@arela/uploader 1.0.19 → 1.0.21

package/src/document-types/pedimento-completo.js

@@ -0,0 +1,68 @@
+import {
+  pedimentoYearFromFields,
+  sharedPedimentoExtractors,
+} from './_pedimento-shared-extractors.js';
+
+/**
+ * "Pedimento Completo" matcher — the 7-page DEF / SEGUNDA / TERCERA copy
+ * (and the "CoveFact" variant). This is distinct from `pedimento_simplificado`
+ * which uses the "FORMA SIMPLIFICADA DE PEDIMENTO" header.
+ *
+ * Match strategy: require the structural fields that the long-form pedimento
+ * always carries (`NUM. PEDIMENTO:`, `CVE.PEDIMENTO:`, `T.OPER:`) plus at
+ * least one of the printed copy markers, while explicitly excluding any
+ * document that already declares itself as a "FORMA SIMPLIFICADA".
+ */
+export const pedimentoCompletoDefinition = {
+  type: 'pedimento_completo',
+  extensions: ['pdf'],
+  match: (source) => {
+    if (/FORMA SIMPLIFICADA DE PEDIMENTO/i.test(source)) return false;
+
+    const hasHeaderFields =
+      /NUM\.?\s*PEDIMENTO:/i.test(source) &&
+      /CVE\.?\s*PEDIMENTO:/i.test(source) &&
+      /T\.?\s*OPER:/i.test(source);
+    if (!hasHeaderFields) return false;
+
+    const hasCopyMarker =
+      /ORIGINAL:\s*ADMINISTRACION GENERAL DE ADUANAS/i.test(source) ||
+      /SEGUNDA COPIA/i.test(source) ||
+      /TERCERA COPIA/i.test(source) ||
+      /COPIA\s+(SIMPLIFICAD[AO])?\s*TRANSPORTISTA/i.test(source) ||
+      /DEFINITIVO/i.test(source);
+
+    return hasCopyMarker;
+  },
+
+  /**
+   * Resolve the final document type after fields have been extracted.
+   * Mirrors the simplificado logic:
+   *   - R1 rectifications require fechaPagoRectificacion
+   *   - Everything else requires paymentDate
+   * No payment evidence ⇒ proforma_completo.
+   */
+  resolveType: (fields) => {
+    const clavePedimento =
+      fields?.find((f) => f.name === 'clavePedimento')?.value ?? null;
+    const paymentDate =
+      fields?.find((f) => f.name === 'paymentDate' && f.found)?.value ?? null;
+    const fechaPagoRectificacion =
+      fields?.find((f) => f.name === 'fechaPagoRectificacion' && f.found)
+        ?.value ?? null;
+
+    if (clavePedimento === 'R1') {
+      return fechaPagoRectificacion
+        ? 'pedimento_completo'
+        : 'proforma_completo';
+    }
+    return paymentDate ? 'pedimento_completo' : 'proforma_completo';
+  },
+
+  extractNumPedimento: (source, fields) => {
+    return fields?.find((f) => f.name === 'numPedimento')?.value ?? null;
+  },
+  extractPedimentoYear: (source, fields) => pedimentoYearFromFields(fields),
+
+  extractors: sharedPedimentoExtractors,
+};

package/src/document-types/pedimento-simplificado.js

@@ -1,4 +1,7 @@
-import { FieldResult } from '../document-type-shared.js';
+import {
+  pedimentoYearFromFields,
+  sharedPedimentoExtractors,
+} from './_pedimento-shared-extractors.js';
 
 export const pedimentoSimplificadoDefinition = {
   type: 'pedimento_simplificado',
@@ -14,7 +17,7 @@ export const pedimentoSimplificadoDefinition = {
    *   - Otherwise: must have paymentDate
    * If no payment evidence is found, it's a "proforma".
    *
-   * @param {FieldResult[]} fields - Extracted fields
+   * @param {import('../document-type-shared.js').FieldResult[]} fields
    * @returns {string} - 'pedimento_simplificado' or 'proforma'
    */
   resolveType: (fields) => {
@@ -27,297 +30,15 @@ export const pedimentoSimplificadoDefinition = {
         ?.value ?? null;
 
     if (clavePedimento === 'R1') {
-      // Rectification pedimentos require fechaPagoRectificacion
       return fechaPagoRectificacion ? 'pedimento_simplificado' : 'proforma';
     }
-
-    // Regular pedimentos require paymentDate
     return paymentDate ? 'pedimento_simplificado' : 'proforma';
   },
 
   extractNumPedimento: (source, fields) => {
     return fields?.find((f) => f.name === 'numPedimento')?.value ?? null;
   },
-  extractPedimentoYear: (source, fields) => {
-    const numPedimento = fields?.find((f) => f.name === 'numPedimento')?.value;
-    if (!numPedimento) {
-      return null;
-    }
-    const year = parseInt(numPedimento.substring(0, 2), 10);
-    return year < 50 ? year + 2000 : year + 1900;
-  },
-  extractors: [
-    // 1) Número de Pedimento (15 dígitos)
-    {
-      field: 'numPedimento',
-      extract: (source) => {
-        const match = source.match(/\d{2}\s?\d{2}\s?\d{4}\s?\d{7}/);
-        return new FieldResult(
-          'numPedimento',
-          !!match,
-          match ? match[0].replace(/\s/g, '') : null,
-        );
-      },
-    },
-
-    // 2) Tipo de Operación: los 3 caracteres justo después del número
-    {
-      field: 'tipoOperacion',
-      extract: (source) => {
-        // Look for the pedimento number pattern followed by operation type
-        // Pattern matches: "22 07 3429 2002089 EXP RT"
-        const match = source.match(
-          /\d{2}\s+\d{2}\s+\d{4}\s+\d{7}\s+([A-Z]{3})/,
-        );
-        return new FieldResult(
-          'tipoOperacion',
-          !!match,
-          match ? match[1] : null,
-        );
-      },
-    },
-
-    // 3) Clave de Pedimento: los 2 caracteres justo después de la operación
-    {
-      field: 'clavePedimento',
-      extract: (source) => {
-        // Look for the pedimento number pattern followed by operation type and then the key
-        // Pattern matches: "22 07 3429 2002089 EXP RT" to capture "RT"
-        const match = source.match(
-          /\d{2}\s+\d{2}\s+\d{4}\s+\d{7}\s+[A-Z]{3}\s+([A-Z0-9]{2})/,
-        );
-        return new FieldResult(
-          'clavePedimento',
-          !!match,
-          match ? match[1] : null,
-        );
-      },
-    },
-
-    // 4) Aduana E/S: Extract the 3-digit aduana code that appears at the end of the peso bruto line
-    {
-      field: 'aduanaEntradaSalida',
-      extract: (source) => {
-        // Look for the peso bruto line format: number followed by decimal amount followed by 3-digit aduana code
-        // Pattern matches formats like: "7 1.100 071" or "7 19,834.260 071" to capture "071"
-        // Uses multiline flag to match line boundaries precisely
-        const match = source.match(/^\s*\d+\s+[\d,.]+\s+(\d{3})\s*$/m);
-        return new FieldResult(
-          'aduanaEntradaSalida',
-          !!match,
-          match ? match[1] : null,
-        );
-      },
-    },
-
-    // 5) RFC: línea con 12-13 caracteres alfanuméricos
-    {
-      field: 'rfc',
-      extract: (source) => {
-        const match = source.match(/\n\s*([A-Z0-9]{12,13})\s*\n/);
-        return new FieldResult('rfc', !!match, match ? match[1] : null);
-      },
-    },
-
-    // 6) Código de Aceptación: línea con 8 caracteres alfanuméricos justo después del RFC
-    {
-      field: 'codigoAceptacion',
-      extract: (source) => {
-        // 1) split into trimmed, non-empty lines
-        const lines = source
-          .split(/\r?\n/)
-          .map((l) => l.trim())
-          .filter((l) => l.length > 0);
-
-        // 2) find the index of an RFC line (12–13 alnum chars)
-        const rfcIndex = lines.findIndex((l) => /^[A-Z0-9]{12,13}$/.test(l));
-        let code = null;
-
-        // 3) if next line exists and is exactly 8 alnum chars, that's the code
-        if (rfcIndex >= 0 && /^[A-Z0-9]{8}$/.test(lines[rfcIndex + 1] || '')) {
-          code = lines[rfcIndex + 1];
-        }
-
-        return new FieldResult('codigoAceptacion', code !== null, code);
-      },
-    },
-
-    // 7) Num. E-Document: exactamente 13 caracteres tras la etiqueta (puede haber múltiples líneas)
-    // {
-    //   field: 'numEDocumento',
-    //   extract: (source) => {
-    //     // Split into lines and find all lines containing NUM. E-DOCUMENT
-    //     const lines = source.split(/\r?\n/);
-    //     const edocLines = lines.filter((line) =>
-    //       /NUM\.?\s*E-DOCUMENT/i.test(line),
-    //     );
-
-    //     if (edocLines.length === 0) {
-    //       return new FieldResult('numEDocumento', false, null);
-    //     }
-
-    //     // Extract all 13-character alphanumeric codes from all NUM. E-DOCUMENT lines
-    //     const extractedCodes = [];
-    //     edocLines.forEach((line) => {
-    //       const afterEdoc = line.replace(/.*NUM\.?\s*E-DOCUMENT\s*/i, '');
-    //       const codes = afterEdoc.match(/[A-Z0-9]{13}/g) || [];
-    //       extractedCodes.push(...codes);
-    //     });
-
-    //     if (extractedCodes.length === 0) {
-    //       return new FieldResult('numEDocumento', false, null);
-    //     }
-
-    //     // Remove duplicates using Set
-    //     const uniqueCodes = [...new Set(extractedCodes)];
-    //     const formattedValue = `[${uniqueCodes.join(',')}]`;
-    //     return new FieldResult('numEDocumento', true, formattedValue);
-    //   },
-    // },
-
-    // 8) Fecha de Pago: Look for various payment date patterns
-    {
-      field: 'paymentDate',
-      extract: (source) => {
-        // Try multiple patterns for payment dates
-        let match = source.match(/2\s+PAGO:\s*(\d{2}\/\d{2}\/\d{4})/);
-        if (!match) {
-          match = source.match(/FECHA DE PAGO:\s*(\d{4}\/\d{2}\/\d{2})/);
-        }
-        if (!match) {
-          match = source.match(/PRESENTACION:\s*(\d{2}\/\d{2}\/\d{4})/);
-        }
-        return new FieldResult('paymentDate', !!match, match ? match[1] : null);
-      },
-    },
-
-    // 9) COVE: NUMERO DE ACUSE DE VALOR (puede haber múltiples líneas)
-    // {
-    //   field: 'cove',
-    //   extract: (source) => {
-    //     // Split into lines and find all lines containing NUMERO DE ACUSE DE VALOR or COVE
-    //     const lines = source.split(/\r?\n/);
-    //     const coveLines = lines.filter(
-    //       (line) =>
-    //         /COVE/i.test(line) || /NUMERO DE ACUSE DE VALOR/i.test(line),
-    //     );
-
-    //     if (coveLines.length === 0) {
-    //       return new FieldResult('cove', false, null);
-    //     }
-
-    //     // Extract all COVE values from all matching lines
-    //     const coveValues = [];
-    //     coveLines.forEach((line) => {
-    //       const coveMatches = line.match(/COVE([A-Z0-9]+)/gi) || [];
-    //       // Extract just the alphanumeric parts (remove the "COVE" prefix)
-    //       const codes = coveMatches.map((match) => match);
-    //       coveValues.push(...codes);
-    //     });
-
-    //     if (coveValues.length === 0) {
-    //       return new FieldResult('cove', false, null);
-    //     }
-
-    //     // Remove duplicates using Set
-    //     const uniqueCoveValues = [...new Set(coveValues)];
-    //     const formattedValue = `[${uniqueCoveValues.join(',')}]`;
-    //     return new FieldResult('cove', true, formattedValue);
-    //   },
-    // },
-
-    // 10) Peso Bruto: Extract weight value
-    // {
-    //   field: 'pesoBruto',
-    //   extract: (source) => {
-    //     // Look for the peso bruto value with decimal format
-    //     const match = source.match(/(\d+\.\d+)\d{3}/);
-    //     return new FieldResult('pesoBruto', !!match, match ? match[1] : null);
-    //   },
-    // },
-
-    // 11) Patente: Extract patent number
-    {
-      field: 'patente',
-      extract: (source) => {
-        // Look for the PATENTE: PEDIMENTO: ADUANA: header line
-        // Then find the corresponding data line with three numbers
-        const lines = source.split(/\r?\n/);
-        const patenteHeaderIndex = lines.findIndex((line) =>
-          /PATENTE:.*PEDIMENTO:.*ADUANA:/i.test(line),
-        );
-
-        if (patenteHeaderIndex >= 0) {
-          // Look for the data line after the header (format: "3429 2002089 07")
-          for (let i = patenteHeaderIndex + 1; i < lines.length; i++) {
-            const line = lines[i].trim();
-            if (/^\d+\s+\d+\s+\d+$/.test(line)) {
-              const parts = line.split(/\s+/);
-              return new FieldResult('patente', true, parts[0]); // First number is the PATENTE
-            }
-          }
-        }
-
-        return new FieldResult('patente', false, null);
-      },
-    },
-
-    // 12) Numero de Operacion Bancaria
-    // {
-    //   field: 'numeroOperacionBancaria',
-    //   extract: (source) => {
-    //     const match = source.match(
-    //       /NUMERO DE OPERACION BANCARIA:\s*([A-Z0-9]+)/i,
-    //     );
-    //     return new FieldResult(
-    //       'numeroOperacionBancaria',
-    //       !!match,
-    //       match ? match[1] : null,
-    //     );
-    //   },
-    // },
-
-    // 13) Numero de Transaccion SAT
-    // {
-    //   field: 'numeroTransaccionSAT',
-    //   extract: (source) => {
-    //     const match = source.match(/NUMERO DE TRANSACCION SAT:\s*([A-Z0-9]+)/i);
-    //     return new FieldResult(
-    //       'numeroTransaccionSAT',
-    //       !!match,
-    //       match ? match[1] : null,
-    //     );
-    //   },
-    // },
-
-    // 14) Fecha de Pago Rectificación
-    {
-      field: 'fechaPagoRectificacion',
-      extract: (source) => {
-        // Look for the RECTIFICACION section header
-        const rectSectionMatch = source.match(
-          /RECTIFICACION[\s\S]{0,500}?(\d{2}\/\d{2}\/\d{4})/i,
-        );
-
-        if (rectSectionMatch) {
-          return new FieldResult(
-            'fechaPagoRectificacion',
-            true,
-            rectSectionMatch[1],
-          );
-        }
-
-        // Fallback: look for any date after FECHA PAGO RECT
-        const fechaMatch = source.match(
-          /FECHA PAGO RECT[\s\S]{0,500}?(\d{2}\/\d{2}\/\d{4})/i,
-        );
-
-        if (fechaMatch) {
-          return new FieldResult('fechaPagoRectificacion', true, fechaMatch[1]);
-        }
+  extractPedimentoYear: (source, fields) => pedimentoYearFromFields(fields),
 
-        return new FieldResult('fechaPagoRectificacion', false, null);
-      },
-    },
-  ],
+  extractors: sharedPedimentoExtractors,
 };

package/src/file-detection.js

@@ -4,6 +4,35 @@ import { PDFParse } from 'pdf-parse';
 
 import { extractDocumentFields } from './document-type-shared.js';
 
+// Document types that participate in arela_path composition. The XML type is
+// kept here even though its matcher is currently disabled — once re-enabled
+// in document-type-shared.js no further changes are needed here.
+const ARELA_PATH_TYPES = new Set([
+  'pedimento_simplificado',
+  'pedimento_completo',
+  'pedimento_completo_xml',
+]);
+
+/**
+ * For `pedimento_completo_xml` the patente is not present in the XML body —
+ * it must be parsed from the filename. Three known patterns are tried.
+ */
+function patenteFromXmlFilename(filePath) {
+  if (!filePath) return null;
+  const fileName = path.basename(filePath);
+
+  let m = fileName.match(/^VU_(\d{4})_\d{3}_\d{7}\.xml$/i);
+  if (m) return m[1];
+
+  m = fileName.match(/^\d{3}-(\d{4})-\d{7}\.xml$/i);
+  if (m) return m[1];
+
+  m = fileName.match(/^\d{4}(\d{4})\d{7}(?:_\d{15})?\.xml$/i);
+  if (m) return m[1];
+
+  return null;
+}
+
 /**
  * Compose arela_path from extracted pedimento fields
  * Format: RFC/Year/Patente/Aduana/Pedimento/
@@ -15,16 +44,21 @@ function composeArelaPath(
   detectedPedimentoYear,
   filePath,
 ) {
-  if (detectedType !== 'pedimento_simplificado') {
+  if (!ARELA_PATH_TYPES.has(detectedType)) {
     return null;
   }
 
   const rfc = fields?.find((f) => f.name === 'rfc')?.value;
-  const patente = fields?.find((f) => f.name === 'patente')?.value;
+  let patente = fields?.find((f) => f.name === 'patente')?.value;
   const aduana = fields?.find((f) => f.name === 'aduanaEntradaSalida')?.value;
   const pedimento = fields?.find((f) => f.name === 'numPedimento')?.value;
   const year = detectedPedimentoYear;
 
+  // XML matcher does not extract patente from the body — derive from filename.
+  if (!patente && detectedType === 'pedimento_completo_xml') {
+    patente = patenteFromXmlFilename(filePath);
+  }
+
   // All components are required for a valid arela_path
   if (!rfc || !year || !patente || !aduana || !pedimento) {
     console.log('⚠️ Missing required fields for arela_path composition:', {
@@ -155,12 +189,12 @@ export class FileDetectionService {
    * @returns {Promise<string>} - Extracted text
    */
   async extractTextFromPDF(filePath) {
+    let parser;
     try {
       const dataBuffer = fs.readFileSync(filePath);
-      // Convert Buffer to Uint8Array as required by pdf-parse
       const uint8Array = new Uint8Array(dataBuffer);
-      const pdfParse = new PDFParse(uint8Array);
-      const result = await pdfParse.getText();
+      parser = new PDFParse({ data: uint8Array });
+      const result = await parser.getText();
       return result.text;
     } catch (error) {
       console.error(
@@ -168,6 +202,10 @@ export class FileDetectionService {
         error.message,
       );
       throw new Error(`Failed to extract text from PDF: ${error.message}`);
+    } finally {
+      if (parser) {
+        await parser.destroy();
+      }
     }
   }
 

package/src/index.js

@@ -1,6 +1,7 @@
 #!/usr/bin/env node
 import { Command } from 'commander';
 
+import gdriveSyncCommand from './commands/GDriveSyncCommand.js';
 import identifyCommand from './commands/IdentifyCommand.js';
 import pollWorkerCommand from './commands/PollWorkerCommand.js';
 import PropagateCommand from './commands/PropagateCommand.js';
@@ -458,6 +459,32 @@ class ArelaUploaderCLI {
     // END OF NEW SIMPLIFIED COMMANDS
     // ============================================================================
 
+    // GDrive sync command - mirror a Google Drive folder to local before scan
+    this.program
+      .command('gdrive-sync')
+      .description(
+        '☁️  Mirror a Google Drive folder to local filesystem (pre-scan source)',
+      )
+      .option(
+        '--root-folder <id>',
+        'Drive folder ID to sync (overrides GDRIVE_ROOT_FOLDER_ID)',
+      )
+      .option(
+        '--dest <path>',
+        'Local mirror destination (overrides GDRIVE_LOCAL_MIRROR_PATH)',
+      )
+      .option('--full', 'Ignore state file and re-verify all files')
+      .option('--dry-run', 'List/plan only, no downloads or writes')
+      .action(async (options) => {
+        try {
+          await gdriveSyncCommand.execute(options);
+        } catch (error) {
+          this.errorHandler.handleFatalError(error, {
+            command: 'gdrive-sync',
+          });
+        }
+      });
+
     // Watch command
     this.program
       .command('watch')

package/src/services/DatabaseService.js

@@ -144,7 +144,9 @@ export class DatabaseService {
         rfc: null,
         message: null,
         file_extension: fileExtension,
-        is_like_simplificado: filename.toLowerCase().includes('simp'),
+        // Flag any PDF whose filename hints at a pedimento (simplificado,
+        // completo, or CoveFact). Column name preserved; semantics broadened.
+        is_like_simplificado: /(simp|pedim|covefact)/i.test(filename),
         year: null,
         created_at: new Date().toISOString(),
         updated_at: new Date().toISOString(),