@ncukondo/search-hub 0.12.0 → 0.12.2

package/dist/config/schema.d.ts

@@ -103,6 +103,8 @@ export declare const ConfigSchema: z.ZodObject<{
         sources: z.ZodDefault<z.ZodObject<{
             unpaywall_email: z.ZodDefault<z.ZodString>;
             core_api_key: z.ZodDefault<z.ZodString>;
+            ncbi_email: z.ZodDefault<z.ZodString>;
+            ncbi_tool: z.ZodDefault<z.ZodString>;
             prefer_sources: z.ZodDefault<z.ZodArray<z.ZodString>>;
         }, z.core.$strip>>;
         download: z.ZodDefault<z.ZodObject<{

package/dist/config/schema.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"schema.d.ts","sourceRoot":"","sources":["../../src/config/schema.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAaxB;;;GAGG;AACH,eAAO,MAAM,oBAAoB;;;;;;;;;iBAc/B,CAAC;AAEH,MAAM,MAAM,cAAc,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,oBAAoB,CAAC,CAAC;AASlE;;;GAGG;AACH,eAAO,MAAM,YAAY;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBAwGvB,CAAC;AAEH,MAAM,MAAM,MAAM,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,YAAY,CAAC,CAAC"}
+{"version":3,"file":"schema.d.ts","sourceRoot":"","sources":["../../src/config/schema.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAaxB;;;GAGG;AACH,eAAO,MAAM,oBAAoB;;;;;;;;;iBAc/B,CAAC;AAEH,MAAM,MAAM,cAAc,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,oBAAoB,CAAC,CAAC;AASlE;;;GAGG;AACH,eAAO,MAAM,YAAY;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA8GvB,CAAC;AAEH,MAAM,MAAM,MAAM,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,YAAY,CAAC,CAAC"}

package/dist/config/schema.js

@@ -55,10 +55,14 @@ const ConfigSchema = z.object({
     sources: z.object({
       unpaywall_email: z.string().default(""),
       core_api_key: z.string().default(""),
+      ncbi_email: z.string().default(""),
+      ncbi_tool: z.string().default("search-hub"),
       prefer_sources: z.array(z.string()).default(["pmc", "arxiv", "unpaywall", "core"])
     }).default(() => ({
       unpaywall_email: "",
       core_api_key: "",
+      ncbi_email: "",
+      ncbi_tool: "search-hub",
       prefer_sources: ["pmc", "arxiv", "unpaywall", "core"]
     })),
     download: z.object({
@@ -75,6 +79,8 @@ const ConfigSchema = z.object({
     sources: {
       unpaywall_email: "",
       core_api_key: "",
+      ncbi_email: "",
+      ncbi_tool: "search-hub",
       prefer_sources: ["pmc", "arxiv", "unpaywall", "core"]
     },
     download: {

package/dist/config/schema.js.map

@@ -1 +1 @@
-{"version":3,"file":"schema.js","sources":["../../src/config/schema.ts"],"sourcesContent":["import { z } from 'zod';\n\n/**\n * Default values for provider configuration.\n */\nconst PROVIDER_DEFAULTS = {\n  enabled: true,\n  rate_limit: 3,\n  timeout: 30000,\n  retries: 3,\n  max_results: 10000,\n} as const;\n\n/**\n * Schema for individual provider configuration.\n * Each provider (PubMed, Scopus, etc.) has similar config options.\n */\nexport const ProviderConfigSchema = z.object({\n  enabled: z.boolean().default(PROVIDER_DEFAULTS.enabled),\n  api_key: z.string().default(''),\n  email: z\n    .string()\n    .refine((val) => val === '' || /^[^@\\s]+@[^@\\s]+\\.[^@\\s]+$/.test(val), {\n      message: 'Invalid email',\n    })\n    .default(''),\n  rate_limit: z.number().positive().default(PROVIDER_DEFAULTS.rate_limit),\n  timeout: z.number().positive().default(PROVIDER_DEFAULTS.timeout),\n  retries: z.number().int().min(0).default(PROVIDER_DEFAULTS.retries),\n  max_results: z.number().int().positive().default(PROVIDER_DEFAULTS.max_results),\n  inst_token: z.string().default(''),\n});\n\nexport type ProviderConfig = z.infer<typeof ProviderConfigSchema>;\n\n/**\n * Get default provider config object.\n */\nfunction getDefaultProviderConfig(): ProviderConfig {\n  return ProviderConfigSchema.parse({});\n}\n\n/**\n * Main configuration schema for search-hub.\n * All fields have defaults, so an empty object is valid input.\n */\nexport const ConfigSchema = z.object({\n  session: z\n    .object({\n      // Empty string means use platform default (resolved in loader.ts)\n      directory: z.string().default(''),\n    })\n    .default(() => ({ directory: '' })),\n\n  log: z\n    .object({\n      level: z.enum(['debug', 'info', 'warn', 'error']).default('info'),\n    })\n    .default(() => ({ level: 'info' as const })),\n\n  output: z\n    .object({\n      color: z.boolean().default(true),\n      progress_bar: z.boolean().default(true),\n    })\n    .default(() => ({ color: true, progress_bar: true })),\n\n  providers: z\n    .object({\n      pubmed: ProviderConfigSchema.default(getDefaultProviderConfig),\n      eric: ProviderConfigSchema.default(getDefaultProviderConfig),\n      arxiv: ProviderConfigSchema.default(getDefaultProviderConfig),\n      scopus: ProviderConfigSchema.default(getDefaultProviderConfig),\n      wos: ProviderConfigSchema.default(getDefaultProviderConfig),\n      embase: ProviderConfigSchema.default(getDefaultProviderConfig),\n    })\n    .default(() => ({\n      pubmed: getDefaultProviderConfig(),\n      eric: getDefaultProviderConfig(),\n      arxiv: getDefaultProviderConfig(),\n      scopus: getDefaultProviderConfig(),\n      wos: getDefaultProviderConfig(),\n      embase: getDefaultProviderConfig(),\n    })),\n\n  fulltext: z\n    .object({\n      enabled: z.boolean().default(true),\n      auto_convert_markdown: z.boolean().default(true),\n      auto_attach_on_register: z.boolean().default(true),\n      sources: z\n        .object({\n          unpaywall_email: z.string().default(''),\n          core_api_key: z.string().default(''),\n          prefer_sources: z.array(z.string()).default(['pmc', 'arxiv', 'unpaywall', 'core']),\n        })\n        .default(() => ({\n          unpaywall_email: '',\n          core_api_key: '',\n          prefer_sources: ['pmc', 'arxiv', 'unpaywall', 'core'],\n        })),\n      download: z\n        .object({\n          concurrent_downloads: z.number().int().positive().default(3),\n          retry_attempts: z.number().int().min(0).default(3),\n        })\n        .default(() => ({\n          concurrent_downloads: 3,\n          retry_attempts: 3,\n        })),\n    })\n    .default(() => ({\n      enabled: true,\n      auto_convert_markdown: true,\n      auto_attach_on_register: true,\n      sources: {\n        unpaywall_email: '',\n        core_api_key: '',\n        prefer_sources: ['pmc', 'arxiv', 'unpaywall', 'core'],\n      },\n      download: {\n        concurrent_downloads: 3,\n        retry_attempts: 3,\n      },\n    })),\n\n  integration: z\n    .object({\n      reference_manager: z\n        .object({\n          enabled: z.boolean().default(true),\n          command: z.string().default('ref'),\n          auto_register: z.boolean().default(false),\n          with_abstracts: z.boolean().default(false),\n        })\n        .default(() => ({\n          enabled: true,\n          command: 'ref',\n          auto_register: false,\n          with_abstracts: false,\n        })),\n    })\n    .default(() => ({\n      reference_manager: {\n        enabled: true,\n        command: 'ref',\n        auto_register: false,\n        with_abstracts: false,\n      },\n    })),\n});\n\nexport type Config = z.infer<typeof ConfigSchema>;\n"],"names":[],"mappings":";AAKA,MAAM,oBAAoB;AAAA,EACxB,SAAS;AAAA,EACT,YAAY;AAAA,EACZ,SAAS;AAAA,EACT,SAAS;AAAA,EACT,aAAa;AACf;AAMO,MAAM,uBAAuB,EAAE,OAAO;AAAA,EAC3C,SAAS,EAAE,QAAA,EAAU,QAAQ,kBAAkB,OAAO;AAAA,EACtD,SAAS,EAAE,SAAS,QAAQ,EAAE;AAAA,EAC9B,OAAO,EACJ,OAAA,EACA,OAAO,CAAC,QAAQ,QAAQ,MAAM,6BAA6B,KAAK,GAAG,GAAG;AAAA,IACrE,SAAS;AAAA,EAAA,CACV,EACA,QAAQ,EAAE;AAAA,EACb,YAAY,EAAE,OAAA,EAAS,WAAW,QAAQ,kBAAkB,UAAU;AAAA,EACtE,SAAS,EAAE,OAAA,EAAS,WAAW,QAAQ,kBAAkB,OAAO;AAAA,EAChE,SAAS,EAAE,OAAA,EAAS,IAAA,EAAM,IAAI,CAAC,EAAE,QAAQ,kBAAkB,OAAO;AAAA,EAClE,aAAa,EAAE,OAAA,EAAS,IAAA,EAAM,WAAW,QAAQ,kBAAkB,WAAW;AAAA,EAC9E,YAAY,EAAE,OAAA,EAAS,QAAQ,EAAE;AACnC,CAAC;AAOD,SAAS,2BAA2C;AAClD,SAAO,qBAAqB,MAAM,EAAE;AACtC;AAMO,MAAM,eAAe,EAAE,OAAO;AAAA,EACnC,SAAS,EACN,OAAO;AAAA;AAAA,IAEN,WAAW,EAAE,OAAA,EAAS,QAAQ,EAAE;AAAA,EAAA,CACjC,EACA,QAAQ,OAAO,EAAE,WAAW,KAAK;AAAA,EAEpC,KAAK,EACF,OAAO;AAAA,IACN,OAAO,EAAE,KAAK,CAAC,SAAS,QAAQ,QAAQ,OAAO,CAAC,EAAE,QAAQ,MAAM;AAAA,EAAA,CACjE,EACA,QAAQ,OAAO,EAAE,OAAO,SAAkB;AAAA,EAE7C,QAAQ,EACL,OAAO;AAAA,IACN,OAAO,EAAE,UAAU,QAAQ,IAAI;AAAA,IAC/B,cAAc,EAAE,QAAA,EAAU,QAAQ,IAAI;AAAA,EAAA,CACvC,EACA,QAAQ,OAAO,EAAE,OAAO,MAAM,cAAc,KAAA,EAAO;AAAA,EAEtD,WAAW,EACR,OAAO;AAAA,IACN,QAAQ,qBAAqB,QAAQ,wBAAwB;AAAA,IAC7D,MAAM,qBAAqB,QAAQ,wBAAwB;AAAA,IAC3D,OAAO,qBAAqB,QAAQ,wBAAwB;AAAA,IAC5D,QAAQ,qBAAqB,QAAQ,wBAAwB;AAAA,IAC7D,KAAK,qBAAqB,QAAQ,wBAAwB;AAAA,IAC1D,QAAQ,qBAAqB,QAAQ,wBAAwB;AAAA,EAAA,CAC9D,EACA,QAAQ,OAAO;AAAA,IACd,QAAQ,yBAAA;AAAA,IACR,MAAM,yBAAA;AAAA,IACN,OAAO,yBAAA;AAAA,IACP,QAAQ,yBAAA;AAAA,IACR,KAAK,yBAAA;AAAA,IACL,QAAQ,yBAAA;AAAA,EAAyB,EACjC;AAAA,EAEJ,UAAU,EACP,OAAO;AAAA,IACN,SAAS,EAAE,UAAU,QAAQ,IAAI;AAAA,IACjC,uBAAuB,EAAE,UAAU,QAAQ,IAAI;AAAA,IAC/C,yBAAyB,EAAE,UAAU,QAAQ,IAAI;AAAA,IACjD,SAAS,EACN,OAAO;AAAA,MACN,iBAAiB,EAAE,SAAS,QAAQ,EAAE;AAAA,MACtC,cAAc,EAAE,SAAS,QAAQ,EAAE;AAAA,MACnC,gBAAgB,EAAE,MAAM,EAAE,OAAA,CAAQ,EAAE,QAAQ,CAAC,OAAO,SAAS,aAAa,MAAM,CAAC;AAAA,IAAA,CAClF,EACA,QAAQ,OAAO;AAAA,MACd,iBAAiB;AAAA,MACjB,cAAc;AAAA,MACd,gBAAgB,CAAC,OAAO,SAAS,aAAa,MAAM;AAAA,IAAA,EACpD;AAAA,IACJ,UAAU,EACP,OAAO;AAAA,MACN,sBAAsB,EAAE,SAAS,MAAM,SAAA,EAAW,QAAQ,CAAC;AAAA,MAC3D,gBAAgB,EAAE,SAAS,IAAA,EAAM,IAAI,CAAC,EAAE,QAAQ,CAAC;AAAA,IAAA,CAClD,EACA,QAAQ,OAAO;AAAA,MACd,sBAAsB;AAAA,MACtB,gBAAgB;AAAA,IAAA,EAChB;AAAA,EAAA,CACL,EACA,QAAQ,OAAO;AAAA,IACd,SAAS;AAAA,IACT,uBAAuB;AAAA,IACvB,yBAAyB;AAAA,IACzB,SAAS;AAAA,MACP,iBAAiB;AAAA,MACjB,cAAc;AAAA,MACd,gBAAgB,CAAC,OAAO,SAAS,aAAa,MAAM;AAAA,IAAA;AAAA,IAEtD,UAAU;AAAA,MACR,sBAAsB;AAAA,MACtB,gBAAgB;AAAA,IAAA;AAAA,EAClB,EACA;AAAA,EAEJ,aAAa,EACV,OAAO;AAAA,IACN,mBAAmB,EAChB,OAAO;AAAA,MACN,SAAS,EAAE,UAAU,QAAQ,IAAI;AAAA,MACjC,SAAS,EAAE,SAAS,QAAQ,KAAK;AAAA,MACjC,eAAe,EAAE,UAAU,QAAQ,KAAK;AAAA,MACxC,gBAAgB,EAAE,QAAA,EAAU,QAAQ,KAAK;AAAA,IAAA,CAC1C,EACA,QAAQ,OAAO;AAAA,MACd,SAAS;AAAA,MACT,SAAS;AAAA,MACT,eAAe;AAAA,MACf,gBAAgB;AAAA,IAAA,EAChB;AAAA,EAAA,CACL,EACA,QAAQ,OAAO;AAAA,IACd,mBAAmB;AAAA,MACjB,SAAS;AAAA,MACT,SAAS;AAAA,MACT,eAAe;AAAA,MACf,gBAAgB;AAAA,IAAA;AAAA,EAClB,EACA;AACN,CAAC;"}
+{"version":3,"file":"schema.js","sources":["../../src/config/schema.ts"],"sourcesContent":["import { z } from 'zod';\n\n/**\n * Default values for provider configuration.\n */\nconst PROVIDER_DEFAULTS = {\n  enabled: true,\n  rate_limit: 3,\n  timeout: 30000,\n  retries: 3,\n  max_results: 10000,\n} as const;\n\n/**\n * Schema for individual provider configuration.\n * Each provider (PubMed, Scopus, etc.) has similar config options.\n */\nexport const ProviderConfigSchema = z.object({\n  enabled: z.boolean().default(PROVIDER_DEFAULTS.enabled),\n  api_key: z.string().default(''),\n  email: z\n    .string()\n    .refine((val) => val === '' || /^[^@\\s]+@[^@\\s]+\\.[^@\\s]+$/.test(val), {\n      message: 'Invalid email',\n    })\n    .default(''),\n  rate_limit: z.number().positive().default(PROVIDER_DEFAULTS.rate_limit),\n  timeout: z.number().positive().default(PROVIDER_DEFAULTS.timeout),\n  retries: z.number().int().min(0).default(PROVIDER_DEFAULTS.retries),\n  max_results: z.number().int().positive().default(PROVIDER_DEFAULTS.max_results),\n  inst_token: z.string().default(''),\n});\n\nexport type ProviderConfig = z.infer<typeof ProviderConfigSchema>;\n\n/**\n * Get default provider config object.\n */\nfunction getDefaultProviderConfig(): ProviderConfig {\n  return ProviderConfigSchema.parse({});\n}\n\n/**\n * Main configuration schema for search-hub.\n * All fields have defaults, so an empty object is valid input.\n */\nexport const ConfigSchema = z.object({\n  session: z\n    .object({\n      // Empty string means use platform default (resolved in loader.ts)\n      directory: z.string().default(''),\n    })\n    .default(() => ({ directory: '' })),\n\n  log: z\n    .object({\n      level: z.enum(['debug', 'info', 'warn', 'error']).default('info'),\n    })\n    .default(() => ({ level: 'info' as const })),\n\n  output: z\n    .object({\n      color: z.boolean().default(true),\n      progress_bar: z.boolean().default(true),\n    })\n    .default(() => ({ color: true, progress_bar: true })),\n\n  providers: z\n    .object({\n      pubmed: ProviderConfigSchema.default(getDefaultProviderConfig),\n      eric: ProviderConfigSchema.default(getDefaultProviderConfig),\n      arxiv: ProviderConfigSchema.default(getDefaultProviderConfig),\n      scopus: ProviderConfigSchema.default(getDefaultProviderConfig),\n      wos: ProviderConfigSchema.default(getDefaultProviderConfig),\n      embase: ProviderConfigSchema.default(getDefaultProviderConfig),\n    })\n    .default(() => ({\n      pubmed: getDefaultProviderConfig(),\n      eric: getDefaultProviderConfig(),\n      arxiv: getDefaultProviderConfig(),\n      scopus: getDefaultProviderConfig(),\n      wos: getDefaultProviderConfig(),\n      embase: getDefaultProviderConfig(),\n    })),\n\n  fulltext: z\n    .object({\n      enabled: z.boolean().default(true),\n      auto_convert_markdown: z.boolean().default(true),\n      auto_attach_on_register: z.boolean().default(true),\n      sources: z\n        .object({\n          unpaywall_email: z.string().default(''),\n          core_api_key: z.string().default(''),\n          ncbi_email: z.string().default(''),\n          ncbi_tool: z.string().default('search-hub'),\n          prefer_sources: z.array(z.string()).default(['pmc', 'arxiv', 'unpaywall', 'core']),\n        })\n        .default(() => ({\n          unpaywall_email: '',\n          core_api_key: '',\n          ncbi_email: '',\n          ncbi_tool: 'search-hub',\n          prefer_sources: ['pmc', 'arxiv', 'unpaywall', 'core'],\n        })),\n      download: z\n        .object({\n          concurrent_downloads: z.number().int().positive().default(3),\n          retry_attempts: z.number().int().min(0).default(3),\n        })\n        .default(() => ({\n          concurrent_downloads: 3,\n          retry_attempts: 3,\n        })),\n    })\n    .default(() => ({\n      enabled: true,\n      auto_convert_markdown: true,\n      auto_attach_on_register: true,\n      sources: {\n        unpaywall_email: '',\n        core_api_key: '',\n        ncbi_email: '',\n        ncbi_tool: 'search-hub',\n        prefer_sources: ['pmc', 'arxiv', 'unpaywall', 'core'],\n      },\n      download: {\n        concurrent_downloads: 3,\n        retry_attempts: 3,\n      },\n    })),\n\n  integration: z\n    .object({\n      reference_manager: z\n        .object({\n          enabled: z.boolean().default(true),\n          command: z.string().default('ref'),\n          auto_register: z.boolean().default(false),\n          with_abstracts: z.boolean().default(false),\n        })\n        .default(() => ({\n          enabled: true,\n          command: 'ref',\n          auto_register: false,\n          with_abstracts: false,\n        })),\n    })\n    .default(() => ({\n      reference_manager: {\n        enabled: true,\n        command: 'ref',\n        auto_register: false,\n        with_abstracts: false,\n      },\n    })),\n});\n\nexport type Config = z.infer<typeof ConfigSchema>;\n"],"names":[],"mappings":";AAKA,MAAM,oBAAoB;AAAA,EACxB,SAAS;AAAA,EACT,YAAY;AAAA,EACZ,SAAS;AAAA,EACT,SAAS;AAAA,EACT,aAAa;AACf;AAMO,MAAM,uBAAuB,EAAE,OAAO;AAAA,EAC3C,SAAS,EAAE,QAAA,EAAU,QAAQ,kBAAkB,OAAO;AAAA,EACtD,SAAS,EAAE,SAAS,QAAQ,EAAE;AAAA,EAC9B,OAAO,EACJ,OAAA,EACA,OAAO,CAAC,QAAQ,QAAQ,MAAM,6BAA6B,KAAK,GAAG,GAAG;AAAA,IACrE,SAAS;AAAA,EAAA,CACV,EACA,QAAQ,EAAE;AAAA,EACb,YAAY,EAAE,OAAA,EAAS,WAAW,QAAQ,kBAAkB,UAAU;AAAA,EACtE,SAAS,EAAE,OAAA,EAAS,WAAW,QAAQ,kBAAkB,OAAO;AAAA,EAChE,SAAS,EAAE,OAAA,EAAS,IAAA,EAAM,IAAI,CAAC,EAAE,QAAQ,kBAAkB,OAAO;AAAA,EAClE,aAAa,EAAE,OAAA,EAAS,IAAA,EAAM,WAAW,QAAQ,kBAAkB,WAAW;AAAA,EAC9E,YAAY,EAAE,OAAA,EAAS,QAAQ,EAAE;AACnC,CAAC;AAOD,SAAS,2BAA2C;AAClD,SAAO,qBAAqB,MAAM,EAAE;AACtC;AAMO,MAAM,eAAe,EAAE,OAAO;AAAA,EACnC,SAAS,EACN,OAAO;AAAA;AAAA,IAEN,WAAW,EAAE,OAAA,EAAS,QAAQ,EAAE;AAAA,EAAA,CACjC,EACA,QAAQ,OAAO,EAAE,WAAW,KAAK;AAAA,EAEpC,KAAK,EACF,OAAO;AAAA,IACN,OAAO,EAAE,KAAK,CAAC,SAAS,QAAQ,QAAQ,OAAO,CAAC,EAAE,QAAQ,MAAM;AAAA,EAAA,CACjE,EACA,QAAQ,OAAO,EAAE,OAAO,SAAkB;AAAA,EAE7C,QAAQ,EACL,OAAO;AAAA,IACN,OAAO,EAAE,UAAU,QAAQ,IAAI;AAAA,IAC/B,cAAc,EAAE,QAAA,EAAU,QAAQ,IAAI;AAAA,EAAA,CACvC,EACA,QAAQ,OAAO,EAAE,OAAO,MAAM,cAAc,KAAA,EAAO;AAAA,EAEtD,WAAW,EACR,OAAO;AAAA,IACN,QAAQ,qBAAqB,QAAQ,wBAAwB;AAAA,IAC7D,MAAM,qBAAqB,QAAQ,wBAAwB;AAAA,IAC3D,OAAO,qBAAqB,QAAQ,wBAAwB;AAAA,IAC5D,QAAQ,qBAAqB,QAAQ,wBAAwB;AAAA,IAC7D,KAAK,qBAAqB,QAAQ,wBAAwB;AAAA,IAC1D,QAAQ,qBAAqB,QAAQ,wBAAwB;AAAA,EAAA,CAC9D,EACA,QAAQ,OAAO;AAAA,IACd,QAAQ,yBAAA;AAAA,IACR,MAAM,yBAAA;AAAA,IACN,OAAO,yBAAA;AAAA,IACP,QAAQ,yBAAA;AAAA,IACR,KAAK,yBAAA;AAAA,IACL,QAAQ,yBAAA;AAAA,EAAyB,EACjC;AAAA,EAEJ,UAAU,EACP,OAAO;AAAA,IACN,SAAS,EAAE,UAAU,QAAQ,IAAI;AAAA,IACjC,uBAAuB,EAAE,UAAU,QAAQ,IAAI;AAAA,IAC/C,yBAAyB,EAAE,UAAU,QAAQ,IAAI;AAAA,IACjD,SAAS,EACN,OAAO;AAAA,MACN,iBAAiB,EAAE,SAAS,QAAQ,EAAE;AAAA,MACtC,cAAc,EAAE,SAAS,QAAQ,EAAE;AAAA,MACnC,YAAY,EAAE,SAAS,QAAQ,EAAE;AAAA,MACjC,WAAW,EAAE,SAAS,QAAQ,YAAY;AAAA,MAC1C,gBAAgB,EAAE,MAAM,EAAE,OAAA,CAAQ,EAAE,QAAQ,CAAC,OAAO,SAAS,aAAa,MAAM,CAAC;AAAA,IAAA,CAClF,EACA,QAAQ,OAAO;AAAA,MACd,iBAAiB;AAAA,MACjB,cAAc;AAAA,MACd,YAAY;AAAA,MACZ,WAAW;AAAA,MACX,gBAAgB,CAAC,OAAO,SAAS,aAAa,MAAM;AAAA,IAAA,EACpD;AAAA,IACJ,UAAU,EACP,OAAO;AAAA,MACN,sBAAsB,EAAE,SAAS,MAAM,SAAA,EAAW,QAAQ,CAAC;AAAA,MAC3D,gBAAgB,EAAE,SAAS,IAAA,EAAM,IAAI,CAAC,EAAE,QAAQ,CAAC;AAAA,IAAA,CAClD,EACA,QAAQ,OAAO;AAAA,MACd,sBAAsB;AAAA,MACtB,gBAAgB;AAAA,IAAA,EAChB;AAAA,EAAA,CACL,EACA,QAAQ,OAAO;AAAA,IACd,SAAS;AAAA,IACT,uBAAuB;AAAA,IACvB,yBAAyB;AAAA,IACzB,SAAS;AAAA,MACP,iBAAiB;AAAA,MACjB,cAAc;AAAA,MACd,YAAY;AAAA,MACZ,WAAW;AAAA,MACX,gBAAgB,CAAC,OAAO,SAAS,aAAa,MAAM;AAAA,IAAA;AAAA,IAEtD,UAAU;AAAA,MACR,sBAAsB;AAAA,MACtB,gBAAgB;AAAA,IAAA;AAAA,EAClB,EACA;AAAA,EAEJ,aAAa,EACV,OAAO;AAAA,IACN,mBAAmB,EAChB,OAAO;AAAA,MACN,SAAS,EAAE,UAAU,QAAQ,IAAI;AAAA,MACjC,SAAS,EAAE,SAAS,QAAQ,KAAK;AAAA,MACjC,eAAe,EAAE,UAAU,QAAQ,KAAK;AAAA,MACxC,gBAAgB,EAAE,QAAA,EAAU,QAAQ,KAAK;AAAA,IAAA,CAC1C,EACA,QAAQ,OAAO;AAAA,MACd,SAAS;AAAA,MACT,SAAS;AAAA,MACT,eAAe;AAAA,MACf,gBAAgB;AAAA,IAAA,EAChB;AAAA,EAAA,CACL,EACA,QAAQ,OAAO;AAAA,IACd,mBAAmB;AAAA,MACjB,SAAS;AAAA,MACT,SAAS;AAAA,MACT,eAAe;AAAA,MACf,gBAAgB;AAAA,IAAA;AAAA,EAClB,EACA;AACN,CAAC;"}

package/dist/{fulltext → integration}/attach-shared.d.ts

@@ -1,5 +1,5 @@
-import { FulltextMeta } from './types.js';
-import { FulltextAttachResult } from '../integration/types.js';
+import { FulltextMeta } from '@ncukondo/academic-fulltext';
+import { FulltextAttachResult } from './types.js';
 /** Files we try to attach, in priority order. */
 export declare const ATTACHABLE_FILES: readonly ["fulltext.pdf", "fulltext.md"];
 /**

package/dist/integration/attach-shared.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"attach-shared.d.ts","sourceRoot":"","sources":["../../src/integration/attach-shared.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAKH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,6BAA6B,CAAC;AAChE,OAAO,KAAK,EAAE,oBAAoB,EAAE,MAAM,YAAY,CAAC;AAEvD,iDAAiD;AACjD,eAAO,MAAM,gBAAgB,0CAA2C,CAAC;AAEzE;;;GAGG;AACH,wBAAgB,SAAS,CAAC,IAAI,EAAE,YAAY,EAAE,SAAS,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,GAAG,MAAM,GAAG,SAAS,CAUhG;AAED,iEAAiE;AACjE,MAAM,WAAW,YAAY;IAC3B,OAAO,EAAE,MAAM,CAAC;IAChB,UAAU,EAAE,MAAM,CAAC;IACnB,IAAI,EAAE,YAAY,GAAG,IAAI,CAAC;CAC3B;AAED;;;GAGG;AACH,wBAAsB,mBAAmB,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,YAAY,EAAE,CAAC,CAwBtF;AAED;;GAEG;AACH,wBAAsB,sBAAsB,CAC1C,UAAU,EAAE,MAAM,EAClB,IAAI,EAAE,YAAY,GACjB,OAAO,CAAC,MAAM,EAAE,CAAC,CAcnB;AAED,0CAA0C;AAC1C,MAAM,WAAW,6BAA6B;IAC5C,2DAA2D;IAC3D,WAAW,EAAE,MAAM,CAAC;IACpB,6CAA6C;IAC7C,SAAS,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IAC/B,sDAAsD;IACtD,UAAU,EAAE,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,KAAK,OAAO,CAAC,IAAI,CAAC,CAAC;IAC/D,+EAA+E;IAC/E,MAAM,CAAC,EAAE,OAAO,CAAC;IACjB,wBAAwB;IACxB,UAAU,CAAC,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,KAAK,IAAI,CAAC;CACvD;AAED;;;GAGG;AACH,wBAAsB,sBAAsB,CAC1C,OAAO,EAAE,6BAA6B,GACrC,OAAO,CAAC,oBAAoB,CAAC,CAsE/B"}

package/dist/integration/attach-shared.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"attach-shared.js","sources":["../../src/integration/attach-shared.ts"],"sourcesContent":["/**\n * Shared fulltext attach utilities.\n * Used by both the integration attach (register flow) and standalone attach command.\n */\n\nimport { readdir, readFile, access } from 'node:fs/promises';\nimport { constants } from 'node:fs';\nimport { join } from 'node:path';\nimport type { FulltextMeta } from '@ncukondo/academic-fulltext';\nimport type { FulltextAttachResult } from './types.js';\n\n/** Files we try to attach, in priority order. */\nexport const ATTACHABLE_FILES = ['fulltext.pdf', 'fulltext.md'] as const;\n\n/**\n * Find the ref ID matching a fulltext meta entry.\n * Tries DOI first, then PMID.\n */\nexport function findRefId(meta: FulltextMeta, refLookup: Map<string, string>): string | undefined {\n  if (meta.doi) {\n    const byDoi = refLookup.get(`doi:${meta.doi}`);\n    if (byDoi) return byDoi;\n  }\n  if (meta.pmid) {\n    const byPmid = refLookup.get(`pmid:${meta.pmid}`);\n    if (byPmid) return byPmid;\n  }\n  return undefined;\n}\n\n/** A single article directory entry with its loaded metadata. */\nexport interface ArticleEntry {\n  dirName: string;\n  articleDir: string;\n  meta: FulltextMeta | null;\n}\n\n/**\n * Scan a fulltext directory and load metadata for all article subdirectories.\n * Returns an empty array if the directory doesn't exist.\n */\nexport async function loadFulltextEntries(fulltextDir: string): Promise<ArticleEntry[]> {\n  let dirNames: string[];\n  try {\n    const dirEntries = await readdir(fulltextDir, { withFileTypes: true });\n    dirNames = dirEntries\n      .filter((e) => e.isDirectory())\n      .map((e) => e.name);\n  } catch {\n    return [];\n  }\n\n  const entries: ArticleEntry[] = [];\n  for (const dirName of dirNames) {\n    const articleDir = join(fulltextDir, dirName);\n    try {\n      const raw = await readFile(join(articleDir, 'meta.json'), 'utf-8');\n      const meta = JSON.parse(raw) as FulltextMeta;\n      entries.push({ dirName, articleDir, meta });\n    } catch {\n      // Will be handled as a failed entry by the caller\n      entries.push({ dirName, articleDir, meta: null });\n    }\n  }\n  return entries;\n}\n\n/**\n * Determine which attachable files exist for an article, verifying on disk.\n */\nexport async function resolveAttachableFiles(\n  articleDir: string,\n  meta: FulltextMeta,\n): Promise<string[]> {\n  const filesToAttach: string[] = [];\n  for (const filename of ATTACHABLE_FILES) {\n    const fileKey = filename === 'fulltext.pdf' ? 'pdf' : 'markdown';\n    if (meta.files[fileKey]) {\n      try {\n        await access(join(articleDir, filename), constants.R_OK);\n        filesToAttach.push(filename);\n      } catch {\n        // File recorded in meta but not on disk — skip\n      }\n    }\n  }\n  return filesToAttach;\n}\n\n/** Options for processFulltextEntries. */\nexport interface ProcessFulltextEntriesOptions {\n  /** Path to the fulltext directory (sessionDir/fulltext) */\n  fulltextDir: string;\n  /** Lookup map from identifiers to ref IDs */\n  refLookup: Map<string, string>;\n  /** Function to attach a single file to a ref entry */\n  attachFile: (refId: string, filePath: string) => Promise<void>;\n  /** If true, skip actual attach calls and just record what would be attached */\n  dryRun?: boolean;\n  /** Progress callback */\n  onProgress?: (current: number, total: number) => void;\n}\n\n/**\n * Shared attach loop: scan fulltext entries, match to refs, and attach files.\n * Both the standalone and integrated attach commands delegate to this function.\n */\nexport async function processFulltextEntries(\n  options: ProcessFulltextEntriesOptions,\n): Promise<FulltextAttachResult> {\n  const { fulltextDir, refLookup, attachFile, dryRun = false, onProgress } = options;\n\n  const result: FulltextAttachResult = {\n    summary: { total: 0, attached: 0, skipped: 0, failed: 0 },\n    attached: [],\n    skipped: [],\n    failed: [],\n  };\n\n  const entries = await loadFulltextEntries(fulltextDir);\n  result.summary.total = entries.length;\n\n  for (let i = 0; i < entries.length; i++) {\n    const entry = entries[i]!;\n    const { dirName, articleDir, meta } = entry;\n\n    if (onProgress) {\n      onProgress(i + 1, entries.length);\n    }\n\n    if (!meta) {\n      result.summary.failed++;\n      result.failed.push({\n        dirName,\n        reason: 'meta_read_error',\n        error: 'Failed to read meta.json',\n      });\n      continue;\n    }\n\n    const refId = findRefId(meta, refLookup);\n    if (!refId) {\n      result.summary.skipped++;\n      result.skipped.push({ dirName, reason: 'not_in_ref' });\n      continue;\n    }\n\n    const filesToAttach = await resolveAttachableFiles(articleDir, meta);\n\n    if (filesToAttach.length === 0) {\n      result.summary.skipped++;\n      result.skipped.push({ dirName, reason: 'no_files' });\n      continue;\n    }\n\n    if (dryRun) {\n      result.summary.attached++;\n      result.attached.push({ refId, files: filesToAttach });\n      continue;\n    }\n\n    try {\n      for (const filename of filesToAttach) {\n        const filePath = join(articleDir, filename);\n        await attachFile(refId, filePath);\n      }\n      result.summary.attached++;\n      result.attached.push({ refId, files: filesToAttach });\n    } catch (error) {\n      result.summary.failed++;\n      result.failed.push({\n        dirName,\n        reason: 'attach_error',\n        error: error instanceof Error ? error.message : 'Unknown error',\n      });\n    }\n  }\n\n  return result;\n}\n"],"names":[],"mappings":";;;AAYO,MAAM,mBAAmB,CAAC,gBAAgB,aAAa;AAMvD,SAAS,UAAU,MAAoB,WAAoD;AAChG,MAAI,KAAK,KAAK;AACZ,UAAM,QAAQ,UAAU,IAAI,OAAO,KAAK,GAAG,EAAE;AAC7C,QAAI,MAAO,QAAO;AAAA,EACpB;AACA,MAAI,KAAK,MAAM;AACb,UAAM,SAAS,UAAU,IAAI,QAAQ,KAAK,IAAI,EAAE;AAChD,QAAI,OAAQ,QAAO;AAAA,EACrB;AACA,SAAO;AACT;AAaA,eAAsB,oBAAoB,aAA8C;AACtF,MAAI;AACJ,MAAI;AACF,UAAM,aAAa,MAAM,QAAQ,aAAa,EAAE,eAAe,MAAM;AACrE,eAAW,WACR,OAAO,CAAC,MAAM,EAAE,YAAA,CAAa,EAC7B,IAAI,CAAC,MAAM,EAAE,IAAI;AAAA,EACtB,QAAQ;AACN,WAAO,CAAA;AAAA,EACT;AAEA,QAAM,UAA0B,CAAA;AAChC,aAAW,WAAW,UAAU;AAC9B,UAAM,aAAa,KAAK,aAAa,OAAO;AAC5C,QAAI;AACF,YAAM,MAAM,MAAM,SAAS,KAAK,YAAY,WAAW,GAAG,OAAO;AACjE,YAAM,OAAO,KAAK,MAAM,GAAG;AAC3B,cAAQ,KAAK,EAAE,SAAS,YAAY,MAAM;AAAA,IAC5C,QAAQ;AAEN,cAAQ,KAAK,EAAE,SAAS,YAAY,MAAM,MAAM;AAAA,IAClD;AAAA,EACF;AACA,SAAO;AACT;AAKA,eAAsB,uBACpB,YACA,MACmB;AACnB,QAAM,gBAA0B,CAAA;AAChC,aAAW,YAAY,kBAAkB;AACvC,UAAM,UAAU,aAAa,iBAAiB,QAAQ;AACtD,QAAI,KAAK,MAAM,OAAO,GAAG;AACvB,UAAI;AACF,cAAM,OAAO,KAAK,YAAY,QAAQ,GAAG,UAAU,IAAI;AACvD,sBAAc,KAAK,QAAQ;AAAA,MAC7B,QAAQ;AAAA,MAER;AAAA,IACF;AAAA,EACF;AACA,SAAO;AACT;AAoBA,eAAsB,uBACpB,SAC+B;AAC/B,QAAM,EAAE,aAAa,WAAW,YAAY,SAAS,OAAO,eAAe;AAE3E,QAAM,SAA+B;AAAA,IACnC,SAAS,EAAE,OAAO,GAAG,UAAU,GAAG,SAAS,GAAG,QAAQ,EAAA;AAAA,IACtD,UAAU,CAAA;AAAA,IACV,SAAS,CAAA;AAAA,IACT,QAAQ,CAAA;AAAA,EAAC;AAGX,QAAM,UAAU,MAAM,oBAAoB,WAAW;AACrD,SAAO,QAAQ,QAAQ,QAAQ;AAE/B,WAAS,IAAI,GAAG,IAAI,QAAQ,QAAQ,KAAK;AACvC,UAAM,QAAQ,QAAQ,CAAC;AACvB,UAAM,EAAE,SAAS,YAAY,KAAA,IAAS;AAEtC,QAAI,YAAY;AACd,iBAAW,IAAI,GAAG,QAAQ,MAAM;AAAA,IAClC;AAEA,QAAI,CAAC,MAAM;AACT,aAAO,QAAQ;AACf,aAAO,OAAO,KAAK;AAAA,QACjB;AAAA,QACA,QAAQ;AAAA,QACR,OAAO;AAAA,MAAA,CACR;AACD;AAAA,IACF;AAEA,UAAM,QAAQ,UAAU,MAAM,SAAS;AACvC,QAAI,CAAC,OAAO;AACV,aAAO,QAAQ;AACf,aAAO,QAAQ,KAAK,EAAE,SAAS,QAAQ,cAAc;AACrD;AAAA,IACF;AAEA,UAAM,gBAAgB,MAAM,uBAAuB,YAAY,IAAI;AAEnE,QAAI,cAAc,WAAW,GAAG;AAC9B,aAAO,QAAQ;AACf,aAAO,QAAQ,KAAK,EAAE,SAAS,QAAQ,YAAY;AACnD;AAAA,IACF;AAEA,QAAI,QAAQ;AACV,aAAO,QAAQ;AACf,aAAO,SAAS,KAAK,EAAE,OAAO,OAAO,eAAe;AACpD;AAAA,IACF;AAEA,QAAI;AACF,iBAAW,YAAY,eAAe;AACpC,cAAM,WAAW,KAAK,YAAY,QAAQ;AAC1C,cAAM,WAAW,OAAO,QAAQ;AAAA,MAClC;AACA,aAAO,QAAQ;AACf,aAAO,SAAS,KAAK,EAAE,OAAO,OAAO,eAAe;AAAA,IACtD,SAAS,OAAO;AACd,aAAO,QAAQ;AACf,aAAO,OAAO,KAAK;AAAA,QACjB;AAAA,QACA,QAAQ;AAAA,QACR,OAAO,iBAAiB,QAAQ,MAAM,UAAU;AAAA,MAAA,CACjD;AAAA,IACH;AAAA,EACF;AAEA,SAAO;AACT;"}

package/dist/integration/fulltext-attach.js

@@ -1,6 +1,6 @@
 import { join } from "node:path";
 import { refFulltextAttach } from "./ref-cli.js";
-import { processFulltextEntries } from "../fulltext/attach-shared.js";
+import { processFulltextEntries } from "./attach-shared.js";
 function extractDoi(source) {
   if (source.startsWith("10.")) return source;
   if (source.startsWith("doi:")) return source.slice(4);

package/dist/integration/fulltext-attach.js.map

@@ -1 +1 @@
-{"version":3,"file":"fulltext-attach.js","sources":["../../src/integration/fulltext-attach.ts"],"sourcesContent":["/**\n * Fulltext attach logic for reference-manager integration.\n * Attaches fulltext files (PDF, Markdown) from session fulltext directories\n * to corresponding entries in the reference-manager library.\n */\n\nimport { join } from 'node:path';\nimport type { FulltextAttachResult } from './types.js';\nimport { refFulltextAttach, type RefCliOptions } from './ref-cli.js';\nimport { processFulltextEntries } from '../fulltext/attach-shared.js';\n\n/**\n * Options for attachFulltexts function.\n */\nexport interface AttachFulltextsOptions {\n  /** Path to the session directory */\n  sessionDir: string;\n  /** Path to the reference library file */\n  libraryPath: string;\n  /** List of refs that were added/exist in the library, with source identifiers */\n  addedRefs: Array<{ id: string; source: string }>;\n  /** Progress callback */\n  onProgress?: (current: number, total: number) => void;\n}\n\n/**\n * Extract DOI from a source string.\n * Sources look like \"10.1234/test\" (bare DOI) or \"doi:10.1234/test\".\n */\nfunction extractDoi(source: string): string | undefined {\n  if (source.startsWith('10.')) return source;\n  if (source.startsWith('doi:')) return source.slice(4);\n  return undefined;\n}\n\n/**\n * Extract PMID from a source string.\n * Sources look like \"pmid:12345678\".\n */\nfunction extractPmid(source: string): string | undefined {\n  if (source.startsWith('pmid:')) return source.slice(5);\n  return undefined;\n}\n\n/**\n * Build a lookup map from identifiers to ref IDs.\n * Keys are normalized: \"doi:10.1234/test\" and \"pmid:12345678\".\n */\nfunction buildRefLookup(addedRefs: Array<{ id: string; source: string }>): Map<string, string> {\n  const lookup = new Map<string, string>();\n  for (const ref of addedRefs) {\n    const doi = extractDoi(ref.source);\n    if (doi) {\n      lookup.set(`doi:${doi}`, ref.id);\n    }\n    const pmid = extractPmid(ref.source);\n    if (pmid) {\n      lookup.set(`pmid:${pmid}`, ref.id);\n    }\n  }\n  return lookup;\n}\n\n/**\n * Attach fulltext files from session fulltext directories to reference-manager entries.\n *\n * For each article directory in sessionDir/fulltext/:\n * 1. Load meta.json to get identifiers\n * 2. Match to a ref entry by DOI or PMID\n * 3. Attach available fulltext files (PDF, Markdown)\n * 4. Record results\n */\nexport async function attachFulltexts(\n  options: AttachFulltextsOptions,\n): Promise<FulltextAttachResult> {\n  const { sessionDir, libraryPath, addedRefs, onProgress } = options;\n  const fulltextDir = join(sessionDir, 'fulltext');\n  const refCliOptions: RefCliOptions = { libraryPath };\n  const refLookup = buildRefLookup(addedRefs);\n\n  return processFulltextEntries({\n    fulltextDir,\n    refLookup,\n    attachFile: (refId, filePath) => refFulltextAttach(refId, filePath, refCliOptions),\n    ...(onProgress ? { onProgress } : {}),\n  });\n}\n"],"names":[],"mappings":";;;AA6BA,SAAS,WAAW,QAAoC;AACtD,MAAI,OAAO,WAAW,KAAK,EAAG,QAAO;AACrC,MAAI,OAAO,WAAW,MAAM,EAAG,QAAO,OAAO,MAAM,CAAC;AACpD,SAAO;AACT;AAMA,SAAS,YAAY,QAAoC;AACvD,MAAI,OAAO,WAAW,OAAO,EAAG,QAAO,OAAO,MAAM,CAAC;AACrD,SAAO;AACT;AAMA,SAAS,eAAe,WAAuE;AAC7F,QAAM,6BAAa,IAAA;AACnB,aAAW,OAAO,WAAW;AAC3B,UAAM,MAAM,WAAW,IAAI,MAAM;AACjC,QAAI,KAAK;AACP,aAAO,IAAI,OAAO,GAAG,IAAI,IAAI,EAAE;AAAA,IACjC;AACA,UAAM,OAAO,YAAY,IAAI,MAAM;AACnC,QAAI,MAAM;AACR,aAAO,IAAI,QAAQ,IAAI,IAAI,IAAI,EAAE;AAAA,IACnC;AAAA,EACF;AACA,SAAO;AACT;AAWA,eAAsB,gBACpB,SAC+B;AAC/B,QAAM,EAAE,YAAY,aAAa,WAAW,eAAe;AAC3D,QAAM,cAAc,KAAK,YAAY,UAAU;AAC/C,QAAM,gBAA+B,EAAE,YAAA;AACvC,QAAM,YAAY,eAAe,SAAS;AAE1C,SAAO,uBAAuB;AAAA,IAC5B;AAAA,IACA;AAAA,IACA,YAAY,CAAC,OAAO,aAAa,kBAAkB,OAAO,UAAU,aAAa;AAAA,IACjF,GAAI,aAAa,EAAE,eAAe,CAAA;AAAA,EAAC,CACpC;AACH;"}
+{"version":3,"file":"fulltext-attach.js","sources":["../../src/integration/fulltext-attach.ts"],"sourcesContent":["/**\n * Fulltext attach logic for reference-manager integration.\n * Attaches fulltext files (PDF, Markdown) from session fulltext directories\n * to corresponding entries in the reference-manager library.\n */\n\nimport { join } from 'node:path';\nimport type { FulltextAttachResult } from './types.js';\nimport { refFulltextAttach, type RefCliOptions } from './ref-cli.js';\nimport { processFulltextEntries } from './attach-shared.js';\n\n/**\n * Options for attachFulltexts function.\n */\nexport interface AttachFulltextsOptions {\n  /** Path to the session directory */\n  sessionDir: string;\n  /** Path to the reference library file */\n  libraryPath: string;\n  /** List of refs that were added/exist in the library, with source identifiers */\n  addedRefs: Array<{ id: string; source: string }>;\n  /** Progress callback */\n  onProgress?: (current: number, total: number) => void;\n}\n\n/**\n * Extract DOI from a source string.\n * Sources look like \"10.1234/test\" (bare DOI) or \"doi:10.1234/test\".\n */\nfunction extractDoi(source: string): string | undefined {\n  if (source.startsWith('10.')) return source;\n  if (source.startsWith('doi:')) return source.slice(4);\n  return undefined;\n}\n\n/**\n * Extract PMID from a source string.\n * Sources look like \"pmid:12345678\".\n */\nfunction extractPmid(source: string): string | undefined {\n  if (source.startsWith('pmid:')) return source.slice(5);\n  return undefined;\n}\n\n/**\n * Build a lookup map from identifiers to ref IDs.\n * Keys are normalized: \"doi:10.1234/test\" and \"pmid:12345678\".\n */\nfunction buildRefLookup(addedRefs: Array<{ id: string; source: string }>): Map<string, string> {\n  const lookup = new Map<string, string>();\n  for (const ref of addedRefs) {\n    const doi = extractDoi(ref.source);\n    if (doi) {\n      lookup.set(`doi:${doi}`, ref.id);\n    }\n    const pmid = extractPmid(ref.source);\n    if (pmid) {\n      lookup.set(`pmid:${pmid}`, ref.id);\n    }\n  }\n  return lookup;\n}\n\n/**\n * Attach fulltext files from session fulltext directories to reference-manager entries.\n *\n * For each article directory in sessionDir/fulltext/:\n * 1. Load meta.json to get identifiers\n * 2. Match to a ref entry by DOI or PMID\n * 3. Attach available fulltext files (PDF, Markdown)\n * 4. Record results\n */\nexport async function attachFulltexts(\n  options: AttachFulltextsOptions,\n): Promise<FulltextAttachResult> {\n  const { sessionDir, libraryPath, addedRefs, onProgress } = options;\n  const fulltextDir = join(sessionDir, 'fulltext');\n  const refCliOptions: RefCliOptions = { libraryPath };\n  const refLookup = buildRefLookup(addedRefs);\n\n  return processFulltextEntries({\n    fulltextDir,\n    refLookup,\n    attachFile: (refId, filePath) => refFulltextAttach(refId, filePath, refCliOptions),\n    ...(onProgress ? { onProgress } : {}),\n  });\n}\n"],"names":[],"mappings":";;;AA6BA,SAAS,WAAW,QAAoC;AACtD,MAAI,OAAO,WAAW,KAAK,EAAG,QAAO;AACrC,MAAI,OAAO,WAAW,MAAM,EAAG,QAAO,OAAO,MAAM,CAAC;AACpD,SAAO;AACT;AAMA,SAAS,YAAY,QAAoC;AACvD,MAAI,OAAO,WAAW,OAAO,EAAG,QAAO,OAAO,MAAM,CAAC;AACrD,SAAO;AACT;AAMA,SAAS,eAAe,WAAuE;AAC7F,QAAM,6BAAa,IAAA;AACnB,aAAW,OAAO,WAAW;AAC3B,UAAM,MAAM,WAAW,IAAI,MAAM;AACjC,QAAI,KAAK;AACP,aAAO,IAAI,OAAO,GAAG,IAAI,IAAI,EAAE;AAAA,IACjC;AACA,UAAM,OAAO,YAAY,IAAI,MAAM;AACnC,QAAI,MAAM;AACR,aAAO,IAAI,QAAQ,IAAI,IAAI,IAAI,EAAE;AAAA,IACnC;AAAA,EACF;AACA,SAAO;AACT;AAWA,eAAsB,gBACpB,SAC+B;AAC/B,QAAM,EAAE,YAAY,aAAa,WAAW,eAAe;AAC3D,QAAM,cAAc,KAAK,YAAY,UAAU;AAC/C,QAAM,gBAA+B,EAAE,YAAA;AACvC,QAAM,YAAY,eAAe,SAAS;AAE1C,SAAO,uBAAuB;AAAA,IAC5B;AAAA,IACA;AAAA,IACA,YAAY,CAAC,OAAO,aAAa,kBAAkB,OAAO,UAAU,aAAa;AAAA,IACjF,GAAI,aAAa,EAAE,eAAe,CAAA;AAAA,EAAC,CACpC;AACH;"}

package/dist/{fulltext → node_modules/@ncukondo/academic-fulltext/dist}/citation-key.js

@@ -1,5 +1,5 @@
-import anyAscii from "../node_modules/any-ascii/any-ascii.js";
 import { randomUUID } from "node:crypto";
+import anyAscii from "../../../any-ascii/any-ascii.js";
 function collisionSuffix(index) {
   let result = "";
   let n = index;

package/dist/node_modules/@ncukondo/academic-fulltext/dist/citation-key.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"citation-key.js","sources":["../../../../../node_modules/@ncukondo/academic-fulltext/dist/citation-key.js"],"sourcesContent":["/**\n * Citation key generation for fulltext directories.\n */\nimport { randomUUID } from \"node:crypto\";\nimport anyAscii from \"any-ascii\";\n/**\n * Generate a collision suffix: a, b, ..., z, aa, ab, ...\n */\nfunction collisionSuffix(index) {\n    let result = \"\";\n    let n = index;\n    do {\n        result = String.fromCodePoint(97 + (n % 26)) + result;\n        n = Math.floor(n / 26) - 1;\n    } while (n >= 0);\n    return result;\n}\n/** CJK Unified Ideographs range: U+4E00–U+9FFF */\nconst CJK_REGEX = /[\\u4e00-\\u9fff]/;\n/**\n * Extract the family name portion from an author string.\n * Handles formats like \"Smith, J.\" → \"Smith\", \"Smith\" → \"Smith\".\n */\nfunction extractFamilyName(author) {\n    const commaIndex = author.indexOf(\",\");\n    if (commaIndex >= 0) {\n        return author.slice(0, commaIndex).trim();\n    }\n    return author.trim();\n}\n/**\n * Generate a citation key from author and year.\n * Format: {family-name-lowercase}{year}\n * With collision handling via letter suffixes (a, b, c, ...).\n */\nexport function generateCitationKey(author, year, existingKeys) {\n    // Extract and normalize author\n    const rawFamily = author?.trim() ? extractFamilyName(author) : \"unknown\";\n    // CJK characters cannot be accurately transliterated to the correct reading\n    // (any-ascii maps them to Chinese pinyin, not Japanese romaji etc.)\n    // Fall back to 'unknown' for names containing CJK ideographs.\n    const normalizedFamily = CJK_REGEX.test(rawFamily)\n        ? \"unknown\"\n        : anyAscii(rawFamily)\n            .toLowerCase()\n            .replace(/[^a-z]/g, \"\") || \"unknown\";\n    // Normalize year\n    const normalizedYear = year?.trim() || \"0000\";\n    const baseKey = `${normalizedFamily}${normalizedYear}`;\n    // Handle collisions\n    if (!existingKeys || !existingKeys.includes(baseKey)) {\n        return baseKey;\n    }\n    // Find the first available suffix\n    for (let i = 0;; i++) {\n        const candidateKey = `${baseKey}${collisionSuffix(i)}`;\n        if (!existingKeys.includes(candidateKey)) {\n            return candidateKey;\n        }\n    }\n}\n/**\n * Generate a directory name from a citation key.\n * Format: {citationKey}-{uuid8}\n */\nexport function generateDirName(citationKey, uuid) {\n    const id = uuid ?? randomUUID();\n    const uuid8 = id.slice(0, 8);\n    return `${citationKey}-${uuid8}`;\n}\n//# sourceMappingURL=citation-key.js.map"],"names":[],"mappings":";;AAQA,SAAS,gBAAgB,OAAO;AAC5B,MAAI,SAAS;AACb,MAAI,IAAI;AACR,KAAG;AACC,aAAS,OAAO,cAAc,KAAM,IAAI,EAAG,IAAI;AAC/C,QAAI,KAAK,MAAM,IAAI,EAAE,IAAI;AAAA,EAC7B,SAAS,KAAK;AACd,SAAO;AACX;AAEA,MAAM,YAAY;AAKlB,SAAS,kBAAkB,QAAQ;AAC/B,QAAM,aAAa,OAAO,QAAQ,GAAG;AACrC,MAAI,cAAc,GAAG;AACjB,WAAO,OAAO,MAAM,GAAG,UAAU,EAAE,KAAI;AAAA,EAC3C;AACA,SAAO,OAAO,KAAI;AACtB;AAMO,SAAS,oBAAoB,QAAQ,MAAM,cAAc;AAE5D,QAAM,YAAY,QAAQ,KAAI,IAAK,kBAAkB,MAAM,IAAI;AAI/D,QAAM,mBAAmB,UAAU,KAAK,SAAS,IAC3C,YACA,SAAS,SAAS,EACf,YAAW,EACX,QAAQ,WAAW,EAAE,KAAK;AAEnC,QAAM,iBAAiB,MAAM,KAAI,KAAM;AACvC,QAAM,UAAU,GAAG,gBAAgB,GAAG,cAAc;AAEpD,MAAI,CAAC,gBAAgB,CAAC,aAAa,SAAS,OAAO,GAAG;AAClD,WAAO;AAAA,EACX;AAEA,WAAS,IAAI,KAAI,KAAK;AAClB,UAAM,eAAe,GAAG,OAAO,GAAG,gBAAgB,CAAC,CAAC;AACpD,QAAI,CAAC,aAAa,SAAS,YAAY,GAAG;AACtC,aAAO;AAAA,IACX;AAAA,EACJ;AACJ;AAKO,SAAS,gBAAgB,aAAa,MAAM;AAC/C,QAAM,KAAK,QAAQ,WAAU;AAC7B,QAAM,QAAQ,GAAG,MAAM,GAAG,CAAC;AAC3B,SAAO,GAAG,WAAW,IAAI,KAAK;AAClC;","x_google_ignoreList":[0]}

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

@@ -0,0 +1,434 @@
+import { d as distExports } from "../../../../../_virtual/index.js";
+function q(root, selector) {
+  return root.querySelector(selector);
+}
+function qa(root, selector) {
+  return root.querySelectorAll(selector);
+}
+function hasClass(el, cls) {
+  return el.classList.contains(cls);
+}
+function escapeRegex(str) {
+  return str.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+}
+function stripAnnotations(el) {
+  const clone = distExports.parse(el.outerHTML);
+  for (const ann of qa(clone, "annotation")) {
+    ann.remove();
+  }
+  return clone.textContent.trim();
+}
+function matchClassInline(el) {
+  if (hasClass(el, "ltx_font_bold")) {
+    return { type: "bold", children: parseInlineChildren(el) };
+  }
+  if (hasClass(el, "ltx_font_italic")) {
+    return { type: "italic", children: parseInlineChildren(el) };
+  }
+  if (hasClass(el, "ltx_font_typewriter")) {
+    return { type: "code", text: el.textContent.trim() };
+  }
+  return null;
+}
+function handleMath(el) {
+  const alt = el.getAttribute("alttext");
+  const text = stripAnnotations(el);
+  const result = { type: "inline-formula", text };
+  if (alt)
+    result.tex = alt;
+  return result;
+}
+function handleAnchor(el) {
+  const href = el.getAttribute("href") ?? "";
+  if (href.startsWith("#bib")) {
+    return { type: "citation", refId: href.slice(1), text: el.textContent.trim() };
+  }
+  if (href.startsWith("http://") || href.startsWith("https://")) {
+    return { type: "link", url: href, children: parseInlineChildren(el) };
+  }
+  return { type: "text", text: el.textContent.trim() };
+}
+const TAG_INLINE_HANDLERS = {
+  math: handleMath,
+  a: handleAnchor,
+  b: (el) => ({ type: "bold", children: parseInlineChildren(el) }),
+  strong: (el) => ({ type: "bold", children: parseInlineChildren(el) }),
+  i: (el) => ({ type: "italic", children: parseInlineChildren(el) }),
+  em: (el) => ({ type: "italic", children: parseInlineChildren(el) }),
+  code: (el) => ({ type: "code", text: el.textContent.trim() }),
+  sup: (el) => ({ type: "superscript", text: el.textContent.trim() }),
+  sub: (el) => ({ type: "subscript", text: el.textContent.trim() })
+};
+const INLINE_CONTAINER_TAGS = /* @__PURE__ */ new Set(["span", "cite"]);
+function processInlineElement(el) {
+  const classMatch = matchClassInline(el);
+  if (classMatch)
+    return [classMatch];
+  const tag = el.tagName?.toLowerCase() ?? "";
+  const handler = TAG_INLINE_HANDLERS[tag];
+  if (handler) {
+    const content = handler(el);
+    return content ? [content] : [];
+  }
+  if (INLINE_CONTAINER_TAGS.has(tag)) {
+    return parseInlineChildren(el);
+  }
+  const text = el.textContent.trim();
+  return text ? [{ type: "text", text }] : [];
+}
+function parseInlineChildren(parent) {
+  const result = [];
+  for (const node of parent.childNodes) {
+    if (node.nodeType === 3) {
+      const text = node.textContent;
+      if (text)
+        result.push({ type: "text", text });
+    } else if (node.nodeType === 1) {
+      result.push(...processInlineElement(node));
+    }
+  }
+  return result;
+}
+function parseParagraph(el) {
+  const innerP = q(el, "p");
+  return { type: "paragraph", content: parseInlineChildren(innerP ?? el) };
+}
+function parseLtxPara(el) {
+  const blocks = [];
+  for (const child of el.childNodes) {
+    if (child.nodeType !== 1)
+      continue;
+    const childEl = child;
+    const tag = childEl.tagName?.toLowerCase() ?? "";
+    if (tag === "section" || /^h[1-6]$/.test(tag))
+      continue;
+    const block = matchBlockByClass(childEl) ?? matchBlockByTag(childEl, tag);
+    if (block) {
+      blocks.push(block);
+    }
+  }
+  if (blocks.length === 0) {
+    blocks.push(parseParagraph(el));
+  }
+  return blocks;
+}
+function parseList(el) {
+  const tag = el.tagName?.toLowerCase() ?? "";
+  const ordered = tag === "ol" || hasClass(el, "ltx_enumerate");
+  const items = [];
+  for (const li of qa(el, "li")) {
+    items.push(parseInlineChildren(li));
+  }
+  return { type: "list", ordered, items };
+}
+function parseFigure(el) {
+  const labelEl = q(el, ".ltx_caption .ltx_tag_figure");
+  const captionEl = q(el, ".ltx_caption");
+  const id = el.getAttribute("id");
+  let label;
+  if (labelEl) {
+    label = labelEl.textContent.trim().replace(/:$/, "");
+  }
+  let caption;
+  if (captionEl) {
+    const captionText = captionEl.textContent.trim();
+    caption = label ? captionText.replace(new RegExp(`^${escapeRegex(label)}[:\\s]*`), "").trim() : captionText;
+  }
+  const block = { type: "figure" };
+  if (id)
+    block.id = id;
+  if (label)
+    block.label = label;
+  if (caption)
+    block.caption = caption;
+  return block;
+}
+function parseTableHeaders(thead) {
+  const headers = [];
+  const headerRow = q(thead, "tr") ?? q(thead, ".ltx_tr");
+  if (headerRow) {
+    for (const th of qa(headerRow, "th, td, .ltx_td")) {
+      headers.push(stripAnnotations(th));
+    }
+  }
+  return headers;
+}
+function parseTableRows(el, thead) {
+  const rows = [];
+  const tbody = q(el, "tbody") ?? q(el, ".ltx_tbody") ?? el;
+  for (const tr of qa(tbody, "tr, .ltx_tr")) {
+    if (thead && tr.parentNode === thead)
+      continue;
+    const cells = [];
+    for (const td of qa(tr, "td, th, .ltx_td")) {
+      cells.push(stripAnnotations(td));
+    }
+    if (cells.length > 0)
+      rows.push(cells);
+  }
+  return rows;
+}
+function parseTable(el) {
+  const captionEl = q(el, ".ltx_caption");
+  const thead = q(el, "thead") ?? q(el, ".ltx_thead");
+  const headers = thead ? parseTableHeaders(thead) : [];
+  const rows = parseTableRows(el, thead);
+  if (captionEl) {
+    return { type: "table", caption: captionEl.textContent.trim(), headers, rows };
+  }
+  return { type: "table", headers, rows };
+}
+function parseFormula(el) {
+  const mathEl = q(el, "math");
+  const id = el.getAttribute("id");
+  const labelEl = q(el, ".ltx_tag_equation");
+  const label = labelEl ? labelEl.textContent.trim() : void 0;
+  let tex;
+  let text;
+  if (mathEl) {
+    const alt = mathEl.getAttribute("alttext");
+    if (alt)
+      tex = alt;
+    text = stripAnnotations(mathEl);
+  } else {
+    text = el.textContent.trim();
+  }
+  const block = { type: "formula" };
+  if (id)
+    block.id = id;
+  if (label)
+    block.label = label;
+  if (tex)
+    block.tex = tex;
+  if (text && !tex)
+    block.text = text;
+  return block;
+}
+function matchBlockByClass(el) {
+  if (hasClass(el, "ltx_equation") || hasClass(el, "ltx_eqn_table"))
+    return parseFormula(el);
+  if (hasClass(el, "ltx_tabular"))
+    return parseTable(el);
+  return null;
+}
+function matchBlockByTag(el, tag) {
+  if (tag === "p")
+    return { type: "paragraph", content: parseInlineChildren(el) };
+  if (tag === "ol" || tag === "ul")
+    return parseList(el);
+  if (tag === "figure" && hasClass(el, "ltx_table"))
+    return parseTable(el);
+  if (tag === "table")
+    return parseTable(el);
+  if (tag === "figure")
+    return parseFigure(el);
+  if (tag === "blockquote")
+    return { type: "blockquote", content: parseInlineChildren(el) };
+  if (tag === "pre")
+    return { type: "preformat", text: el.textContent.trim() };
+  return null;
+}
+function matchBlock(el) {
+  const tag = el.tagName?.toLowerCase() ?? "";
+  if (tag === "section" || /^h[1-6]$/.test(tag))
+    return null;
+  return matchBlockByClass(el) ?? matchBlockByTag(el, tag);
+}
+function parseBlockContent(parent) {
+  const blocks = [];
+  for (const child of parent.childNodes) {
+    if (child.nodeType !== 1)
+      continue;
+    const el = child;
+    if (hasClass(el, "ltx_para")) {
+      blocks.push(...parseLtxPara(el));
+      continue;
+    }
+    const block = matchBlock(el);
+    if (block) {
+      blocks.push(block);
+      continue;
+    }
+    const tag = el.tagName?.toLowerCase() ?? "";
+    if (tag !== "section" && tag !== "nav" && !/^h[1-6]$/.test(tag)) {
+      const text = el.textContent.trim();
+      if (text)
+        blocks.push({ type: "paragraph", content: parseInlineChildren(el) });
+    }
+  }
+  return blocks;
+}
+function getSectionLevel(el) {
+  if (hasClass(el, "ltx_subsection"))
+    return 3;
+  if (hasClass(el, "ltx_subsubsection"))
+    return 4;
+  if (hasClass(el, "ltx_paragraph"))
+    return 5;
+  return 2;
+}
+function findHeading(el, level) {
+  return q(el, `:scope > h${level}`) ?? q(el, ":scope > h2") ?? q(el, ":scope > h3") ?? q(el, ":scope > h4") ?? q(el, ":scope > h5") ?? q(el, ":scope > h6");
+}
+function parseSection(el) {
+  const level = getSectionLevel(el);
+  const headingEl = findHeading(el, level);
+  const title = headingEl ? headingEl.textContent.trim() : "";
+  const content = parseBlockContent(el);
+  const subsections = [];
+  for (const child of el.childNodes) {
+    if (child.nodeType !== 1)
+      continue;
+    const childEl = child;
+    if (childEl.tagName?.toLowerCase() === "section") {
+      subsections.push(parseSection(childEl));
+    }
+  }
+  return { title, level, content, subsections };
+}
+function parseTitle(root) {
+  const titleEl = q(root, ".ltx_title.ltx_title_document");
+  if (!titleEl)
+    return "";
+  const titleClone = distExports.parse(titleEl.outerHTML);
+  for (const authorInTitle of qa(titleClone, ".ltx_authors")) {
+    authorInTitle.remove();
+  }
+  return titleClone.textContent.trim();
+}
+function parseAuthorName(commaPart) {
+  const lines = commaPart.split(/\n/);
+  const firstLine = lines.find((l) => l.trim() !== "")?.trim() ?? "";
+  if (!firstLine)
+    return null;
+  if (/^\d/.test(firstLine))
+    return null;
+  const name = firstLine.replace(/\d+$/, "").trim();
+  if (!name)
+    return null;
+  const words = name.split(/\s+/);
+  const surname = words.pop() ?? name;
+  const author = { surname };
+  if (words.length > 0)
+    author.givenNames = words.join(" ");
+  return author;
+}
+function parseAuthors(root) {
+  const authors = [];
+  for (const authorEl of qa(root, ".ltx_authors .ltx_personname")) {
+    const fullText = authorEl.textContent.trim();
+    if (!fullText)
+      continue;
+    for (const part of fullText.split(",")) {
+      const author = parseAuthorName(part);
+      if (author)
+        authors.push(author);
+    }
+  }
+  return authors;
+}
+function parseAbstract(root) {
+  const abstractEl = q(root, ".ltx_abstract");
+  if (!abstractEl)
+    return void 0;
+  const clone = distExports.parse(abstractEl.outerHTML);
+  const absTitle = q(clone, ".ltx_title");
+  if (absTitle)
+    absTitle.remove();
+  for (const note of qa(clone, ".ltx_note")) {
+    note.remove();
+  }
+  const text = clone.textContent.trim();
+  return text || void 0;
+}
+function parseMetadata(root) {
+  const title = parseTitle(root);
+  const authors = parseAuthors(root);
+  const abstract = parseAbstract(root);
+  const keywordEls = qa(root, ".ltx_keywords .ltx_text");
+  const keywords = keywordEls.map((kw) => kw.textContent.trim()).filter(Boolean);
+  const metadata = { title, authors };
+  if (abstract)
+    metadata.abstract = abstract;
+  if (keywords.length > 0)
+    metadata.keywords = keywords;
+  return metadata;
+}
+function extractRefText(item) {
+  const bibBlock = q(item, ".ltx_bibblock");
+  if (bibBlock)
+    return bibBlock.textContent.trim();
+  let text = item.textContent.trim();
+  const labelEl = q(item, ".ltx_tag_bibitem");
+  if (labelEl) {
+    const labelText = labelEl.textContent.trim();
+    if (text.startsWith(labelText)) {
+      text = text.slice(labelText.length).trim();
+    }
+  }
+  return text;
+}
+function extractDoi(item) {
+  for (const link of qa(item, "a")) {
+    const href = link.getAttribute("href") ?? "";
+    const doiMatch = href.match(/doi\.org\/(.+)/);
+    if (doiMatch)
+      return doiMatch[1];
+  }
+  return void 0;
+}
+function parseReferences(root) {
+  const refs = [];
+  for (const item of qa(root, ".ltx_bibitem")) {
+    const id = item.getAttribute("id") ?? `ref${refs.length + 1}`;
+    const text = extractRefText(item);
+    const doi = extractDoi(item);
+    const ref = { id, text };
+    if (doi)
+      ref.doi = doi;
+    refs.push(ref);
+  }
+  return refs;
+}
+function parseAcknowledgments(root) {
+  const ackEl = q(root, ".ltx_acknowledgement");
+  if (!ackEl)
+    return void 0;
+  const clone = distExports.parse(ackEl.outerHTML);
+  const heading = q(clone, ".ltx_title");
+  if (heading)
+    heading.remove();
+  const text = clone.textContent.trim();
+  return text || void 0;
+}
+function isBodySection(el) {
+  return el.tagName?.toLowerCase() === "section" && !hasClass(el, "ltx_bibliography") && !hasClass(el, "ltx_acknowledgement");
+}
+function parseBodySections(article) {
+  const sections = [];
+  for (const child of article.childNodes) {
+    if (child.nodeType !== 1)
+      continue;
+    const el = child;
+    if (isBodySection(el)) {
+      sections.push(parseSection(el));
+    }
+  }
+  return sections;
+}
+function parseArxivHtml(html) {
+  const root = distExports.parse(html);
+  const metadata = parseMetadata(root);
+  const references = parseReferences(root);
+  const article = q(root, "article.ltx_document") ?? root;
+  const sections = parseBodySections(article);
+  const acknowledgments = parseAcknowledgments(root);
+  const doc = { metadata, sections, references };
+  if (acknowledgments)
+    doc.acknowledgments = acknowledgments;
+  return doc;
+}
+export {
+  parseArxivHtml
+};
+//# sourceMappingURL=arxiv-html-parser.js.map