;AACzB,YAAM,OAAO,WAAW,KAAK;AAC7B,UAAI,KAAM,QAAO;AAAA,IACnB;AACA,WAAO;AAAA,EACT;AACA,MAAI,SAAS,OAAO,UAAU,UAAU;AACtC,UAAM,OAAQ,MAAkC,MAAM;AACtD,QAAI,OAAO,SAAS,SAAU,QAAO,YAAY,IAAI;AAAA,EACvD;AACA,SAAO;AACT;","names":[]}

package/dist/extraction/v1/routed.d.ts

@@ -0,0 +1,17 @@
+import type { ExtractionResult } from '../../types.js';
+export interface RoutedExtractInput {
+    html: string;
+    url: string;
+    cleanedHtml?: string;
+    contentType?: string;
+}
+/**
+ * V1 routed extractor — picks a category-specific extractor based on the
+ * classifier output, with defuddle → readability → turndown fallbacks. Site
+ * extractors (github/stackoverflow/mdn/docs-generic + plugins) run first and
+ * short-circuit on match, matching legacy behavior.
+ *
+ * PDF handling lives in V1Extractor — this router assumes HTML.
+ */
+export declare function routedExtract(input: RoutedExtractInput): Promise<ExtractionResult>;
+//# sourceMappingURL=routed.d.ts.map

package/dist/extraction/v1/routed.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"routed.d.ts","sourceRoot":"","sources":["../../../src/extraction/v1/routed.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,gBAAgB,EAAE,MAAM,gBAAgB,CAAC;AAcvD,MAAM,WAAW,kBAAkB;IACjC,IAAI,EAAE,MAAM,CAAC;IACb,GAAG,EAAE,MAAM,CAAC;IACZ,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,WAAW,CAAC,EAAE,MAAM,CAAC;CACtB;AAED;;;;;;;GAOG;AACH,wBAAsB,aAAa,CAAC,KAAK,EAAE,kBAAkB,GAAG,OAAO,CAAC,gBAAgB,CAAC,CA2BxF"}

package/dist/extraction/v1/routed.js

@@ -0,0 +1,68 @@
+import { parseHTML } from "linkedom";
+import { defuddleExtract } from "../defuddle.js";
+import { readabilityExtract } from "../readability.js";
+import { htmlToMarkdown } from "../markdown.js";
+import { stripBoilerplateDom } from "../boilerplate.js";
+import { createLogger } from "../../logger.js";
+import { classifyContent } from "./classifier.js";
+import { extractRecipe } from "./recipe.js";
+import { extractProduct } from "./product.js";
+import { extractNews } from "./news.js";
+import { getSiteExtractors } from "./site-extractors.js";
+const log = createLogger("extract");
+async function routedExtract(input) {
+  const { html, url } = input;
+  const cleanedHtml = input.cleanedHtml ?? cleanHtml(html, url);
+  const siteHit = trySiteExtractors(cleanedHtml, url, html);
+  if (siteHit) return siteHit;
+  const type = classifyContent(url, html);
+  log.debug("classified content", { url, type });
+  switch (type) {
+    case "recipe":
+      return await extractRecipe(cleanedHtml, url) ?? await fallbackChain(cleanedHtml, url);
+    case "product":
+      return await extractProduct(cleanedHtml, url) ?? await fallbackChain(cleanedHtml, url);
+    case "news":
+      return await extractNews(cleanedHtml, url) ?? await fallbackChain(cleanedHtml, url);
+    case "code":
+    case "docs":
+    case "generic":
+    default:
+      return fallbackChain(cleanedHtml, url, type);
+  }
+}
+function cleanHtml(html, url) {
+  try {
+    const { document } = parseHTML(html);
+    stripBoilerplateDom(document);
+    return document.toString();
+  } catch (err) {
+    log.warn("boilerplate DOM pre-pass failed", { url, error: String(err) });
+    return html;
+  }
+}
+function trySiteExtractors(cleanedHtml, url, originalHtml) {
+  const extractors = getSiteExtractors();
+  const match = extractors.find((e) => e.canHandle(url, originalHtml));
+  if (!match) return null;
+  const out = match.extract(cleanedHtml, url);
+  return out ?? null;
+}
+async function fallbackChain(cleanedHtml, url, _type) {
+  const fromDefuddle = await defuddleExtract(cleanedHtml, url);
+  if (fromDefuddle) return fromDefuddle;
+  const fromReadability = readabilityExtract(cleanedHtml, url);
+  if (fromReadability) return fromReadability;
+  return {
+    title: "",
+    markdown: htmlToMarkdown(cleanedHtml),
+    metadata: {},
+    links: [],
+    images: [],
+    extractor: "turndown"
+  };
+}
+export {
+  routedExtract
+};
+//# sourceMappingURL=routed.js.map

package/dist/extraction/v1/routed.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../src/extraction/v1/routed.ts"],"sourcesContent":["import { parseHTML } from 'linkedom';\nimport type { ExtractionResult } from '../../types.js';\nimport { defuddleExtract } from '../defuddle.js';\nimport { readabilityExtract } from '../readability.js';\nimport { htmlToMarkdown } from '../markdown.js';\nimport { stripBoilerplateDom } from '../boilerplate.js';\nimport { createLogger } from '../../logger.js';\nimport { classifyContent, type ContentType } from './classifier.js';\nimport { extractRecipe } from './recipe.js';\nimport { extractProduct } from './product.js';\nimport { extractNews } from './news.js';\nimport { getSiteExtractors } from './site-extractors.js';\n\nconst log = createLogger('extract');\n\nexport interface RoutedExtractInput {\n  html: string;\n  url: string;\n  cleanedHtml?: string;\n  contentType?: string;\n}\n\n/**\n * V1 routed extractor — picks a category-specific extractor based on the\n * classifier output, with defuddle → readability → turndown fallbacks. Site\n * extractors (github/stackoverflow/mdn/docs-generic + plugins) run first and\n * short-circuit on match, matching legacy behavior.\n *\n * PDF handling lives in V1Extractor — this router assumes HTML.\n */\nexport async function routedExtract(input: RoutedExtractInput): Promise<ExtractionResult> {\n  const { html, url } = input;\n  const cleanedHtml = input.cleanedHtml ?? cleanHtml(html, url);\n\n  const siteHit = trySiteExtractors(cleanedHtml, url, html);\n  if (siteHit) return siteHit;\n\n  const type = classifyContent(url, html);\n  log.debug('classified content', { url, type });\n\n  switch (type) {\n    case 'recipe':\n      return (\n        (await extractRecipe(cleanedHtml, url)) ?? (await fallbackChain(cleanedHtml, url))\n      );\n    case 'product':\n      return (\n        (await extractProduct(cleanedHtml, url)) ?? (await fallbackChain(cleanedHtml, url))\n      );\n    case 'news':\n      return (await extractNews(cleanedHtml, url)) ?? (await fallbackChain(cleanedHtml, url));\n    case 'code':\n    case 'docs':\n    case 'generic':\n    default:\n      return fallbackChain(cleanedHtml, url, type);\n  }\n}\n\nfunction cleanHtml(html: string, url: string): string {\n  try {\n    const { document } = parseHTML(html);\n    stripBoilerplateDom(document);\n    return document.toString();\n  } catch (err) {\n    log.warn('boilerplate DOM pre-pass failed', { url, error: String(err) });\n    return html;\n  }\n}\n\nfunction trySiteExtractors(\n  cleanedHtml: string,\n  url: string,\n  originalHtml: string,\n): ExtractionResult | null {\n  const extractors = getSiteExtractors();\n  const match = extractors.find((e) => e.canHandle(url, originalHtml));\n  if (!match) return null;\n  const out = match.extract(cleanedHtml, url);\n  return out ?? null;\n}\n\nasync function fallbackChain(\n  cleanedHtml: string,\n  url: string,\n  _type?: ContentType,\n): Promise<ExtractionResult> {\n  const fromDefuddle = await defuddleExtract(cleanedHtml, url);\n  if (fromDefuddle) return fromDefuddle;\n\n  const fromReadability = readabilityExtract(cleanedHtml, url);\n  if (fromReadability) return fromReadability;\n\n  return {\n    title: '',\n    markdown: htmlToMarkdown(cleanedHtml),\n    metadata: {},\n    links: [],\n    images: [],\n    extractor: 'turndown',\n  };\n}\n"],"mappings":"AAAA,SAAS,iBAAiB;AAE1B,SAAS,uBAAuB;AAChC,SAAS,0BAA0B;AACnC,SAAS,sBAAsB;AAC/B,SAAS,2BAA2B;AACpC,SAAS,oBAAoB;AAC7B,SAAS,uBAAyC;AAClD,SAAS,qBAAqB;AAC9B,SAAS,sBAAsB;AAC/B,SAAS,mBAAmB;AAC5B,SAAS,yBAAyB;AAElC,MAAM,MAAM,aAAa,SAAS;AAiBlC,eAAsB,cAAc,OAAsD;AACxF,QAAM,EAAE,MAAM,IAAI,IAAI;AACtB,QAAM,cAAc,MAAM,eAAe,UAAU,MAAM,GAAG;AAE5D,QAAM,UAAU,kBAAkB,aAAa,KAAK,IAAI;AACxD,MAAI,QAAS,QAAO;AAEpB,QAAM,OAAO,gBAAgB,KAAK,IAAI;AACtC,MAAI,MAAM,sBAAsB,EAAE,KAAK,KAAK,CAAC;AAE7C,UAAQ,MAAM;AAAA,IACZ,KAAK;AACH,aACG,MAAM,cAAc,aAAa,GAAG,KAAO,MAAM,cAAc,aAAa,GAAG;AAAA,IAEpF,KAAK;AACH,aACG,MAAM,eAAe,aAAa,GAAG,KAAO,MAAM,cAAc,aAAa,GAAG;AAAA,IAErF,KAAK;AACH,aAAQ,MAAM,YAAY,aAAa,GAAG,KAAO,MAAM,cAAc,aAAa,GAAG;AAAA,IACvF,KAAK;AAAA,IACL,KAAK;AAAA,IACL,KAAK;AAAA,IACL;AACE,aAAO,cAAc,aAAa,KAAK,IAAI;AAAA,EAC/C;AACF;AAEA,SAAS,UAAU,MAAc,KAAqB;AACpD,MAAI;AACF,UAAM,EAAE,SAAS,IAAI,UAAU,IAAI;AACnC,wBAAoB,QAAQ;AAC5B,WAAO,SAAS,SAAS;AAAA,EAC3B,SAAS,KAAK;AACZ,QAAI,KAAK,mCAAmC,EAAE,KAAK,OAAO,OAAO,GAAG,EAAE,CAAC;AACvE,WAAO;AAAA,EACT;AACF;AAEA,SAAS,kBACP,aACA,KACA,cACyB;AACzB,QAAM,aAAa,kBAAkB;AACrC,QAAM,QAAQ,WAAW,KAAK,CAAC,MAAM,EAAE,UAAU,KAAK,YAAY,CAAC;AACnE,MAAI,CAAC,MAAO,QAAO;AACnB,QAAM,MAAM,MAAM,QAAQ,aAAa,GAAG;AAC1C,SAAO,OAAO;AAChB;AAEA,eAAe,cACb,aACA,KACA,OAC2B;AAC3B,QAAM,eAAe,MAAM,gBAAgB,aAAa,GAAG;AAC3D,MAAI,aAAc,QAAO;AAEzB,QAAM,kBAAkB,mBAAmB,aAAa,GAAG;AAC3D,MAAI,gBAAiB,QAAO;AAE5B,SAAO;AAAA,IACL,OAAO;AAAA,IACP,UAAU,eAAe,WAAW;AAAA,IACpC,UAAU,CAAC;AAAA,IACX,OAAO,CAAC;AAAA,IACR,QAAQ,CAAC;AAAA,IACT,WAAW;AAAA,EACb;AACF;","names":[]}

package/dist/extraction/v1/schemas/Article.d.ts

@@ -0,0 +1,11 @@
+export interface ArticleData {
+    title: string;
+    body: string;
+    url: string;
+    author?: string;
+    date?: string;
+    description?: string;
+    language?: string;
+}
+export declare function extractArticle(html: string, url: string): Promise<ArticleData | null>;
+//# sourceMappingURL=Article.d.ts.map

package/dist/extraction/v1/schemas/Article.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"Article.d.ts","sourceRoot":"","sources":["../../../../src/extraction/v1/schemas/Article.ts"],"names":[],"mappings":"AAEA,MAAM,WAAW,WAAW;IAC1B,KAAK,EAAE,MAAM,CAAC;IACd,IAAI,EAAE,MAAM,CAAC;IACb,GAAG,EAAE,MAAM,CAAC;IACZ,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,wBAAsB,cAAc,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,OAAO,CAAC,WAAW,GAAG,IAAI,CAAC,CAoB3F"}

package/dist/extraction/v1/schemas/Article.js

@@ -0,0 +1,23 @@
+import { extractNews } from "../news.js";
+async function extractArticle(html, url) {
+  const result = await extractNews(html, url);
+  if (!result) return null;
+  const title = (result.title ?? "").trim();
+  const body = (result.markdown ?? "").trim();
+  if (!title && !body) return null;
+  const meta = result.metadata ?? {};
+  const data = {
+    title,
+    body,
+    url
+  };
+  if (meta.author) data.author = meta.author;
+  if (meta.date) data.date = meta.date;
+  if (meta.description) data.description = meta.description;
+  if (meta.language) data.language = meta.language;
+  return data;
+}
+export {
+  extractArticle
+};
+//# sourceMappingURL=Article.js.map

package/dist/extraction/v1/schemas/Article.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/extraction/v1/schemas/Article.ts"],"sourcesContent":["import { extractNews } from '../news.js';\n\nexport interface ArticleData {\n  title: string;\n  body: string;\n  url: string;\n  author?: string;\n  date?: string;\n  description?: string;\n  language?: string;\n}\n\nexport async function extractArticle(html: string, url: string): Promise<ArticleData | null> {\n  const result = await extractNews(html, url);\n  if (!result) return null;\n\n  const title = (result.title ?? '').trim();\n  const body = (result.markdown ?? '').trim();\n  if (!title && !body) return null;\n\n  const meta = result.metadata ?? {};\n  const data: ArticleData = {\n    title,\n    body,\n    url,\n  };\n  if (meta.author) data.author = meta.author;\n  if (meta.date) data.date = meta.date;\n  if (meta.description) data.description = meta.description;\n  if (meta.language) data.language = meta.language;\n\n  return data;\n}\n"],"mappings":"AAAA,SAAS,mBAAmB;AAY5B,eAAsB,eAAe,MAAc,KAA0C;AAC3F,QAAM,SAAS,MAAM,YAAY,MAAM,GAAG;AAC1C,MAAI,CAAC,OAAQ,QAAO;AAEpB,QAAM,SAAS,OAAO,SAAS,IAAI,KAAK;AACxC,QAAM,QAAQ,OAAO,YAAY,IAAI,KAAK;AAC1C,MAAI,CAAC,SAAS,CAAC,KAAM,QAAO;AAE5B,QAAM,OAAO,OAAO,YAAY,CAAC;AACjC,QAAM,OAAoB;AAAA,IACxB;AAAA,IACA;AAAA,IACA;AAAA,EACF;AACA,MAAI,KAAK,OAAQ,MAAK,SAAS,KAAK;AACpC,MAAI,KAAK,KAAM,MAAK,OAAO,KAAK;AAChC,MAAI,KAAK,YAAa,MAAK,cAAc,KAAK;AAC9C,MAAI,KAAK,SAAU,MAAK,WAAW,KAAK;AAExC,SAAO;AACT;","names":[]}

package/dist/extraction/v1/schemas/CodeSnippet.d.ts

@@ -0,0 +1,9 @@
+export interface CodeSnippetData {
+    language?: string;
+    code: string;
+    filename?: string;
+    description?: string;
+    url: string;
+}
+export declare function extractCodeSnippet(html: string, url: string): Promise<CodeSnippetData | null>;
+//# sourceMappingURL=CodeSnippet.d.ts.map

package/dist/extraction/v1/schemas/CodeSnippet.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"CodeSnippet.d.ts","sourceRoot":"","sources":["../../../../src/extraction/v1/schemas/CodeSnippet.ts"],"names":[],"mappings":"AAEA,MAAM,WAAW,eAAe;IAC9B,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,IAAI,EAAE,MAAM,CAAC;IACb,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,GAAG,EAAE,MAAM,CAAC;CACb;AAID,wBAAsB,kBAAkB,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,OAAO,CAAC,eAAe,GAAG,IAAI,CAAC,CAwCnG"}

package/dist/extraction/v1/schemas/CodeSnippet.js

@@ -0,0 +1,90 @@
+import { parseHTML } from "linkedom";
+const MIN_CODE_LENGTH = 30;
+async function extractCodeSnippet(html, url) {
+  if (!html) return null;
+  let document;
+  try {
+    ({ document } = parseHTML(html));
+  } catch {
+    return null;
+  }
+  const blocks = Array.from(document.querySelectorAll("pre > code, pre code"));
+  if (blocks.length === 0) return null;
+  let largest = null;
+  let largestLen = 0;
+  for (const block of blocks) {
+    const text = block.textContent ?? "";
+    if (text.length > largestLen) {
+      largestLen = text.length;
+      largest = block;
+    }
+  }
+  if (!largest || largestLen < MIN_CODE_LENGTH) return null;
+  const code = (largest.textContent ?? "").trim();
+  if (!code) return null;
+  const data = { code, url };
+  const language = detectLanguage(largest);
+  if (language) data.language = language;
+  const pre = closestPre(largest);
+  const filename = detectFilename(pre);
+  if (filename) data.filename = filename;
+  const description = detectDescription(pre);
+  if (description) data.description = description;
+  return data;
+}
+function detectLanguage(codeEl) {
+  const fromCode = languageFromClass(codeEl.getAttribute("class"));
+  if (fromCode) return fromCode;
+  const pre = closestPre(codeEl);
+  if (pre) {
+    const fromPre = languageFromClass(pre.getAttribute("class"));
+    if (fromPre) return fromPre;
+  }
+  return void 0;
+}
+function languageFromClass(cls) {
+  if (!cls) return void 0;
+  for (const token of cls.split(/\s+/)) {
+    const m = /^(?:language|lang|hljs|highlight)-(.+)$/.exec(token);
+    if (m) return m[1];
+  }
+  return void 0;
+}
+function closestPre(el) {
+  let current = el;
+  while (current) {
+    if (current.tagName === "PRE") return current;
+    current = current.parentElement;
+  }
+  return null;
+}
+function detectFilename(pre) {
+  if (!pre) return void 0;
+  const figcaption = pre.parentElement?.querySelector("figcaption") ?? pre.previousElementSibling;
+  if (figcaption && figcaption.tagName === "FIGCAPTION") {
+    const text = (figcaption.textContent ?? "").trim();
+    if (text) return text;
+  }
+  const header = pre.parentElement?.querySelector("header, .filename, div.filename");
+  if (header) {
+    const text = (header.textContent ?? "").trim();
+    if (text) return text;
+  }
+  return void 0;
+}
+function detectDescription(pre) {
+  if (!pre) return void 0;
+  let sib = pre.previousElementSibling;
+  while (sib && (sib.tagName === "FIGCAPTION" || sib.tagName === "HEADER")) {
+    sib = sib.previousElementSibling;
+  }
+  if (sib && sib.tagName === "P") {
+    const text = (sib.textContent ?? "").trim();
+    if (text) return text.slice(0, 300);
+  }
+  return void 0;
+}
+export {
+  extractCodeSnippet
+};
+//# sourceMappingURL=CodeSnippet.js.map

package/dist/extraction/v1/schemas/CodeSnippet.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/extraction/v1/schemas/CodeSnippet.ts"],"sourcesContent":["import { parseHTML } from 'linkedom';\n\nexport interface CodeSnippetData {\n  language?: string;\n  code: string;\n  filename?: string;\n  description?: string;\n  url: string;\n}\n\nconst MIN_CODE_LENGTH = 30;\n\nexport async function extractCodeSnippet(html: string, url: string): Promise<CodeSnippetData | null> {\n  if (!html) return null;\n\n  let document: Document;\n  try {\n    ({ document } = parseHTML(html));\n  } catch {\n    return null;\n  }\n\n  const blocks = Array.from(document.querySelectorAll('pre > code, pre code'));\n  if (blocks.length === 0) return null;\n\n  let largest: Element | null = null;\n  let largestLen = 0;\n  for (const block of blocks) {\n    const text = block.textContent ?? '';\n    if (text.length > largestLen) {\n      largestLen = text.length;\n      largest = block;\n    }\n  }\n  if (!largest || largestLen < MIN_CODE_LENGTH) return null;\n\n  const code = (largest.textContent ?? '').trim();\n  if (!code) return null;\n\n  const data: CodeSnippetData = { code, url };\n\n  const language = detectLanguage(largest);\n  if (language) data.language = language;\n\n  const pre = closestPre(largest);\n  const filename = detectFilename(pre);\n  if (filename) data.filename = filename;\n\n  const description = detectDescription(pre);\n  if (description) data.description = description;\n\n  return data;\n}\n\nfunction detectLanguage(codeEl: Element): string | undefined {\n  const fromCode = languageFromClass(codeEl.getAttribute('class'));\n  if (fromCode) return fromCode;\n  const pre = closestPre(codeEl);\n  if (pre) {\n    const fromPre = languageFromClass(pre.getAttribute('class'));\n    if (fromPre) return fromPre;\n  }\n  return undefined;\n}\n\nfunction languageFromClass(cls: string | null | undefined): string | undefined {\n  if (!cls) return undefined;\n  for (const token of cls.split(/\\s+/)) {\n    const m = /^(?:language|lang|hljs|highlight)-(.+)$/.exec(token);\n    if (m) return m[1];\n  }\n  return undefined;\n}\n\nfunction closestPre(el: Element): Element | null {\n  let current: Element | null = el;\n  while (current) {\n    if (current.tagName === 'PRE') return current;\n    current = current.parentElement;\n  }\n  return null;\n}\n\nfunction detectFilename(pre: Element | null): string | undefined {\n  if (!pre) return undefined;\n  const figcaption =\n    pre.parentElement?.querySelector('figcaption') ?? pre.previousElementSibling;\n  if (figcaption && figcaption.tagName === 'FIGCAPTION') {\n    const text = (figcaption.textContent ?? '').trim();\n    if (text) return text;\n  }\n  const header = pre.parentElement?.querySelector('header, .filename, div.filename');\n  if (header) {\n    const text = (header.textContent ?? '').trim();\n    if (text) return text;\n  }\n  return undefined;\n}\n\nfunction detectDescription(pre: Element | null): string | undefined {\n  if (!pre) return undefined;\n  let sib = pre.previousElementSibling;\n  // skip over figcaption/header used for filename\n  while (sib && (sib.tagName === 'FIGCAPTION' || sib.tagName === 'HEADER')) {\n    sib = sib.previousElementSibling;\n  }\n  if (sib && sib.tagName === 'P') {\n    const text = (sib.textContent ?? '').trim();\n    if (text) return text.slice(0, 300);\n  }\n  return undefined;\n}\n"],"mappings":"AAAA,SAAS,iBAAiB;AAU1B,MAAM,kBAAkB;AAExB,eAAsB,mBAAmB,MAAc,KAA8C;AACnG,MAAI,CAAC,KAAM,QAAO;AAElB,MAAI;AACJ,MAAI;AACF,KAAC,EAAE,SAAS,IAAI,UAAU,IAAI;AAAA,EAChC,QAAQ;AACN,WAAO;AAAA,EACT;AAEA,QAAM,SAAS,MAAM,KAAK,SAAS,iBAAiB,sBAAsB,CAAC;AAC3E,MAAI,OAAO,WAAW,EAAG,QAAO;AAEhC,MAAI,UAA0B;AAC9B,MAAI,aAAa;AACjB,aAAW,SAAS,QAAQ;AAC1B,UAAM,OAAO,MAAM,eAAe;AAClC,QAAI,KAAK,SAAS,YAAY;AAC5B,mBAAa,KAAK;AAClB,gBAAU;AAAA,IACZ;AAAA,EACF;AACA,MAAI,CAAC,WAAW,aAAa,gBAAiB,QAAO;AAErD,QAAM,QAAQ,QAAQ,eAAe,IAAI,KAAK;AAC9C,MAAI,CAAC,KAAM,QAAO;AAElB,QAAM,OAAwB,EAAE,MAAM,IAAI;AAE1C,QAAM,WAAW,eAAe,OAAO;AACvC,MAAI,SAAU,MAAK,WAAW;AAE9B,QAAM,MAAM,WAAW,OAAO;AAC9B,QAAM,WAAW,eAAe,GAAG;AACnC,MAAI,SAAU,MAAK,WAAW;AAE9B,QAAM,cAAc,kBAAkB,GAAG;AACzC,MAAI,YAAa,MAAK,cAAc;AAEpC,SAAO;AACT;AAEA,SAAS,eAAe,QAAqC;AAC3D,QAAM,WAAW,kBAAkB,OAAO,aAAa,OAAO,CAAC;AAC/D,MAAI,SAAU,QAAO;AACrB,QAAM,MAAM,WAAW,MAAM;AAC7B,MAAI,KAAK;AACP,UAAM,UAAU,kBAAkB,IAAI,aAAa,OAAO,CAAC;AAC3D,QAAI,QAAS,QAAO;AAAA,EACtB;AACA,SAAO;AACT;AAEA,SAAS,kBAAkB,KAAoD;AAC7E,MAAI,CAAC,IAAK,QAAO;AACjB,aAAW,SAAS,IAAI,MAAM,KAAK,GAAG;AACpC,UAAM,IAAI,0CAA0C,KAAK,KAAK;AAC9D,QAAI,EAAG,QAAO,EAAE,CAAC;AAAA,EACnB;AACA,SAAO;AACT;AAEA,SAAS,WAAW,IAA6B;AAC/C,MAAI,UAA0B;AAC9B,SAAO,SAAS;AACd,QAAI,QAAQ,YAAY,MAAO,QAAO;AACtC,cAAU,QAAQ;AAAA,EACpB;AACA,SAAO;AACT;AAEA,SAAS,eAAe,KAAyC;AAC/D,MAAI,CAAC,IAAK,QAAO;AACjB,QAAM,aACJ,IAAI,eAAe,cAAc,YAAY,KAAK,IAAI;AACxD,MAAI,cAAc,WAAW,YAAY,cAAc;AACrD,UAAM,QAAQ,WAAW,eAAe,IAAI,KAAK;AACjD,QAAI,KAAM,QAAO;AAAA,EACnB;AACA,QAAM,SAAS,IAAI,eAAe,cAAc,iCAAiC;AACjF,MAAI,QAAQ;AACV,UAAM,QAAQ,OAAO,eAAe,IAAI,KAAK;AAC7C,QAAI,KAAM,QAAO;AAAA,EACnB;AACA,SAAO;AACT;AAEA,SAAS,kBAAkB,KAAyC;AAClE,MAAI,CAAC,IAAK,QAAO;AACjB,MAAI,MAAM,IAAI;AAEd,SAAO,QAAQ,IAAI,YAAY,gBAAgB,IAAI,YAAY,WAAW;AACxE,UAAM,IAAI;AAAA,EACZ;AACA,MAAI,OAAO,IAAI,YAAY,KAAK;AAC9B,UAAM,QAAQ,IAAI,eAAe,IAAI,KAAK;AAC1C,QAAI,KAAM,QAAO,KAAK,MAAM,GAAG,GAAG;AAAA,EACpC;AACA,SAAO;AACT;","names":[]}

package/dist/extraction/v1/schemas/EventListing.d.ts

@@ -0,0 +1,10 @@
+export interface EventListingData {
+    name: string;
+    startDate: string;
+    endDate?: string;
+    location?: string;
+    description?: string;
+    url: string;
+}
+export declare function extractEventListing(html: string, url: string): Promise<EventListingData | null>;
+//# sourceMappingURL=EventListing.d.ts.map

package/dist/extraction/v1/schemas/EventListing.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"EventListing.d.ts","sourceRoot":"","sources":["../../../../src/extraction/v1/schemas/EventListing.ts"],"names":[],"mappings":"AAGA,MAAM,WAAW,gBAAgB;IAC/B,IAAI,EAAE,MAAM,CAAC;IACb,SAAS,EAAE,MAAM,CAAC;IAClB,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,GAAG,EAAE,MAAM,CAAC;CACb;AAED,wBAAsB,mBAAmB,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,OAAO,CAAC,gBAAgB,GAAG,IAAI,CAAC,CAOrG"}

package/dist/extraction/v1/schemas/EventListing.js

@@ -0,0 +1,122 @@
+import { parseHTML } from "linkedom";
+import { extractJsonLd } from "../../jsonld.js";
+async function extractEventListing(html, url) {
+  if (!html) return null;
+  const fromJsonLd = tryJsonLd(html, url);
+  if (fromJsonLd) return fromJsonLd;
+  return tryMetaFallback(html, url);
+}
+function tryJsonLd(html, url) {
+  let blocks;
+  try {
+    blocks = extractJsonLd(html);
+  } catch {
+    return null;
+  }
+  const ev = blocks.find((block) => typeIncludes(block["@type"], "event"));
+  if (!ev) return null;
+  const name = stringField(ev["name"]);
+  const startDate = stringField(ev["startDate"]);
+  if (!name && !startDate) return null;
+  const data = {
+    name: name ?? "",
+    startDate: startDate ?? "",
+    url
+  };
+  const endDate = stringField(ev["endDate"]);
+  if (endDate) data.endDate = endDate;
+  const description = stringField(ev["description"]);
+  if (description) data.description = description;
+  const location = readLocation(ev["location"]);
+  if (location) data.location = location;
+  return data;
+}
+function tryMetaFallback(html, url) {
+  let document;
+  try {
+    ({ document } = parseHTML(html));
+  } catch {
+    return null;
+  }
+  const startMeta = metaContent(document, 'meta[property="event:start_time"]');
+  const endMeta = metaContent(document, 'meta[property="event:end_time"]');
+  const locationMeta = metaContent(document, 'meta[property="event:location"]');
+  const description = metaContent(document, 'meta[property="og:description"]');
+  let start = startMeta;
+  if (!start) {
+    const timeEl = document.querySelector('time[itemprop="startDate"]');
+    const dt = timeEl?.getAttribute("datetime")?.trim();
+    if (dt) start = dt;
+  }
+  let name;
+  const h1 = document.querySelector("h1");
+  const h1Text = h1?.textContent?.trim();
+  if (h1Text) name = h1Text;
+  if (!name) {
+    name = metaContent(document, 'meta[property="og:title"]');
+  }
+  if (!name && !start) return null;
+  const data = {
+    name: name ?? "",
+    startDate: start ?? "",
+    url
+  };
+  if (endMeta) data.endDate = endMeta;
+  if (locationMeta) data.location = locationMeta;
+  if (description) data.description = description;
+  return data;
+}
+function metaContent(document, selector) {
+  const el = document.querySelector(selector);
+  const content = el?.getAttribute("content")?.trim();
+  return content && content.length > 0 ? content : void 0;
+}
+function typeIncludes(raw, want) {
+  const target = want.toLowerCase();
+  if (typeof raw === "string") return normalizeType(raw) === target;
+  if (Array.isArray(raw)) {
+    return raw.some((entry) => typeof entry === "string" && normalizeType(entry) === target);
+  }
+  return false;
+}
+function normalizeType(raw) {
+  const tail = raw.split(/[/#:]/).pop() ?? raw;
+  return tail.toLowerCase();
+}
+function stringField(value) {
+  if (typeof value !== "string") return void 0;
+  const trimmed = value.trim();
+  return trimmed.length > 0 ? trimmed : void 0;
+}
+function readLocation(value) {
+  if (typeof value === "string") return stringField(value);
+  if (Array.isArray(value)) {
+    for (const entry of value) {
+      const loc = readLocation(entry);
+      if (loc) return loc;
+    }
+    return void 0;
+  }
+  if (value && typeof value === "object") {
+    const obj = value;
+    const name = stringField(obj["name"]);
+    if (name) return name;
+    const address = obj["address"];
+    if (typeof address === "string") return stringField(address);
+    if (address && typeof address === "object") {
+      const aobj = address;
+      const parts = [
+        stringField(aobj["streetAddress"]),
+        stringField(aobj["addressLocality"]),
+        stringField(aobj["addressRegion"]),
+        stringField(aobj["addressCountry"])
+      ].filter((s) => Boolean(s));
+      if (parts.length > 0) return parts.join(", ");
+    }
+  }
+  return void 0;
+}
+export {
+  extractEventListing
+};
+//# sourceMappingURL=EventListing.js.map

package/dist/extraction/v1/schemas/EventListing.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/extraction/v1/schemas/EventListing.ts"],"sourcesContent":["import { parseHTML } from 'linkedom';\nimport { extractJsonLd } from '../../jsonld.js';\n\nexport interface EventListingData {\n  name: string;\n  startDate: string;\n  endDate?: string;\n  location?: string;\n  description?: string;\n  url: string;\n}\n\nexport async function extractEventListing(html: string, url: string): Promise<EventListingData | null> {\n  if (!html) return null;\n\n  const fromJsonLd = tryJsonLd(html, url);\n  if (fromJsonLd) return fromJsonLd;\n\n  return tryMetaFallback(html, url);\n}\n\nfunction tryJsonLd(html: string, url: string): EventListingData | null {\n  let blocks: Record<string, unknown>[];\n  try {\n    blocks = extractJsonLd(html);\n  } catch {\n    return null;\n  }\n\n  const ev = blocks.find((block) => typeIncludes(block['@type'], 'event'));\n  if (!ev) return null;\n\n  const name = stringField(ev['name']);\n  const startDate = stringField(ev['startDate']);\n  if (!name && !startDate) return null;\n\n  const data: EventListingData = {\n    name: name ?? '',\n    startDate: startDate ?? '',\n    url,\n  };\n  const endDate = stringField(ev['endDate']);\n  if (endDate) data.endDate = endDate;\n  const description = stringField(ev['description']);\n  if (description) data.description = description;\n  const location = readLocation(ev['location']);\n  if (location) data.location = location;\n  return data;\n}\n\nfunction tryMetaFallback(html: string, url: string): EventListingData | null {\n  let document: Document;\n  try {\n    ({ document } = parseHTML(html));\n  } catch {\n    return null;\n  }\n\n  const startMeta = metaContent(document, 'meta[property=\"event:start_time\"]');\n  const endMeta = metaContent(document, 'meta[property=\"event:end_time\"]');\n  const locationMeta = metaContent(document, 'meta[property=\"event:location\"]');\n  const description = metaContent(document, 'meta[property=\"og:description\"]');\n\n  let start = startMeta;\n  if (!start) {\n    const timeEl = document.querySelector('time[itemprop=\"startDate\"]');\n    const dt = timeEl?.getAttribute('datetime')?.trim();\n    if (dt) start = dt;\n  }\n\n  let name: string | undefined;\n  const h1 = document.querySelector('h1');\n  const h1Text = h1?.textContent?.trim();\n  if (h1Text) name = h1Text;\n  if (!name) {\n    name = metaContent(document, 'meta[property=\"og:title\"]');\n  }\n\n  if (!name && !start) return null;\n\n  const data: EventListingData = {\n    name: name ?? '',\n    startDate: start ?? '',\n    url,\n  };\n  if (endMeta) data.endDate = endMeta;\n  if (locationMeta) data.location = locationMeta;\n  if (description) data.description = description;\n  return data;\n}\n\nfunction metaContent(document: Document, selector: string): string | undefined {\n  const el = document.querySelector(selector);\n  const content = el?.getAttribute('content')?.trim();\n  return content && content.length > 0 ? content : undefined;\n}\n\nfunction typeIncludes(raw: unknown, want: string): boolean {\n  const target = want.toLowerCase();\n  if (typeof raw === 'string') return normalizeType(raw) === target;\n  if (Array.isArray(raw)) {\n    return raw.some((entry) => typeof entry === 'string' && normalizeType(entry) === target);\n  }\n  return false;\n}\n\nfunction normalizeType(raw: string): string {\n  const tail = raw.split(/[/#:]/).pop() ?? raw;\n  return tail.toLowerCase();\n}\n\nfunction stringField(value: unknown): string | undefined {\n  if (typeof value !== 'string') return undefined;\n  const trimmed = value.trim();\n  return trimmed.length > 0 ? trimmed : undefined;\n}\n\nfunction readLocation(value: unknown): string | undefined {\n  if (typeof value === 'string') return stringField(value);\n  if (Array.isArray(value)) {\n    for (const entry of value) {\n      const loc = readLocation(entry);\n      if (loc) return loc;\n    }\n    return undefined;\n  }\n  if (value && typeof value === 'object') {\n    const obj = value as Record<string, unknown>;\n    const name = stringField(obj['name']);\n    if (name) return name;\n    const address = obj['address'];\n    if (typeof address === 'string') return stringField(address);\n    if (address && typeof address === 'object') {\n      const aobj = address as Record<string, unknown>;\n      const parts = [\n        stringField(aobj['streetAddress']),\n        stringField(aobj['addressLocality']),\n        stringField(aobj['addressRegion']),\n        stringField(aobj['addressCountry']),\n      ].filter((s): s is string => Boolean(s));\n      if (parts.length > 0) return parts.join(', ');\n    }\n  }\n  return undefined;\n}\n"],"mappings":"AAAA,SAAS,iBAAiB;AAC1B,SAAS,qBAAqB;AAW9B,eAAsB,oBAAoB,MAAc,KAA+C;AACrG,MAAI,CAAC,KAAM,QAAO;AAElB,QAAM,aAAa,UAAU,MAAM,GAAG;AACtC,MAAI,WAAY,QAAO;AAEvB,SAAO,gBAAgB,MAAM,GAAG;AAClC;AAEA,SAAS,UAAU,MAAc,KAAsC;AACrE,MAAI;AACJ,MAAI;AACF,aAAS,cAAc,IAAI;AAAA,EAC7B,QAAQ;AACN,WAAO;AAAA,EACT;AAEA,QAAM,KAAK,OAAO,KAAK,CAAC,UAAU,aAAa,MAAM,OAAO,GAAG,OAAO,CAAC;AACvE,MAAI,CAAC,GAAI,QAAO;AAEhB,QAAM,OAAO,YAAY,GAAG,MAAM,CAAC;AACnC,QAAM,YAAY,YAAY,GAAG,WAAW,CAAC;AAC7C,MAAI,CAAC,QAAQ,CAAC,UAAW,QAAO;AAEhC,QAAM,OAAyB;AAAA,IAC7B,MAAM,QAAQ;AAAA,IACd,WAAW,aAAa;AAAA,IACxB;AAAA,EACF;AACA,QAAM,UAAU,YAAY,GAAG,SAAS,CAAC;AACzC,MAAI,QAAS,MAAK,UAAU;AAC5B,QAAM,cAAc,YAAY,GAAG,aAAa,CAAC;AACjD,MAAI,YAAa,MAAK,cAAc;AACpC,QAAM,WAAW,aAAa,GAAG,UAAU,CAAC;AAC5C,MAAI,SAAU,MAAK,WAAW;AAC9B,SAAO;AACT;AAEA,SAAS,gBAAgB,MAAc,KAAsC;AAC3E,MAAI;AACJ,MAAI;AACF,KAAC,EAAE,SAAS,IAAI,UAAU,IAAI;AAAA,EAChC,QAAQ;AACN,WAAO;AAAA,EACT;AAEA,QAAM,YAAY,YAAY,UAAU,mCAAmC;AAC3E,QAAM,UAAU,YAAY,UAAU,iCAAiC;AACvE,QAAM,eAAe,YAAY,UAAU,iCAAiC;AAC5E,QAAM,cAAc,YAAY,UAAU,iCAAiC;AAE3E,MAAI,QAAQ;AACZ,MAAI,CAAC,OAAO;AACV,UAAM,SAAS,SAAS,cAAc,4BAA4B;AAClE,UAAM,KAAK,QAAQ,aAAa,UAAU,GAAG,KAAK;AAClD,QAAI,GAAI,SAAQ;AAAA,EAClB;AAEA,MAAI;AACJ,QAAM,KAAK,SAAS,cAAc,IAAI;AACtC,QAAM,SAAS,IAAI,aAAa,KAAK;AACrC,MAAI,OAAQ,QAAO;AACnB,MAAI,CAAC,MAAM;AACT,WAAO,YAAY,UAAU,2BAA2B;AAAA,EAC1D;AAEA,MAAI,CAAC,QAAQ,CAAC,MAAO,QAAO;AAE5B,QAAM,OAAyB;AAAA,IAC7B,MAAM,QAAQ;AAAA,IACd,WAAW,SAAS;AAAA,IACpB;AAAA,EACF;AACA,MAAI,QAAS,MAAK,UAAU;AAC5B,MAAI,aAAc,MAAK,WAAW;AAClC,MAAI,YAAa,MAAK,cAAc;AACpC,SAAO;AACT;AAEA,SAAS,YAAY,UAAoB,UAAsC;AAC7E,QAAM,KAAK,SAAS,cAAc,QAAQ;AAC1C,QAAM,UAAU,IAAI,aAAa,SAAS,GAAG,KAAK;AAClD,SAAO,WAAW,QAAQ,SAAS,IAAI,UAAU;AACnD;AAEA,SAAS,aAAa,KAAc,MAAuB;AACzD,QAAM,SAAS,KAAK,YAAY;AAChC,MAAI,OAAO,QAAQ,SAAU,QAAO,cAAc,GAAG,MAAM;AAC3D,MAAI,MAAM,QAAQ,GAAG,GAAG;AACtB,WAAO,IAAI,KAAK,CAAC,UAAU,OAAO,UAAU,YAAY,cAAc,KAAK,MAAM,MAAM;AAAA,EACzF;AACA,SAAO;AACT;AAEA,SAAS,cAAc,KAAqB;AAC1C,QAAM,OAAO,IAAI,MAAM,OAAO,EAAE,IAAI,KAAK;AACzC,SAAO,KAAK,YAAY;AAC1B;AAEA,SAAS,YAAY,OAAoC;AACvD,MAAI,OAAO,UAAU,SAAU,QAAO;AACtC,QAAM,UAAU,MAAM,KAAK;AAC3B,SAAO,QAAQ,SAAS,IAAI,UAAU;AACxC;AAEA,SAAS,aAAa,OAAoC;AACxD,MAAI,OAAO,UAAU,SAAU,QAAO,YAAY,KAAK;AACvD,MAAI,MAAM,QAAQ,KAAK,GAAG;AACxB,eAAW,SAAS,OAAO;AACzB,YAAM,MAAM,aAAa,KAAK;AAC9B,UAAI,IAAK,QAAO;AAAA,IAClB;AACA,WAAO;AAAA,EACT;AACA,MAAI,SAAS,OAAO,UAAU,UAAU;AACtC,UAAM,MAAM;AACZ,UAAM,OAAO,YAAY,IAAI,MAAM,CAAC;AACpC,QAAI,KAAM,QAAO;AACjB,UAAM,UAAU,IAAI,SAAS;AAC7B,QAAI,OAAO,YAAY,SAAU,QAAO,YAAY,OAAO;AAC3D,QAAI,WAAW,OAAO,YAAY,UAAU;AAC1C,YAAM,OAAO;AACb,YAAM,QAAQ;AAAA,QACZ,YAAY,KAAK,eAAe,CAAC;AAAA,QACjC,YAAY,KAAK,iBAAiB,CAAC;AAAA,QACnC,YAAY,KAAK,eAAe,CAAC;AAAA,QACjC,YAAY,KAAK,gBAAgB,CAAC;AAAA,MACpC,EAAE,OAAO,CAAC,MAAmB,QAAQ,CAAC,CAAC;AACvC,UAAI,MAAM,SAAS,EAAG,QAAO,MAAM,KAAK,IAAI;AAAA,IAC9C;AAAA,EACF;AACA,SAAO;AACT;","names":[]}

package/dist/extraction/v1/schemas/Paper.d.ts

@@ -0,0 +1,10 @@
+export interface PaperData {
+    title: string;
+    authors: string[];
+    abstract: string;
+    publishedDate?: string;
+    doi?: string;
+    arxivId?: string;
+}
+export declare function extractPaper(html: string, url: string): Promise<PaperData | null>;
+//# sourceMappingURL=Paper.d.ts.map

package/dist/extraction/v1/schemas/Paper.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"Paper.d.ts","sourceRoot":"","sources":["../../../../src/extraction/v1/schemas/Paper.ts"],"names":[],"mappings":"AAGA,MAAM,WAAW,SAAS;IACxB,KAAK,EAAE,MAAM,CAAC;IACd,OAAO,EAAE,MAAM,EAAE,CAAC;IAClB,QAAQ,EAAE,MAAM,CAAC;IACjB,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,GAAG,CAAC,EAAE,MAAM,CAAC;IACb,OAAO,CAAC,EAAE,MAAM,CAAC;CAClB;AAID,wBAAsB,YAAY,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,OAAO,CAAC,SAAS,GAAG,IAAI,CAAC,CAOvF"}