const ExcelJS = require('exceljs'); const { PDFParse } = require('pdf-parse'); const { PDFDocument } = require('pdf-lib'); const readline = require('node:readline/promises'); const { stdin: input, stdout: output } = require('node:process'); const fs = require('fs'); const os = require('os'); const path = require('path'); const sfdc = require('./sfdc2'); require('dotenv').config(); // excel1 => /Users/ddatinguinoo/Downloads/Estratto IPP 5 Sic.xlsx // excel2 => /Users/ddatinguinoo/Downloads/Estratto Haccp IPP 5 (1).xlsx // excel3 di prova => /Users/ddatinguinoo/Downloads/File di prova.xlsx // pdf1 => /Users/ddatinguinoo/Downloads/G-S IPP 5.pdf // pdf2 => /Users/ddatinguinoo/Downloads/H IPP 5.pdf const authConfig = { sfdcTokenFile: process.env.SFDC_TOKEN_FILE, // Path to store/retrieve OAuth tokens sfdcClientId: process.env.SFDC_CLIENT_ID, sfdcClientSecret: process.env.SFDC_CLIENT_SECRET }; async function initializeSalesforceConnection() { // sfdc.setProductionBaseUrl(); // or sfdc.setSandboxBaseUrl(); sfdc.setSandboxBaseUrl(); // Example: using sandbox sfdc.setTokenFile(authConfig.sfdcTokenFile); sfdc.setClientId(authConfig.sfdcClientId); sfdc.setClientSecret(authConfig.sfdcClientSecret); // Try to load existing token or initiate OAuth flow if (!await sfdc.getSalesforceToken()) { console.log('No token found. Initiating OAuth2 flow...'); // For command-line scripts where a browser can be opened: await sfdc.initToken((loginUrl) => { console.log(`Please open this URL to authenticate: ${loginUrl}`); }); } else { console.log('Token loaded successfully.'); } // Ensure the access token is valid and refresh if necessary await sfdc.checkToken(); console.log('Salesforce connection is ready.'); } // Escape single quotes for SOQL queries (e.g., D'Angelo -> D\'Angelo) function escapeSoql(str) { return str ? str.replace(/'/g, "\\'") : ''; } // the pdf files contain the identifying ids of the users such as 'Codice Fiscale' (CF) and we use them to attach // the page/pages to the specific account that has that id. If CFs are not available we search based on the // name an lastname of the user async function attachPdfToAccount(sfdc, tempFiles, isDryRun = false) { let parser; try { let contentVersion; const rl = readline.createInterface({ input, output }); for (let i = 0; i < tempFiles.length; i++) { // 1. Read the local PDF file and convert it to Base64 const pdfBuffer = fs.readFileSync(tempFiles[i]); const base64Pdf = pdfBuffer.toString('base64'); parser = new PDFParse({ data: pdfBuffer }); try { const result = await parser.getText(); let conditions = []; const userID = result.text.match(/Codice fiscale:\s*([A-Za-z0-9]+)/i); const codiceFiscale = userID ? userID[1].trim() : null; if (codiceFiscale) { conditions.push(`CF__c='${ codiceFiscale }'`); } // since the name and lastname of the user are reversed in some files, we need to put it in order // to match the name from the database const nameMatch = result.text.match(/Il\s+Sig\.\/La\s+Sig\.ra[ \t]+([A-Za-zÀ-ÿ']+(?:[ \t]+[A-Za-zÀ-ÿ']+)+)/i); if (nameMatch) { const rawName = nameMatch[1].trim() const parts = rawName.split(/\s+/); if (parts.length >= 2) { const fullName = parts.join(' '); const reversedName = [...parts.slice(1), parts[0]].join(' '); if (fullName !== reversedName) { conditions.push(`Name='${ escapeSoql(reversedName) }'`); } } conditions.push(`Name='${ escapeSoql(rawName) }'`); } // use the soql command to search in the database // if codice fiscale is not available, I search the user's account based on their name and lastname const soql1 = `SELECT Id, Name FROM Account WHERE (${conditions.join(' OR ')}) ORDER BY CreatedDate DESC LIMIT 1`; const result_soql = await sfdc.query(soql1); if (result_soql && result_soql.records && result_soql.records.length > 0) { // if we find something let matchFound = result_soql.records[0]; console.log('--- Match found! ---'); if (isDryRun) { console.log('DRY RUN: non creato documento'); } else { const title = await rl.question('Choose a title for the document: '); const nowIsoString = new Date().toISOString(); // current timestamp in UTC const newDocument = await sfdc.create('Documento__c', { Account__c: `${ result_soql.records[0].Id }`, Name: `${ title }`, Data_Caricamento__c: nowIsoString, Tipo_Documento__c: "Stage doc", Stato__c: 'Valido' }); // this is the query that is needed to extract the Id of the client and assign it // to FirstPublishLocationId field const soql2 = `SELECT Account__c FROM Documento__c WHERE Id='${ newDocument.id }'`; const docQuery = await sfdc.query(soql2); console.log(`Doc Id: ${ newDocument.id }`); contentVersion = await sfdc.create('ContentVersion', { Title: `${ result_soql.records[0].Name }_Document.pdf`, PathOnClient: `${ result_soql.records[0].Name }_Document.pdf`, VersionData: base64Pdf, // Must be Base64 string FirstPublishLocationId: `${ docQuery.records[0].Account__c }` }); } } else { console.log('Account non trovato su Salesforce!'); } } finally { if (parser) { await parser.destroy(); } } } console.log(`PDFs successfully attached!`); console.log('Dry run completed!'); rl.close(); } catch (error) { console.error(`Error uploading PDF: `, error); } } // this is where i need to store each page in a temporary pdf file async function processPage(pdfDoc, pageIndex, tempFiles) { try { const newDoc = await PDFDocument.create(); const [ copiedPage ] = await newDoc.copyPages(pdfDoc, [pageIndex]); newDoc.addPage(copiedPage); const pdfBytes = await newDoc.save(); const filepath = path.join(os.tmpdir(), `fileTemp_page_${pageIndex}.pdf`); fs.writeFileSync(filepath, pdfBytes); tempFiles.push(filepath); } catch (error) { console.error('Error in processPage:', error); } } async function splitExcelFile(inputFile) { let numRows; // this parameter is needed when we need to analyze the pdf file in case it has duplicate pages try { console.log(`Loading excel file: ${ inputFile }\n`); const workbook = new ExcelJS.Workbook(); await workbook.xlsx.readFile(inputFile); const worksheet = workbook.getWorksheet(1); const headerRow = worksheet.getRow(1).actualCellCount; //number of columns of the table numRows = (worksheet.actualRowCount) - 1; // since it counts the row that represents the header, I need to subtract -1 // to get the actual number of rows that contains the data return numRows; } catch (error) { console.error('Error processing the excel file: ', error); } } const getCodiceFiscale = (pageText) => { if (!pageText) return null; const match = pageText.match(/Codice fiscale:\s*([A-Za-z0-9]+)/i); return match ? match[1].toUpperCase() : null; // match[1] is the captured code }; async function splitPDF(numRows, inputFile, tempFiles) { console.log(`Loading pdf file: ${ inputFile }...\n`); let parser; try { // 1. we use pdfParse to read and extract the content of a pdf file const existingPdfBytes = fs.readFileSync(inputFile); parser = new PDFParse({ data: existingPdfBytes }); const result = await parser.getText(); const pagesNumber = result.pages.length; //pages number of the pdf file // 2. we use pdf-lib to create and edit a pdf file const srcPdf = await PDFDocument.load(existingPdfBytes); // if there are duplicates, I only have to print one copy if (pagesNumber > numRows) { // comparison between the number of pages of the pdf file and those of the excel file let index = 1; for (let i = 0; i < pagesNumber; i++) { // we use optional chaining ?. so that when we encounter null or undefined values, the code doesn't crash const prevCF = getCodiceFiscale(result.pages[i + 1]?.text); const currentCF = getCodiceFiscale(result.pages[i].text); if (prevCF && prevCF === currentCF) { continue; } await processPage(srcPdf, i, tempFiles); index++; } } else { for (let i = 0; i < pagesNumber; i++) { await processPage(srcPdf, i, tempFiles); // Pass 0-based index i } } return tempFiles; } catch (error) { console.error(`Error processing the pdf file: `, error); } finally { if (parser) { await parser.destroy(); } } } // cleans up temporary files from OS temp directory function cleanUp(tempFiles) { tempFiles.forEach((file) => { try { if (fs.existsSync(file)) { fs.unlinkSync(file); } } catch (err) { console.error(`Failed in deleting the temporary file ${ file }: `, err); } }) } async function main() { const isMyFlag = process.argv.includes('--myFlag') || process.env.MY_FLAG === 'true'; const isDryRun = process.argv.includes('--dryRun') || process.env.DRY_RUN === 'true'; if (isMyFlag) { console.log(`--- Test di attachPdfToAccount ${isDryRun ? '(MODALITÀ DRY RUN)' : ''} --- \n`); let excelFile = process.argv[2]; let pdfFile = process.argv[3]; if (!pdfFile || !excelFile) { console.error('Errore: Fornisci il percorso di un file PDF per il test.'); process.exit(1); } const tempFilePath = []; try { let resultExcel = await splitExcelFile(excelFile); let resultPdf = await splitPDF(resultExcel, pdfFile, tempFilePath); await initializeSalesforceConnection(); await attachPdfToAccount(sfdc, resultPdf, isDryRun); } catch (err) { console.error('An error occurred in test mode: ', err); } finally { cleanUp(tempFilePath); } return; } } main();