@ncukondo/search-hub 0.12.1 → 0.12.2

package/dist/node_modules/@ncukondo/academic-fulltext/dist/discovery/index.js

@@ -0,0 +1,139 @@
+import { checkArxiv } from "./arxiv.js";
+import { checkCore } from "./core.js";
+import { resolveDoiToPmcid } from "./ncbi-id-converter.js";
+import { checkPmc } from "./pmc.js";
+import { checkUnpaywallDetailed } from "./unpaywall.js";
+const DEFAULT_SOURCE_ORDER = ["pmc", "arxiv", "unpaywall", "core"];
+async function checkPmcSource(article) {
+  if (!article.pmid && !article.pmcid)
+    return void 0;
+  const ids = {};
+  if (article.pmid)
+    ids.pmid = article.pmid;
+  if (article.pmcid)
+    ids.pmcid = article.pmcid;
+  return await checkPmc(ids);
+}
+function checkArxivSource(article) {
+  if (!article.arxivId)
+    return void 0;
+  return checkArxiv(article.arxivId);
+}
+async function checkCoreSource(article, config) {
+  if (!config.coreApiKey || !article.doi)
+    return void 0;
+  return await checkCore(article.doi, config.coreApiKey);
+}
+const sourceCheckers = {
+  pmc: checkPmcSource,
+  arxiv: checkArxivSource,
+  core: checkCoreSource
+};
+function determineOAStatus(locations, errors, sourcesChecked) {
+  if (locations.length > 0)
+    return "open";
+  if (errors.length > 0 && errors.length >= sourcesChecked)
+    return "unknown";
+  return "closed";
+}
+async function enrichArticleIds(article, config) {
+  const discoveredIds = {};
+  if (!article.doi || article.pmcid || article.pmid) {
+    return { enriched: article, discoveredIds };
+  }
+  try {
+    const email = config.ncbiEmail || config.unpaywallEmail;
+    const options = {};
+    if (config.ncbiTool)
+      options.tool = config.ncbiTool;
+    if (email)
+      options.email = email;
+    const result = await resolveDoiToPmcid(article.doi, options);
+    if (result) {
+      const enriched = { ...article };
+      if (result.pmcid) {
+        enriched.pmcid = result.pmcid;
+        discoveredIds.pmcid = result.pmcid;
+      }
+      if (result.pmid) {
+        enriched.pmid = result.pmid;
+        discoveredIds.pmid = result.pmid;
+      }
+      return { enriched, discoveredIds };
+    }
+  } catch {
+  }
+  return { enriched: article, discoveredIds };
+}
+async function checkUnpaywallSource(enriched, config, state) {
+  if (!config.unpaywallEmail || !enriched.doi)
+    return;
+  state.sourcesChecked++;
+  try {
+    const detailed = await checkUnpaywallDetailed(enriched.doi, config.unpaywallEmail);
+    if (!detailed)
+      return;
+    state.locations.push(...detailed.locations);
+    if (detailed.pmcid) {
+      state.unpaywallPmcid = detailed.pmcid;
+    }
+  } catch (err) {
+    state.errors.push({ source: "unpaywall", error: String(err) });
+  }
+}
+async function checkGenericSource(source, enriched, config, state) {
+  const checker = sourceCheckers[source];
+  if (!checker)
+    return;
+  const result = await runSourceChecker(checker, enriched, config);
+  if (result.skipped)
+    return;
+  state.sourcesChecked++;
+  if (result.error) {
+    state.errors.push({ source, error: result.error });
+  } else if (result.locations) {
+    state.locations.push(...result.locations);
+  }
+}
+async function lazyPmcCheck(enriched, state, discoveredIds) {
+  if (!state.unpaywallPmcid || enriched.pmcid)
+    return;
+  discoveredIds.pmcid = discoveredIds.pmcid ?? state.unpaywallPmcid;
+  try {
+    const pmcLocations = await checkPmc({ pmcid: state.unpaywallPmcid });
+    if (pmcLocations) {
+      state.locations.push(...pmcLocations);
+    }
+  } catch (err) {
+    state.errors.push({ source: "pmc-lazy", error: String(err) });
+  }
+}
+async function discoverOA(article, config) {
+  const { enriched, discoveredIds } = await enrichArticleIds(article, config);
+  const state = { locations: [], errors: [], sourcesChecked: 0 };
+  const sourceOrder = config.preferSources.length > 0 ? config.preferSources : DEFAULT_SOURCE_ORDER;
+  for (const source of sourceOrder) {
+    if (source === "unpaywall") {
+      await checkUnpaywallSource(enriched, config, state);
+    } else {
+      await checkGenericSource(source, enriched, config, state);
+    }
+  }
+  await lazyPmcCheck(enriched, state, discoveredIds);
+  const oaStatus = determineOAStatus(state.locations, state.errors, state.sourcesChecked);
+  return { oaStatus, locations: state.locations, errors: state.errors, discoveredIds };
+}
+async function runSourceChecker(checker, article, config) {
+  try {
+    const result = await checker(article, config);
+    if (result === void 0)
+      return { skipped: true };
+    return { skipped: false, locations: result ?? [] };
+  } catch (err) {
+    return { skipped: false, error: String(err) };
+  }
+}
+export {
+  discoverOA
+};
+//# sourceMappingURL=index.js.map

package/dist/node_modules/@ncukondo/academic-fulltext/dist/discovery/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sources":["../../../../../../node_modules/@ncukondo/academic-fulltext/dist/discovery/index.js"],"sourcesContent":["/**\n * OA Discovery Aggregator.\n * Combines results from multiple OA discovery sources.\n */\nimport { checkArxiv } from \"./arxiv.js\";\nimport { checkCore } from \"./core.js\";\nimport { resolveDoiToPmcid } from \"./ncbi-id-converter.js\";\nimport { checkPmc } from \"./pmc.js\";\nimport { checkUnpaywallDetailed } from \"./unpaywall.js\";\n/** Default source order when preferSources is empty or not specified */\nconst DEFAULT_SOURCE_ORDER = [\"pmc\", \"arxiv\", \"unpaywall\", \"core\"];\nasync function checkPmcSource(article) {\n    if (!article.pmid && !article.pmcid)\n        return undefined;\n    const ids = {};\n    if (article.pmid)\n        ids.pmid = article.pmid;\n    if (article.pmcid)\n        ids.pmcid = article.pmcid;\n    return await checkPmc(ids);\n}\nfunction checkArxivSource(article) {\n    if (!article.arxivId)\n        return undefined;\n    return checkArxiv(article.arxivId);\n}\nasync function checkCoreSource(article, config) {\n    if (!config.coreApiKey || !article.doi)\n        return undefined;\n    return await checkCore(article.doi, config.coreApiKey);\n}\n/** Map of source name to its checker function (excluding unpaywall, handled specially). */\nconst sourceCheckers = {\n    pmc: checkPmcSource,\n    arxiv: checkArxivSource,\n    core: checkCoreSource,\n};\n/** Determine OA status from collected locations and errors. */\nfunction determineOAStatus(locations, errors, sourcesChecked) {\n    if (locations.length > 0)\n        return \"open\";\n    if (errors.length > 0 && errors.length >= sourcesChecked)\n        return \"unknown\";\n    return \"closed\";\n}\n/**\n * Pre-enrich article with PMCID from NCBI ID Converter when only DOI is available.\n */\nasync function enrichArticleIds(article, config) {\n    const discoveredIds = {};\n    // Only enrich if we have a DOI but no pmcid/pmid\n    if (!article.doi || article.pmcid || article.pmid) {\n        return { enriched: article, discoveredIds };\n    }\n    try {\n        const email = config.ncbiEmail || config.unpaywallEmail;\n        const options = {};\n        if (config.ncbiTool)\n            options.tool = config.ncbiTool;\n        if (email)\n            options.email = email;\n        const result = await resolveDoiToPmcid(article.doi, options);\n        if (result) {\n            const enriched = { ...article };\n            if (result.pmcid) {\n                enriched.pmcid = result.pmcid;\n                discoveredIds.pmcid = result.pmcid;\n            }\n            if (result.pmid) {\n                enriched.pmid = result.pmid;\n                discoveredIds.pmid = result.pmid;\n            }\n            return { enriched, discoveredIds };\n        }\n    }\n    catch {\n        // NCBI ID Converter failure is non-fatal; continue without enrichment\n    }\n    return { enriched: article, discoveredIds };\n}\n/** Check Unpaywall with PMCID extraction. */\nasync function checkUnpaywallSource(enriched, config, state) {\n    if (!config.unpaywallEmail || !enriched.doi)\n        return;\n    state.sourcesChecked++;\n    try {\n        const detailed = await checkUnpaywallDetailed(enriched.doi, config.unpaywallEmail);\n        if (!detailed)\n            return;\n        state.locations.push(...detailed.locations);\n        if (detailed.pmcid) {\n            state.unpaywallPmcid = detailed.pmcid;\n        }\n    }\n    catch (err) {\n        state.errors.push({ source: \"unpaywall\", error: String(err) });\n    }\n}\n/** Check a single non-unpaywall source, updating state. */\nasync function checkGenericSource(source, enriched, config, state) {\n    const checker = sourceCheckers[source];\n    if (!checker)\n        return;\n    const result = await runSourceChecker(checker, enriched, config);\n    if (result.skipped)\n        return;\n    state.sourcesChecked++;\n    if (result.error) {\n        state.errors.push({ source, error: result.error });\n    }\n    else if (result.locations) {\n        state.locations.push(...result.locations);\n    }\n}\n/** Perform lazy PMC check if Unpaywall revealed a PMCID not already known. */\nasync function lazyPmcCheck(enriched, state, discoveredIds) {\n    if (!state.unpaywallPmcid || enriched.pmcid)\n        return;\n    discoveredIds.pmcid = discoveredIds.pmcid ?? state.unpaywallPmcid;\n    try {\n        const pmcLocations = await checkPmc({ pmcid: state.unpaywallPmcid });\n        if (pmcLocations) {\n            state.locations.push(...pmcLocations);\n        }\n    }\n    catch (err) {\n        state.errors.push({ source: \"pmc-lazy\", error: String(err) });\n    }\n}\n/**\n * Discover OA availability for an article across all configured sources.\n * Checks sources in the order specified by config.preferSources, falling back\n * to the default order. Individual source errors are caught and reported\n * without failing the whole discovery.\n *\n * When only a DOI is provided, pre-enriches with NCBI ID Converter to resolve\n * PMCID/PMID, enabling PMC discovery. Also extracts PMCID from Unpaywall URLs\n * as a fallback, performing a lazy PMC check if a PMCID is discovered.\n */\nexport async function discoverOA(article, config) {\n    const { enriched, discoveredIds } = await enrichArticleIds(article, config);\n    const state = { locations: [], errors: [], sourcesChecked: 0 };\n    const sourceOrder = config.preferSources.length > 0 ? config.preferSources : DEFAULT_SOURCE_ORDER;\n    for (const source of sourceOrder) {\n        if (source === \"unpaywall\") {\n            await checkUnpaywallSource(enriched, config, state);\n        }\n        else {\n            await checkGenericSource(source, enriched, config, state);\n        }\n    }\n    await lazyPmcCheck(enriched, state, discoveredIds);\n    const oaStatus = determineOAStatus(state.locations, state.errors, state.sourcesChecked);\n    return { oaStatus, locations: state.locations, errors: state.errors, discoveredIds };\n}\n/** Run a single source checker with error handling. */\nasync function runSourceChecker(checker, article, config) {\n    try {\n        const result = await checker(article, config);\n        if (result === undefined)\n            return { skipped: true };\n        return { skipped: false, locations: result ?? [] };\n    }\n    catch (err) {\n        return { skipped: false, error: String(err) };\n    }\n}\n//# sourceMappingURL=index.js.map"],"names":[],"mappings":";;;;;AAUA,MAAM,uBAAuB,CAAC,OAAO,SAAS,aAAa,MAAM;AACjE,eAAe,eAAe,SAAS;AACnC,MAAI,CAAC,QAAQ,QAAQ,CAAC,QAAQ;AAC1B,WAAO;AACX,QAAM,MAAM,CAAA;AACZ,MAAI,QAAQ;AACR,QAAI,OAAO,QAAQ;AACvB,MAAI,QAAQ;AACR,QAAI,QAAQ,QAAQ;AACxB,SAAO,MAAM,SAAS,GAAG;AAC7B;AACA,SAAS,iBAAiB,SAAS;AAC/B,MAAI,CAAC,QAAQ;AACT,WAAO;AACX,SAAO,WAAW,QAAQ,OAAO;AACrC;AACA,eAAe,gBAAgB,SAAS,QAAQ;AAC5C,MAAI,CAAC,OAAO,cAAc,CAAC,QAAQ;AAC/B,WAAO;AACX,SAAO,MAAM,UAAU,QAAQ,KAAK,OAAO,UAAU;AACzD;AAEA,MAAM,iBAAiB;AAAA,EACnB,KAAK;AAAA,EACL,OAAO;AAAA,EACP,MAAM;AACV;AAEA,SAAS,kBAAkB,WAAW,QAAQ,gBAAgB;AAC1D,MAAI,UAAU,SAAS;AACnB,WAAO;AACX,MAAI,OAAO,SAAS,KAAK,OAAO,UAAU;AACtC,WAAO;AACX,SAAO;AACX;AAIA,eAAe,iBAAiB,SAAS,QAAQ;AAC7C,QAAM,gBAAgB,CAAA;AAEtB,MAAI,CAAC,QAAQ,OAAO,QAAQ,SAAS,QAAQ,MAAM;AAC/C,WAAO,EAAE,UAAU,SAAS,cAAa;AAAA,EAC7C;AACA,MAAI;AACA,UAAM,QAAQ,OAAO,aAAa,OAAO;AACzC,UAAM,UAAU,CAAA;AAChB,QAAI,OAAO;AACP,cAAQ,OAAO,OAAO;AAC1B,QAAI;AACA,cAAQ,QAAQ;AACpB,UAAM,SAAS,MAAM,kBAAkB,QAAQ,KAAK,OAAO;AAC3D,QAAI,QAAQ;AACR,YAAM,WAAW,EAAE,GAAG,QAAO;AAC7B,UAAI,OAAO,OAAO;AACd,iBAAS,QAAQ,OAAO;AACxB,sBAAc,QAAQ,OAAO;AAAA,MACjC;AACA,UAAI,OAAO,MAAM;AACb,iBAAS,OAAO,OAAO;AACvB,sBAAc,OAAO,OAAO;AAAA,MAChC;AACA,aAAO,EAAE,UAAU,cAAa;AAAA,IACpC;AAAA,EACJ,QACM;AAAA,EAEN;AACA,SAAO,EAAE,UAAU,SAAS,cAAa;AAC7C;AAEA,eAAe,qBAAqB,UAAU,QAAQ,OAAO;AACzD,MAAI,CAAC,OAAO,kBAAkB,CAAC,SAAS;AACpC;AACJ,QAAM;AACN,MAAI;AACA,UAAM,WAAW,MAAM,uBAAuB,SAAS,KAAK,OAAO,cAAc;AACjF,QAAI,CAAC;AACD;AACJ,UAAM,UAAU,KAAK,GAAG,SAAS,SAAS;AAC1C,QAAI,SAAS,OAAO;AAChB,YAAM,iBAAiB,SAAS;AAAA,IACpC;AAAA,EACJ,SACO,KAAK;AACR,UAAM,OAAO,KAAK,EAAE,QAAQ,aAAa,OAAO,OAAO,GAAG,GAAG;AAAA,EACjE;AACJ;AAEA,eAAe,mBAAmB,QAAQ,UAAU,QAAQ,OAAO;AAC/D,QAAM,UAAU,eAAe,MAAM;AACrC,MAAI,CAAC;AACD;AACJ,QAAM,SAAS,MAAM,iBAAiB,SAAS,UAAU,MAAM;AAC/D,MAAI,OAAO;AACP;AACJ,QAAM;AACN,MAAI,OAAO,OAAO;AACd,UAAM,OAAO,KAAK,EAAE,QAAQ,OAAO,OAAO,OAAO;AAAA,EACrD,WACS,OAAO,WAAW;AACvB,UAAM,UAAU,KAAK,GAAG,OAAO,SAAS;AAAA,EAC5C;AACJ;AAEA,eAAe,aAAa,UAAU,OAAO,eAAe;AACxD,MAAI,CAAC,MAAM,kBAAkB,SAAS;AAClC;AACJ,gBAAc,QAAQ,cAAc,SAAS,MAAM;AACnD,MAAI;AACA,UAAM,eAAe,MAAM,SAAS,EAAE,OAAO,MAAM,gBAAgB;AACnE,QAAI,cAAc;AACd,YAAM,UAAU,KAAK,GAAG,YAAY;AAAA,IACxC;AAAA,EACJ,SACO,KAAK;AACR,UAAM,OAAO,KAAK,EAAE,QAAQ,YAAY,OAAO,OAAO,GAAG,GAAG;AAAA,EAChE;AACJ;AAWO,eAAe,WAAW,SAAS,QAAQ;AAC9C,QAAM,EAAE,UAAU,cAAa,IAAK,MAAM,iBAAiB,SAAS,MAAM;AAC1E,QAAM,QAAQ,EAAE,WAAW,CAAA,GAAI,QAAQ,CAAA,GAAI,gBAAgB,EAAC;AAC5D,QAAM,cAAc,OAAO,cAAc,SAAS,IAAI,OAAO,gBAAgB;AAC7E,aAAW,UAAU,aAAa;AAC9B,QAAI,WAAW,aAAa;AACxB,YAAM,qBAAqB,UAAU,QAAQ,KAAK;AAAA,IACtD,OACK;AACD,YAAM,mBAAmB,QAAQ,UAAU,QAAQ,KAAK;AAAA,IAC5D;AAAA,EACJ;AACA,QAAM,aAAa,UAAU,OAAO,aAAa;AACjD,QAAM,WAAW,kBAAkB,MAAM,WAAW,MAAM,QAAQ,MAAM,cAAc;AACtF,SAAO,EAAE,UAAU,WAAW,MAAM,WAAW,QAAQ,MAAM,QAAQ,cAAa;AACtF;AAEA,eAAe,iBAAiB,SAAS,SAAS,QAAQ;AACtD,MAAI;AACA,UAAM,SAAS,MAAM,QAAQ,SAAS,MAAM;AAC5C,QAAI,WAAW;AACX,aAAO,EAAE,SAAS,KAAI;AAC1B,WAAO,EAAE,SAAS,OAAO,WAAW,UAAU,CAAA,EAAE;AAAA,EACpD,SACO,KAAK;AACR,WAAO,EAAE,SAAS,OAAO,OAAO,OAAO,GAAG,EAAC;AAAA,EAC/C;AACJ;","x_google_ignoreList":[0]}

package/dist/node_modules/@ncukondo/academic-fulltext/dist/discovery/ncbi-id-converter.js

@@ -0,0 +1,46 @@
+const IDCONV_BASE_URL = "https://www.ncbi.nlm.nih.gov/pmc/utils/idconv/v1.0/";
+function buildUrl(ids, options) {
+  const params = new URLSearchParams({
+    ids: ids.join(","),
+    format: "json"
+  });
+  if (options?.tool)
+    params.set("tool", options.tool);
+  if (options?.email)
+    params.set("email", options.email);
+  return `${IDCONV_BASE_URL}?${params.toString()}`;
+}
+function parseRecord(record) {
+  if (record.errmsg)
+    return null;
+  const result = {};
+  if (record.pmcid)
+    result.pmcid = record.pmcid;
+  if (record.pmid)
+    result.pmid = record.pmid;
+  if (record.doi)
+    result.doi = record.doi;
+  return Object.keys(result).length > 0 ? result : null;
+}
+async function resolveDoiToPmcid(doi, options) {
+  if (!doi)
+    return null;
+  const url = buildUrl([doi], options);
+  const response = await fetch(url);
+  if (!response.ok) {
+    if (response.status === 404)
+      return null;
+    throw new Error(`NCBI ID Converter API error: HTTP ${response.status} ${response.statusText}`);
+  }
+  const data = await response.json();
+  if (!data.records || data.records.length === 0)
+    return null;
+  const record = data.records[0];
+  if (!record)
+    return null;
+  return parseRecord(record);
+}
+export {
+  resolveDoiToPmcid
+};
+//# sourceMappingURL=ncbi-id-converter.js.map

package/dist/node_modules/@ncukondo/academic-fulltext/dist/discovery/ncbi-id-converter.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"ncbi-id-converter.js","sources":["../../../../../../node_modules/@ncukondo/academic-fulltext/dist/discovery/ncbi-id-converter.js"],"sourcesContent":["/**\n * NCBI ID Converter API client.\n * Resolves DOI → PMCID/PMID using the NCBI ID Converter API.\n *\n * API: https://www.ncbi.nlm.nih.gov/pmc/utils/idconv/v1.0/?ids={doi}&format=json\n */\nconst IDCONV_BASE_URL = \"https://www.ncbi.nlm.nih.gov/pmc/utils/idconv/v1.0/\";\nfunction buildUrl(ids, options) {\n    const params = new URLSearchParams({\n        ids: ids.join(\",\"),\n        format: \"json\",\n    });\n    if (options?.tool)\n        params.set(\"tool\", options.tool);\n    if (options?.email)\n        params.set(\"email\", options.email);\n    return `${IDCONV_BASE_URL}?${params.toString()}`;\n}\nfunction parseRecord(record) {\n    if (record.errmsg)\n        return null;\n    const result = {};\n    if (record.pmcid)\n        result.pmcid = record.pmcid;\n    if (record.pmid)\n        result.pmid = record.pmid;\n    if (record.doi)\n        result.doi = record.doi;\n    return Object.keys(result).length > 0 ? result : null;\n}\n/**\n * Resolve a single DOI to PMCID/PMID via the NCBI ID Converter API.\n *\n * @param doi - The article's DOI\n * @param options - Optional tool name and email for API identification\n * @returns Conversion result with pmcid/pmid, or null if not found\n */\nexport async function resolveDoiToPmcid(doi, options) {\n    if (!doi)\n        return null;\n    const url = buildUrl([doi], options);\n    const response = await fetch(url);\n    if (!response.ok) {\n        if (response.status === 404)\n            return null;\n        throw new Error(`NCBI ID Converter API error: HTTP ${response.status} ${response.statusText}`);\n    }\n    const data = (await response.json());\n    if (!data.records || data.records.length === 0)\n        return null;\n    const record = data.records[0];\n    if (!record)\n        return null;\n    return parseRecord(record);\n}\n/**\n * Batch resolve multiple IDs (DOIs, PMIDs, PMCIDs) to their cross-references.\n *\n * @param ids - Array of identifiers to resolve\n * @param options - Optional tool name and email for API identification\n * @returns Map of input ID → conversion result\n */\nexport async function batchResolveIds(ids, options) {\n    const results = new Map();\n    if (ids.length === 0)\n        return results;\n    const url = buildUrl(ids, options);\n    const response = await fetch(url);\n    if (!response.ok) {\n        throw new Error(`NCBI ID Converter API error: HTTP ${response.status} ${response.statusText}`);\n    }\n    const data = (await response.json());\n    if (!data.records)\n        return results;\n    for (const record of data.records) {\n        const parsed = parseRecord(record);\n        if (!parsed)\n            continue;\n        // Map back to the input ID: try doi, pmid, pmcid\n        const key = record.doi ?? record.pmid ?? record.pmcid;\n        if (key)\n            results.set(key, parsed);\n    }\n    return results;\n}\n//# sourceMappingURL=ncbi-id-converter.js.map"],"names":[],"mappings":"AAMA,MAAM,kBAAkB;AACxB,SAAS,SAAS,KAAK,SAAS;AAC5B,QAAM,SAAS,IAAI,gBAAgB;AAAA,IAC/B,KAAK,IAAI,KAAK,GAAG;AAAA,IACjB,QAAQ;AAAA,EAChB,CAAK;AACD,MAAI,SAAS;AACT,WAAO,IAAI,QAAQ,QAAQ,IAAI;AACnC,MAAI,SAAS;AACT,WAAO,IAAI,SAAS,QAAQ,KAAK;AACrC,SAAO,GAAG,eAAe,IAAI,OAAO,SAAQ,CAAE;AAClD;AACA,SAAS,YAAY,QAAQ;AACzB,MAAI,OAAO;AACP,WAAO;AACX,QAAM,SAAS,CAAA;AACf,MAAI,OAAO;AACP,WAAO,QAAQ,OAAO;AAC1B,MAAI,OAAO;AACP,WAAO,OAAO,OAAO;AACzB,MAAI,OAAO;AACP,WAAO,MAAM,OAAO;AACxB,SAAO,OAAO,KAAK,MAAM,EAAE,SAAS,IAAI,SAAS;AACrD;AAQO,eAAe,kBAAkB,KAAK,SAAS;AAClD,MAAI,CAAC;AACD,WAAO;AACX,QAAM,MAAM,SAAS,CAAC,GAAG,GAAG,OAAO;AACnC,QAAM,WAAW,MAAM,MAAM,GAAG;AAChC,MAAI,CAAC,SAAS,IAAI;AACd,QAAI,SAAS,WAAW;AACpB,aAAO;AACX,UAAM,IAAI,MAAM,qCAAqC,SAAS,MAAM,IAAI,SAAS,UAAU,EAAE;AAAA,EACjG;AACA,QAAM,OAAQ,MAAM,SAAS;AAC7B,MAAI,CAAC,KAAK,WAAW,KAAK,QAAQ,WAAW;AACzC,WAAO;AACX,QAAM,SAAS,KAAK,QAAQ,CAAC;AAC7B,MAAI,CAAC;AACD,WAAO;AACX,SAAO,YAAY,MAAM;AAC7B;","x_google_ignoreList":[0]}

package/dist/{fulltext → node_modules/@ncukondo/academic-fulltext/dist}/discovery/pmc.js

@@ -31,11 +31,14 @@ async function lookupPmcid(pmid, options) {
   }
   const data = await response.json();
   const linksets = data.linksets;
-  if (!linksets || linksets.length === 0) return null;
+  if (!linksets || linksets.length === 0)
+    return null;
   const firstLinkset = linksets[0];
-  if (!firstLinkset?.linksetdbs) return null;
+  if (!firstLinkset?.linksetdbs)
+    return null;
   const pmcLink = firstLinkset.linksetdbs.find((db) => db.dbto === "pmc");
-  if (!pmcLink?.links || pmcLink.links.length === 0) return null;
+  if (!pmcLink?.links || pmcLink.links.length === 0)
+    return null;
   const pmcNumericId = pmcLink.links[0];
   return pmcNumericId ? `PMC${pmcNumericId}` : null;
 }
@@ -45,7 +48,8 @@ async function checkPmc(ids, options) {
   }
   if (ids.pmid) {
     const pmcid = await lookupPmcid(ids.pmid);
-    if (!pmcid) return null;
+    if (!pmcid)
+      return null;
     return getPmcUrls(pmcid);
   }
   return null;

package/dist/node_modules/@ncukondo/academic-fulltext/dist/discovery/pmc.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"pmc.js","sources":["../../../../../../node_modules/@ncukondo/academic-fulltext/dist/discovery/pmc.js"],"sourcesContent":["/**\n * PMC OA discovery client.\n * Checks PubMed Central availability and generates download URLs.\n *\n * PDF: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC{id}/pdf/\n * XML: https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pmc&id={pmcid}&rettype=xml\n * PMID→PMCID: https://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&db=pmc&id={pmid}&retmode=json\n */\nconst EUTILS_BASE = \"https://eutils.ncbi.nlm.nih.gov/entrez/eutils\";\n/** Strip \"PMC\" prefix from PMCID, returning just the numeric part */\nfunction stripPmcPrefix(pmcid) {\n    return pmcid.replace(/^PMC/i, \"\");\n}\n/** Ensure PMCID has the \"PMC\" prefix */\nfunction ensurePmcPrefix(pmcid) {\n    return pmcid.startsWith(\"PMC\") ? pmcid : `PMC${pmcid}`;\n}\n/**\n * Generate PMC download URLs from a known PMCID.\n */\nexport function getPmcUrls(pmcid) {\n    const numericId = stripPmcPrefix(pmcid);\n    const fullPmcid = ensurePmcPrefix(pmcid);\n    return [\n        {\n            source: \"pmc\",\n            url: `https://www.ncbi.nlm.nih.gov/pmc/articles/${fullPmcid}/pdf/`,\n            urlType: \"pdf\",\n            version: \"published\",\n        },\n        {\n            source: \"pmc\",\n            url: `${EUTILS_BASE}/efetch.fcgi?db=pmc&id=${numericId}&rettype=xml`,\n            urlType: \"xml\",\n            version: \"published\",\n        },\n    ];\n}\n/**\n * Look up PMCID from PMID via E-utilities elink API.\n * @returns PMCID string or null if not in PMC\n */\nasync function lookupPmcid(pmid, options) {\n    let url = `${EUTILS_BASE}/elink.fcgi?dbfrom=pubmed&db=pmc&id=${pmid}&retmode=json`;\n    if (options?.apiKey) {\n        url += `&api_key=${options.apiKey}`;\n    }\n    const response = await fetch(url);\n    if (!response.ok) {\n        throw new Error(`PMC elink API error: HTTP ${response.status}`);\n    }\n    const data = (await response.json());\n    const linksets = data.linksets;\n    if (!linksets || linksets.length === 0)\n        return null;\n    const firstLinkset = linksets[0];\n    if (!firstLinkset?.linksetdbs)\n        return null;\n    const pmcLink = firstLinkset.linksetdbs.find((db) => db.dbto === \"pmc\");\n    if (!pmcLink?.links || pmcLink.links.length === 0)\n        return null;\n    const pmcNumericId = pmcLink.links[0];\n    return pmcNumericId ? `PMC${pmcNumericId}` : null;\n}\n/**\n * Check PMC availability for an article.\n * If PMCID is known, generates URLs directly.\n * If only PMID is known, looks up PMCID via E-utilities elink.\n */\nexport async function checkPmc(ids, options) {\n    // If PMCID is already known, generate URLs directly\n    if (ids.pmcid) {\n        return getPmcUrls(ids.pmcid);\n    }\n    // If only PMID is known, look up PMCID\n    if (ids.pmid) {\n        const pmcid = await lookupPmcid(ids.pmid, options);\n        if (!pmcid)\n            return null;\n        return getPmcUrls(pmcid);\n    }\n    // No identifiers provided\n    return null;\n}\n//# sourceMappingURL=pmc.js.map"],"names":[],"mappings":"AAQA,MAAM,cAAc;AAEpB,SAAS,eAAe,OAAO;AAC3B,SAAO,MAAM,QAAQ,SAAS,EAAE;AACpC;AAEA,SAAS,gBAAgB,OAAO;AAC5B,SAAO,MAAM,WAAW,KAAK,IAAI,QAAQ,MAAM,KAAK;AACxD;AAIO,SAAS,WAAW,OAAO;AAC9B,QAAM,YAAY,eAAe,KAAK;AACtC,QAAM,YAAY,gBAAgB,KAAK;AACvC,SAAO;AAAA,IACH;AAAA,MACI,QAAQ;AAAA,MACR,KAAK,6CAA6C,SAAS;AAAA,MAC3D,SAAS;AAAA,MACT,SAAS;AAAA,IACrB;AAAA,IACQ;AAAA,MACI,QAAQ;AAAA,MACR,KAAK,GAAG,WAAW,0BAA0B,SAAS;AAAA,MACtD,SAAS;AAAA,MACT,SAAS;AAAA,IACrB;AAAA,EACA;AACA;AAKA,eAAe,YAAY,MAAM,SAAS;AACtC,MAAI,MAAM,GAAG,WAAW,uCAAuC,IAAI;AAInE,QAAM,WAAW,MAAM,MAAM,GAAG;AAChC,MAAI,CAAC,SAAS,IAAI;AACd,UAAM,IAAI,MAAM,6BAA6B,SAAS,MAAM,EAAE;AAAA,EAClE;AACA,QAAM,OAAQ,MAAM,SAAS;AAC7B,QAAM,WAAW,KAAK;AACtB,MAAI,CAAC,YAAY,SAAS,WAAW;AACjC,WAAO;AACX,QAAM,eAAe,SAAS,CAAC;AAC/B,MAAI,CAAC,cAAc;AACf,WAAO;AACX,QAAM,UAAU,aAAa,WAAW,KAAK,CAAC,OAAO,GAAG,SAAS,KAAK;AACtE,MAAI,CAAC,SAAS,SAAS,QAAQ,MAAM,WAAW;AAC5C,WAAO;AACX,QAAM,eAAe,QAAQ,MAAM,CAAC;AACpC,SAAO,eAAe,MAAM,YAAY,KAAK;AACjD;AAMO,eAAe,SAAS,KAAK,SAAS;AAEzC,MAAI,IAAI,OAAO;AACX,WAAO,WAAW,IAAI,KAAK;AAAA,EAC/B;AAEA,MAAI,IAAI,MAAM;AACV,UAAM,QAAQ,MAAM,YAAY,IAAI,IAAa;AACjD,QAAI,CAAC;AACD,aAAO;AACX,WAAO,WAAW,KAAK;AAAA,EAC3B;AAEA,SAAO;AACX;","x_google_ignoreList":[0]}

package/dist/{fulltext → node_modules/@ncukondo/academic-fulltext/dist}/discovery/unpaywall.js

@@ -1,4 +1,5 @@
 const UNPAYWALL_BASE_URL = "https://api.unpaywall.org/v2";
+const PMC_URL_PATTERN = /\/pmc\/articles\/(PMC\d+)/i;
 function mapVersion(version) {
   switch (version) {
     case "publishedVersion":
@@ -14,7 +15,8 @@ function mapVersion(version) {
 function toOALocation(loc) {
   const hasPdf = loc.url_for_pdf != null;
   const url = hasPdf ? loc.url_for_pdf : loc.url_for_landing_page;
-  if (!url) return null;
+  if (!url)
+    return null;
   const result = {
     source: "unpaywall",
     url,
@@ -26,32 +28,64 @@ function toOALocation(loc) {
   }
   return result;
 }
-async function checkUnpaywall(doi, email) {
-  if (!doi) return null;
+function extractPmcidFromUrl(url) {
+  const match = PMC_URL_PATTERN.exec(url);
+  return match ? match[1] ?? null : null;
+}
+function extractPmcidFromLocations(locations) {
+  for (const loc of locations) {
+    if (loc.url_for_pdf) {
+      const pmcid = extractPmcidFromUrl(loc.url_for_pdf);
+      if (pmcid)
+        return pmcid;
+    }
+    if (loc.url_for_landing_page) {
+      const pmcid = extractPmcidFromUrl(loc.url_for_landing_page);
+      if (pmcid)
+        return pmcid;
+    }
+  }
+  return null;
+}
+async function fetchUnpaywallData(doi, email) {
+  if (!doi)
+    return null;
   if (!email) {
     throw new Error("Unpaywall email is required for API access");
   }
   const url = `${UNPAYWALL_BASE_URL}/${doi}?email=${encodeURIComponent(email)}`;
   const response = await fetch(url);
   if (!response.ok) {
-    if (response.status === 404) return null;
+    if (response.status === 404)
+      return null;
     if (response.status === 429) {
       throw new Error("Unpaywall rate limit exceeded");
     }
     throw new Error(`Unpaywall API error: HTTP ${response.status} ${response.statusText}`);
   }
-  const data = await response.json();
-  if (!data.is_oa || !data.oa_locations || data.oa_locations.length === 0) {
+  return await response.json();
+}
+async function checkUnpaywallDetailed(doi, email) {
+  const data = await fetchUnpaywallData(doi, email);
+  if (!data || !data.is_oa || !data.oa_locations || data.oa_locations.length === 0) {
     return null;
   }
   const locations = [];
   for (const loc of data.oa_locations) {
     const oaLoc = toOALocation(loc);
-    if (oaLoc) locations.push(oaLoc);
+    if (oaLoc)
+      locations.push(oaLoc);
   }
-  return locations.length > 0 ? locations : null;
+  if (locations.length === 0)
+    return null;
+  const pmcid = extractPmcidFromLocations(data.oa_locations);
+  const result = { locations };
+  if (pmcid)
+    result.pmcid = pmcid;
+  return result;
 }
 export {
-  checkUnpaywall
+  checkUnpaywallDetailed,
+  extractPmcidFromUrl
 };
 //# sourceMappingURL=unpaywall.js.map

package/dist/node_modules/@ncukondo/academic-fulltext/dist/discovery/unpaywall.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"unpaywall.js","sources":["../../../../../../node_modules/@ncukondo/academic-fulltext/dist/discovery/unpaywall.js"],"sourcesContent":["/**\n * Unpaywall OA discovery client.\n * Checks Open Access availability via the Unpaywall API.\n *\n * API: https://api.unpaywall.org/v2/{doi}?email={email}\n * Rate limit: 100,000 requests/day (no per-second limit documented)\n */\nconst UNPAYWALL_BASE_URL = \"https://api.unpaywall.org/v2\";\n/** Regex to extract PMCID from PMC URLs */\nconst PMC_URL_PATTERN = /\\/pmc\\/articles\\/(PMC\\d+)/i;\n/** Map Unpaywall version strings to our OALocation version format */\nfunction mapVersion(version) {\n    switch (version) {\n        case \"publishedVersion\":\n            return \"published\";\n        case \"acceptedVersion\":\n            return \"accepted\";\n        case \"submittedVersion\":\n            return \"submitted\";\n        default:\n            return \"published\";\n    }\n}\n/** Convert an Unpaywall location to our OALocation format */\nfunction toOALocation(loc) {\n    const hasPdf = loc.url_for_pdf != null;\n    const url = hasPdf ? loc.url_for_pdf : loc.url_for_landing_page;\n    if (!url)\n        return null;\n    const result = {\n        source: \"unpaywall\",\n        url,\n        urlType: hasPdf ? \"pdf\" : \"html\",\n        version: mapVersion(loc.version),\n    };\n    if (loc.license) {\n        result.license = loc.license;\n    }\n    return result;\n}\n/**\n * Extract PMCID from a URL containing a PMC article link.\n * E.g., \"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1234567/pdf/\" → \"PMC1234567\"\n */\nexport function extractPmcidFromUrl(url) {\n    const match = PMC_URL_PATTERN.exec(url);\n    return match ? (match[1] ?? null) : null;\n}\n/**\n * Scan Unpaywall locations for a PMCID embedded in any URL.\n */\nfunction extractPmcidFromLocations(locations) {\n    for (const loc of locations) {\n        if (loc.url_for_pdf) {\n            const pmcid = extractPmcidFromUrl(loc.url_for_pdf);\n            if (pmcid)\n                return pmcid;\n        }\n        if (loc.url_for_landing_page) {\n            const pmcid = extractPmcidFromUrl(loc.url_for_landing_page);\n            if (pmcid)\n                return pmcid;\n        }\n    }\n    return null;\n}\n/**\n * Fetch Unpaywall API and return raw response data.\n */\nasync function fetchUnpaywallData(doi, email) {\n    if (!doi)\n        return null;\n    if (!email) {\n        throw new Error(\"Unpaywall email is required for API access\");\n    }\n    const url = `${UNPAYWALL_BASE_URL}/${doi}?email=${encodeURIComponent(email)}`;\n    const response = await fetch(url);\n    if (!response.ok) {\n        if (response.status === 404)\n            return null;\n        if (response.status === 429) {\n            throw new Error(\"Unpaywall rate limit exceeded\");\n        }\n        throw new Error(`Unpaywall API error: HTTP ${response.status} ${response.statusText}`);\n    }\n    return (await response.json());\n}\n/**\n * Check Unpaywall for Open Access availability of an article.\n *\n * @param doi - The article's DOI\n * @param email - Email address required by Unpaywall API (free, no registration)\n * @returns Array of OALocations if OA, null if closed/not found\n * @throws On rate limit (429) or network errors\n */\nexport async function checkUnpaywall(doi, email) {\n    const data = await fetchUnpaywallData(doi, email);\n    if (!data || !data.is_oa || !data.oa_locations || data.oa_locations.length === 0) {\n        return null;\n    }\n    const locations = [];\n    for (const loc of data.oa_locations) {\n        const oaLoc = toOALocation(loc);\n        if (oaLoc)\n            locations.push(oaLoc);\n    }\n    return locations.length > 0 ? locations : null;\n}\n/**\n * Check Unpaywall with detailed results including extracted PMCID.\n * Returns both OA locations and any PMCID found in the location URLs.\n *\n * @param doi - The article's DOI\n * @param email - Email address required by Unpaywall API (free, no registration)\n * @returns Detailed result with locations and optional pmcid, or null if closed/not found\n * @throws On rate limit (429) or network errors\n */\nexport async function checkUnpaywallDetailed(doi, email) {\n    const data = await fetchUnpaywallData(doi, email);\n    if (!data || !data.is_oa || !data.oa_locations || data.oa_locations.length === 0) {\n        return null;\n    }\n    const locations = [];\n    for (const loc of data.oa_locations) {\n        const oaLoc = toOALocation(loc);\n        if (oaLoc)\n            locations.push(oaLoc);\n    }\n    if (locations.length === 0)\n        return null;\n    const pmcid = extractPmcidFromLocations(data.oa_locations);\n    const result = { locations };\n    if (pmcid)\n        result.pmcid = pmcid;\n    return result;\n}\n//# sourceMappingURL=unpaywall.js.map"],"names":[],"mappings":"AAOA,MAAM,qBAAqB;AAE3B,MAAM,kBAAkB;AAExB,SAAS,WAAW,SAAS;AACzB,UAAQ,SAAO;AAAA,IACX,KAAK;AACD,aAAO;AAAA,IACX,KAAK;AACD,aAAO;AAAA,IACX,KAAK;AACD,aAAO;AAAA,IACX;AACI,aAAO;AAAA,EACnB;AACA;AAEA,SAAS,aAAa,KAAK;AACvB,QAAM,SAAS,IAAI,eAAe;AAClC,QAAM,MAAM,SAAS,IAAI,cAAc,IAAI;AAC3C,MAAI,CAAC;AACD,WAAO;AACX,QAAM,SAAS;AAAA,IACX,QAAQ;AAAA,IACR;AAAA,IACA,SAAS,SAAS,QAAQ;AAAA,IAC1B,SAAS,WAAW,IAAI,OAAO;AAAA,EACvC;AACI,MAAI,IAAI,SAAS;AACb,WAAO,UAAU,IAAI;AAAA,EACzB;AACA,SAAO;AACX;AAKO,SAAS,oBAAoB,KAAK;AACrC,QAAM,QAAQ,gBAAgB,KAAK,GAAG;AACtC,SAAO,QAAS,MAAM,CAAC,KAAK,OAAQ;AACxC;AAIA,SAAS,0BAA0B,WAAW;AAC1C,aAAW,OAAO,WAAW;AACzB,QAAI,IAAI,aAAa;AACjB,YAAM,QAAQ,oBAAoB,IAAI,WAAW;AACjD,UAAI;AACA,eAAO;AAAA,IACf;AACA,QAAI,IAAI,sBAAsB;AAC1B,YAAM,QAAQ,oBAAoB,IAAI,oBAAoB;AAC1D,UAAI;AACA,eAAO;AAAA,IACf;AAAA,EACJ;AACA,SAAO;AACX;AAIA,eAAe,mBAAmB,KAAK,OAAO;AAC1C,MAAI,CAAC;AACD,WAAO;AACX,MAAI,CAAC,OAAO;AACR,UAAM,IAAI,MAAM,4CAA4C;AAAA,EAChE;AACA,QAAM,MAAM,GAAG,kBAAkB,IAAI,GAAG,UAAU,mBAAmB,KAAK,CAAC;AAC3E,QAAM,WAAW,MAAM,MAAM,GAAG;AAChC,MAAI,CAAC,SAAS,IAAI;AACd,QAAI,SAAS,WAAW;AACpB,aAAO;AACX,QAAI,SAAS,WAAW,KAAK;AACzB,YAAM,IAAI,MAAM,+BAA+B;AAAA,IACnD;AACA,UAAM,IAAI,MAAM,6BAA6B,SAAS,MAAM,IAAI,SAAS,UAAU,EAAE;AAAA,EACzF;AACA,SAAQ,MAAM,SAAS,KAAI;AAC/B;AA+BO,eAAe,uBAAuB,KAAK,OAAO;AACrD,QAAM,OAAO,MAAM,mBAAmB,KAAK,KAAK;AAChD,MAAI,CAAC,QAAQ,CAAC,KAAK,SAAS,CAAC,KAAK,gBAAgB,KAAK,aAAa,WAAW,GAAG;AAC9E,WAAO;AAAA,EACX;AACA,QAAM,YAAY,CAAA;AAClB,aAAW,OAAO,KAAK,cAAc;AACjC,UAAM,QAAQ,aAAa,GAAG;AAC9B,QAAI;AACA,gBAAU,KAAK,KAAK;AAAA,EAC5B;AACA,MAAI,UAAU,WAAW;AACrB,WAAO;AACX,QAAM,QAAQ,0BAA0B,KAAK,YAAY;AACzD,QAAM,SAAS,EAAE,UAAS;AAC1B,MAAI;AACA,WAAO,QAAQ;AACnB,SAAO;AACX;","x_google_ignoreList":[0]}

package/dist/node_modules/@ncukondo/academic-fulltext/dist/download/arxiv-html.js

@@ -0,0 +1,48 @@
+import { mkdir, writeFile } from "node:fs/promises";
+import { dirname } from "node:path";
+const ARXIV_HTML_BASE = "https://arxiv.org/html/";
+const USER_AGENT = "search-hub/0.8.0 (https://github.com/ncukondo/search-hub)";
+function normalizeArxivId(id) {
+  return id.replace(/^arXiv:/i, "");
+}
+function isHtmlContentType(contentType) {
+  if (!contentType)
+    return false;
+  const base = (contentType.split(";")[0] ?? "").trim().toLowerCase();
+  return base === "text/html";
+}
+async function downloadArxivHtml(arxivId, destPath) {
+  const id = normalizeArxivId(arxivId);
+  const url = `${ARXIV_HTML_BASE}${id}`;
+  try {
+    const response = await fetch(url, {
+      headers: { "User-Agent": USER_AGENT }
+    });
+    if (!response.ok) {
+      return {
+        success: false,
+        error: `HTTP ${response.status} ${response.statusText}`
+      };
+    }
+    const contentType = response.headers.get("content-type");
+    if (!isHtmlContentType(contentType)) {
+      return {
+        success: false,
+        error: `Unexpected Content-Type: ${contentType ?? "none"} (expected text/html)`
+      };
+    }
+    const text = await response.text();
+    await mkdir(dirname(destPath), { recursive: true });
+    await writeFile(destPath, text, "utf-8");
+    return { success: true, size: Buffer.byteLength(text) };
+  } catch (err) {
+    return {
+      success: false,
+      error: err instanceof Error ? err.message : String(err)
+    };
+  }
+}
+export {
+  downloadArxivHtml
+};
+//# sourceMappingURL=arxiv-html.js.map

package/dist/node_modules/@ncukondo/academic-fulltext/dist/download/arxiv-html.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"arxiv-html.js","sources":["../../../../../../node_modules/@ncukondo/academic-fulltext/dist/download/arxiv-html.js"],"sourcesContent":["/**\n * arXiv HTML downloader.\n *\n * Downloads LaTeXML-generated HTML from arXiv for conversion to Markdown.\n * Not all arXiv papers have HTML versions; 404 is handled gracefully.\n */\nimport { mkdir, writeFile } from \"node:fs/promises\";\nimport { dirname } from \"node:path\";\nconst ARXIV_HTML_BASE = \"https://arxiv.org/html/\";\nconst USER_AGENT = \"search-hub/0.8.0 (https://github.com/ncukondo/search-hub)\";\n/** Strip common prefixes from arXiv IDs */\nfunction normalizeArxivId(id) {\n    return id.replace(/^arXiv:/i, \"\");\n}\nfunction isHtmlContentType(contentType) {\n    if (!contentType)\n        return false;\n    const base = (contentType.split(\";\")[0] ?? \"\").trim().toLowerCase();\n    return base === \"text/html\";\n}\n/**\n * Download arXiv HTML for a given arXiv ID.\n *\n * @param arxivId - arXiv identifier (e.g., \"2301.13867\", \"arXiv:2301.13867\")\n * @param destPath - Destination file path for the HTML\n * @returns Result with success status and file size\n */\nexport async function downloadArxivHtml(arxivId, destPath) {\n    const id = normalizeArxivId(arxivId);\n    const url = `${ARXIV_HTML_BASE}${id}`;\n    try {\n        const response = await fetch(url, {\n            headers: { \"User-Agent\": USER_AGENT },\n        });\n        if (!response.ok) {\n            return {\n                success: false,\n                error: `HTTP ${response.status} ${response.statusText}`,\n            };\n        }\n        const contentType = response.headers.get(\"content-type\");\n        if (!isHtmlContentType(contentType)) {\n            return {\n                success: false,\n                error: `Unexpected Content-Type: ${contentType ?? \"none\"} (expected text/html)`,\n            };\n        }\n        const text = await response.text();\n        await mkdir(dirname(destPath), { recursive: true });\n        await writeFile(destPath, text, \"utf-8\");\n        return { success: true, size: Buffer.byteLength(text) };\n    }\n    catch (err) {\n        return {\n            success: false,\n            error: err instanceof Error ? err.message : String(err),\n        };\n    }\n}\n//# sourceMappingURL=arxiv-html.js.map"],"names":[],"mappings":";;AAQA,MAAM,kBAAkB;AACxB,MAAM,aAAa;AAEnB,SAAS,iBAAiB,IAAI;AAC1B,SAAO,GAAG,QAAQ,YAAY,EAAE;AACpC;AACA,SAAS,kBAAkB,aAAa;AACpC,MAAI,CAAC;AACD,WAAO;AACX,QAAM,QAAQ,YAAY,MAAM,GAAG,EAAE,CAAC,KAAK,IAAI,KAAI,EAAG,YAAW;AACjE,SAAO,SAAS;AACpB;AAQO,eAAe,kBAAkB,SAAS,UAAU;AACvD,QAAM,KAAK,iBAAiB,OAAO;AACnC,QAAM,MAAM,GAAG,eAAe,GAAG,EAAE;AACnC,MAAI;AACA,UAAM,WAAW,MAAM,MAAM,KAAK;AAAA,MAC9B,SAAS,EAAE,cAAc,WAAU;AAAA,IAC/C,CAAS;AACD,QAAI,CAAC,SAAS,IAAI;AACd,aAAO;AAAA,QACH,SAAS;AAAA,QACT,OAAO,QAAQ,SAAS,MAAM,IAAI,SAAS,UAAU;AAAA,MACrE;AAAA,IACQ;AACA,UAAM,cAAc,SAAS,QAAQ,IAAI,cAAc;AACvD,QAAI,CAAC,kBAAkB,WAAW,GAAG;AACjC,aAAO;AAAA,QACH,SAAS;AAAA,QACT,OAAO,4BAA4B,eAAe,MAAM;AAAA,MACxE;AAAA,IACQ;AACA,UAAM,OAAO,MAAM,SAAS,KAAI;AAChC,UAAM,MAAM,QAAQ,QAAQ,GAAG,EAAE,WAAW,MAAM;AAClD,UAAM,UAAU,UAAU,MAAM,OAAO;AACvC,WAAO,EAAE,SAAS,MAAM,MAAM,OAAO,WAAW,IAAI,EAAC;AAAA,EACzD,SACO,KAAK;AACR,WAAO;AAAA,MACH,SAAS;AAAA,MACT,OAAO,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAAA,IAClE;AAAA,EACI;AACJ;","x_google_ignoreList":[0]}

package/dist/node_modules/@ncukondo/academic-fulltext/dist/download/downloader.js

@@ -0,0 +1,64 @@
+import { mkdir, writeFile } from "node:fs/promises";
+import { dirname } from "node:path";
+const NON_RETRYABLE_STATUSES = /* @__PURE__ */ new Set([400, 401, 403, 404, 405, 410]);
+const VALID_CONTENT_TYPES = ["application/pdf", "application/octet-stream"];
+const USER_AGENT = "search-hub/0.8.0 (https://github.com/ncukondo/search-hub)";
+function isValidPdfContentType(contentType) {
+  if (!contentType)
+    return false;
+  const base = (contentType.split(";")[0] ?? "").trim().toLowerCase();
+  return VALID_CONTENT_TYPES.includes(base);
+}
+function sleep(ms) {
+  return new Promise((resolve) => setTimeout(resolve, ms));
+}
+async function attemptDownload(url, destPath) {
+  const response = await fetch(url, {
+    headers: { "User-Agent": USER_AGENT }
+  });
+  if (!response.ok) {
+    const error = `HTTP ${response.status} ${response.statusText}`;
+    if (NON_RETRYABLE_STATUSES.has(response.status)) {
+      return { kind: "fail", result: { success: false, error } };
+    }
+    return { kind: "retry", error };
+  }
+  const contentType = response.headers.get("content-type");
+  if (!isValidPdfContentType(contentType)) {
+    return {
+      kind: "fail",
+      result: {
+        success: false,
+        error: `Unexpected Content-Type: ${contentType ?? "none"}`
+      }
+    };
+  }
+  const buffer = await response.arrayBuffer();
+  await mkdir(dirname(destPath), { recursive: true });
+  await writeFile(destPath, Buffer.from(buffer));
+  return { kind: "success", result: { success: true, size: buffer.byteLength } };
+}
+async function downloadPdf(url, destPath, options) {
+  const retries = options?.retries ?? 3;
+  const retryDelay = options?.retryDelay ?? 1e3;
+  let lastError;
+  for (let attempt = 1; attempt <= retries; attempt++) {
+    try {
+      const outcome = await attemptDownload(url, destPath);
+      if (outcome.kind === "success" || outcome.kind === "fail") {
+        return outcome.result;
+      }
+      lastError = outcome.error;
+    } catch (err) {
+      lastError = err instanceof Error ? err.message : String(err);
+    }
+    if (attempt < retries) {
+      await sleep(retryDelay * attempt);
+    }
+  }
+  return { success: false, error: lastError ?? "Download failed" };
+}
+export {
+  downloadPdf
+};
+//# sourceMappingURL=downloader.js.map

package/dist/node_modules/@ncukondo/academic-fulltext/dist/download/downloader.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"downloader.js","sources":["../../../../../../node_modules/@ncukondo/academic-fulltext/dist/download/downloader.js"],"sourcesContent":["/**\n * PDF downloader with retry and error handling.\n */\nimport { mkdir, writeFile } from \"node:fs/promises\";\nimport { dirname } from \"node:path\";\n/** HTTP status codes that should not be retried */\nconst NON_RETRYABLE_STATUSES = new Set([400, 401, 403, 404, 405, 410]);\n/** Content types accepted as valid PDF responses */\nconst VALID_CONTENT_TYPES = [\"application/pdf\", \"application/octet-stream\"];\nconst USER_AGENT = \"search-hub/0.8.0 (https://github.com/ncukondo/search-hub)\";\nfunction isValidPdfContentType(contentType) {\n    if (!contentType)\n        return false;\n    const base = (contentType.split(\";\")[0] ?? \"\").trim().toLowerCase();\n    return VALID_CONTENT_TYPES.includes(base);\n}\nfunction sleep(ms) {\n    return new Promise((resolve) => setTimeout(resolve, ms));\n}\nasync function attemptDownload(url, destPath) {\n    const response = await fetch(url, {\n        headers: { \"User-Agent\": USER_AGENT },\n    });\n    if (!response.ok) {\n        const error = `HTTP ${response.status} ${response.statusText}`;\n        if (NON_RETRYABLE_STATUSES.has(response.status)) {\n            return { kind: \"fail\", result: { success: false, error } };\n        }\n        return { kind: \"retry\", error };\n    }\n    const contentType = response.headers.get(\"content-type\");\n    if (!isValidPdfContentType(contentType)) {\n        return {\n            kind: \"fail\",\n            result: {\n                success: false,\n                error: `Unexpected Content-Type: ${contentType ?? \"none\"}`,\n            },\n        };\n    }\n    const buffer = await response.arrayBuffer();\n    await mkdir(dirname(destPath), { recursive: true });\n    await writeFile(destPath, Buffer.from(buffer));\n    return { kind: \"success\", result: { success: true, size: buffer.byteLength } };\n}\n/**\n * Download a PDF from a URL to a local file path.\n * Retries on network errors and 429 responses with exponential backoff.\n * Does not retry on 403/404 or other client errors.\n */\nexport async function downloadPdf(url, destPath, options) {\n    const retries = options?.retries ?? 3;\n    const retryDelay = options?.retryDelay ?? 1000;\n    let lastError;\n    for (let attempt = 1; attempt <= retries; attempt++) {\n        try {\n            const outcome = await attemptDownload(url, destPath);\n            if (outcome.kind === \"success\" || outcome.kind === \"fail\") {\n                return outcome.result;\n            }\n            lastError = outcome.error;\n        }\n        catch (err) {\n            lastError = err instanceof Error ? err.message : String(err);\n        }\n        if (attempt < retries) {\n            await sleep(retryDelay * attempt);\n        }\n    }\n    return { success: false, error: lastError ?? \"Download failed\" };\n}\n//# sourceMappingURL=downloader.js.map"],"names":[],"mappings":";;AAMA,MAAM,yBAAyB,oBAAI,IAAI,CAAC,KAAK,KAAK,KAAK,KAAK,KAAK,GAAG,CAAC;AAErE,MAAM,sBAAsB,CAAC,mBAAmB,0BAA0B;AAC1E,MAAM,aAAa;AACnB,SAAS,sBAAsB,aAAa;AACxC,MAAI,CAAC;AACD,WAAO;AACX,QAAM,QAAQ,YAAY,MAAM,GAAG,EAAE,CAAC,KAAK,IAAI,KAAI,EAAG,YAAW;AACjE,SAAO,oBAAoB,SAAS,IAAI;AAC5C;AACA,SAAS,MAAM,IAAI;AACf,SAAO,IAAI,QAAQ,CAAC,YAAY,WAAW,SAAS,EAAE,CAAC;AAC3D;AACA,eAAe,gBAAgB,KAAK,UAAU;AAC1C,QAAM,WAAW,MAAM,MAAM,KAAK;AAAA,IAC9B,SAAS,EAAE,cAAc,WAAU;AAAA,EAC3C,CAAK;AACD,MAAI,CAAC,SAAS,IAAI;AACd,UAAM,QAAQ,QAAQ,SAAS,MAAM,IAAI,SAAS,UAAU;AAC5D,QAAI,uBAAuB,IAAI,SAAS,MAAM,GAAG;AAC7C,aAAO,EAAE,MAAM,QAAQ,QAAQ,EAAE,SAAS,OAAO,QAAO;AAAA,IAC5D;AACA,WAAO,EAAE,MAAM,SAAS,MAAK;AAAA,EACjC;AACA,QAAM,cAAc,SAAS,QAAQ,IAAI,cAAc;AACvD,MAAI,CAAC,sBAAsB,WAAW,GAAG;AACrC,WAAO;AAAA,MACH,MAAM;AAAA,MACN,QAAQ;AAAA,QACJ,SAAS;AAAA,QACT,OAAO,4BAA4B,eAAe,MAAM;AAAA,MACxE;AAAA,IACA;AAAA,EACI;AACA,QAAM,SAAS,MAAM,SAAS,YAAW;AACzC,QAAM,MAAM,QAAQ,QAAQ,GAAG,EAAE,WAAW,MAAM;AAClD,QAAM,UAAU,UAAU,OAAO,KAAK,MAAM,CAAC;AAC7C,SAAO,EAAE,MAAM,WAAW,QAAQ,EAAE,SAAS,MAAM,MAAM,OAAO,aAAY;AAChF;AAMO,eAAe,YAAY,KAAK,UAAU,SAAS;AACtD,QAAM,UAAU,SAAS,WAAW;AACpC,QAAM,aAAa,SAAS,cAAc;AAC1C,MAAI;AACJ,WAAS,UAAU,GAAG,WAAW,SAAS,WAAW;AACjD,QAAI;AACA,YAAM,UAAU,MAAM,gBAAgB,KAAK,QAAQ;AACnD,UAAI,QAAQ,SAAS,aAAa,QAAQ,SAAS,QAAQ;AACvD,eAAO,QAAQ;AAAA,MACnB;AACA,kBAAY,QAAQ;AAAA,IACxB,SACO,KAAK;AACR,kBAAY,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAAA,IAC/D;AACA,QAAI,UAAU,SAAS;AACnB,YAAM,MAAM,aAAa,OAAO;AAAA,IACpC;AAAA,EACJ;AACA,SAAO,EAAE,SAAS,OAAO,OAAO,aAAa,kBAAiB;AAClE;","x_google_ignoreList":[0]}