315 lines
12 KiB
JavaScript
315 lines
12 KiB
JavaScript
const ExcelJS = require('exceljs');
|
|
const { PDFParse } = require('pdf-parse');
|
|
const { PDFDocument } = require('pdf-lib');
|
|
const fs = require('fs');
|
|
const os = require('os');
|
|
const path = require('path');
|
|
const sfdc = require('./sfdc2');
|
|
require('dotenv').config();
|
|
let excelFile = process.argv[2];
|
|
let pdfFile = process.argv[3];
|
|
// excel1 => /Users/ddatinguinoo/Downloads/Estratto IPP 5 Sic.xlsx
|
|
// excel2 => /Users/ddatinguinoo/Downloads/Estratto Haccp IPP 5 (1).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: ${loginUrl}`);
|
|
});
|
|
} else {
|
|
console.log('Token loaded successfully.');
|
|
}
|
|
// Ensure the access token is valid and refresh if necessary
|
|
await sfdc.checkToken();
|
|
console.log('Salesforce connection is ready.');
|
|
}
|
|
|
|
// Escape single quotes for SOQL queries (e.g., D'Angelo -> D\'Angelo)
|
|
function escapeSoql(str) {
|
|
return str ? str.replace(/'/g, "\\'") : '';
|
|
}
|
|
|
|
// the pdf files contain the identifying ids of the users such as 'Codice Fiscale' (CF) and we use them to attach
|
|
// the page/pages to the specific account that has that id. If CFs are not available we search based on the
|
|
// name an lastname of the user
|
|
async function attachPdfToAccount(sfdc, tempFiles) {
|
|
let parser;
|
|
try {
|
|
let contentVersion;
|
|
for (let i = 0; i < tempFiles.length; i++) {
|
|
// 1. Read the local PDF file and convert it to Base64
|
|
const pdfBuffer = fs.readFileSync(tempFiles[i]);
|
|
const base64Pdf = pdfBuffer.toString('base64');
|
|
parser = new PDFParse({ data: pdfBuffer });
|
|
try {
|
|
const result = await parser.getText();
|
|
let conditions = [];
|
|
|
|
const userID = result.text.match(/Codice fiscale:\s*([A-Za-z0-9]+)/i);
|
|
const codiceFiscale = userID ? userID[1].trim() : null;
|
|
if (codiceFiscale) {
|
|
conditions.push(`CF__c='${ 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(' ');
|
|
|
|
//conditions.push(`Name='${fullName}'`);
|
|
if (fullName !== reversedName) {
|
|
conditions.push(`Name='${ escapeSoql(reversedName) }'`);
|
|
}
|
|
}
|
|
conditions.push(`Name='${ escapeSoql(rawName) }'`);
|
|
}
|
|
|
|
// use the soql command to search in the database
|
|
// if codice fiscale is not available, I search the user's account based on their name and lastname
|
|
const soql = `SELECT Id, Name FROM Account WHERE (${conditions.join(' OR ')}) ORDER BY CreatedDate DESC LIMIT 1`;
|
|
const result_soql = await sfdc.query(soql);
|
|
if (result_soql && result_soql.records && result_soql.records.length > 0) { // if we find something
|
|
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: result_soql.records[0].Id// Automatically creates the link to the Account!
|
|
});
|
|
}
|
|
}
|
|
finally {
|
|
if (parser) {
|
|
await parser.destroy();
|
|
}
|
|
}
|
|
}
|
|
console.log(`PDFs successfully attached!`);
|
|
|
|
} catch (error) {
|
|
console.error(`Error uploading PDF: `, error);
|
|
}
|
|
}
|
|
|
|
// here we create an Account for each client and upload their pdf file
|
|
async function uploadUserAccount(data) {
|
|
await initializeSalesforceConnection();
|
|
|
|
// now for each client we need to create an Account on Salesforce Sandbox
|
|
try {
|
|
for (let i = 0; i < data.length; i++) {
|
|
const firstName = (data[i].NOME || '').toString().trim();
|
|
const lastName = (data[i].COGNOME || '').toString().trim();
|
|
|
|
let userData = {
|
|
Name: `${ firstName } ${ lastName }`.trim(),
|
|
CF__c: data[i].CODICE_FISCALE || null,
|
|
};
|
|
const newAccount = await sfdc.create('Account', userData); // it creates the Account
|
|
}
|
|
console.log('Accounts created successfully!');
|
|
}
|
|
catch (error) {
|
|
console.log('Something went wrong while creating the accounts: ', error);
|
|
}
|
|
}
|
|
// this is where i need to store each page in a temporary pdf file
|
|
async function processPage(pdfDoc, pageIndex, tempFiles) {
|
|
try {
|
|
const newDoc = await PDFDocument.create();
|
|
const [ copiedPage ] = await newDoc.copyPages(pdfDoc, [pageIndex]);
|
|
newDoc.addPage(copiedPage);
|
|
const pdfBytes = await newDoc.save();
|
|
|
|
const filepath = path.join(os.tmpdir(), `fileTemp_page_${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
|
|
const columnNames = []; // contains the names of the columns (header names)
|
|
const data = []; // contains the info of all the clients
|
|
|
|
try {
|
|
console.log(`Loading excel file: ${ inputFile }\n`);
|
|
|
|
const workbook = new ExcelJS.Workbook();
|
|
await workbook.xlsx.readFile(inputFile);
|
|
const worksheet = workbook.getWorksheet(1);
|
|
|
|
const headerRow = worksheet.getRow(1).actualCellCount; //number of columns of the table
|
|
numRows = (worksheet.actualRowCount) - 1;
|
|
|
|
worksheet.getRow(1) // to convert every title of the first row to upper case
|
|
.eachCell((col, col_num) => {
|
|
if (typeof col.value === 'string') {
|
|
col.value = col.value.toUpperCase().trim();
|
|
}
|
|
})
|
|
|
|
// convert every letter of CF to uppercase and store the CFs in the array cf
|
|
if (worksheet.getRow(1).getCell(6).value.trim() === 'CODICE FISCALE') {
|
|
worksheet.getColumn(6)
|
|
.eachCell((row, row_num) => {
|
|
row.value = (row.value).toString().toUpperCase();
|
|
})
|
|
}
|
|
|
|
worksheet.getColumn(8) // of the first file
|
|
.eachCell((cell, rownumber) => {
|
|
if (typeof cell.value === 'string') {
|
|
cell.value = (cell.value).replace(/\b\w/g, char => char.toUpperCase());
|
|
/* \b finds the start of a word (the boundary of a word)
|
|
\w targets the first letter following that boundary
|
|
/g (global flag) ensures it updates every word
|
|
*/
|
|
}
|
|
})
|
|
|
|
// to store the names of the columns in columnNames
|
|
worksheet.getRow(1).eachCell({ includeEmpty: true }, (cell, cell_num) => {
|
|
columnNames.push(cell.value);
|
|
})
|
|
|
|
// now we need to store the data of each user
|
|
for (let i = 2; i <= numRows + 1; i++) {
|
|
let dataUser = {};
|
|
worksheet.getRow(i)
|
|
.eachCell({ includeEmpty: true }, (cell, cell_num) => {
|
|
const rawColumnName = columnNames[cell_num - 1] || `column_${cell_num}`;
|
|
|
|
// 1. Trim whitespace and replace spaces with underscores
|
|
const cleanKey = rawColumnName
|
|
.toString()
|
|
.trim()
|
|
.replace(/\s+/g, '_'); // \s => matches any whitespaces
|
|
// + => groups multiple consecutive spaces together into a
|
|
// single match so you don't end up with multiple underscores.
|
|
|
|
dataUser[cleanKey] = cell.value;
|
|
})
|
|
data.push(dataUser);
|
|
}
|
|
return { numRows, data };
|
|
}
|
|
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() {
|
|
if (!excelFile || !pdfFile) {
|
|
console.error('Error: Please provide both Excel and PDF paths.\nUsage: node script.js <path-to-excel> <path-to-pdf>');
|
|
process.exit(1);
|
|
}
|
|
|
|
const tempFilePath = []; // contains the temporary files
|
|
let resultExcel, resultPdf;
|
|
try {
|
|
resultExcel = await splitExcelFile(excelFile);
|
|
resultPdf = await splitPDF(resultExcel.numRows, pdfFile, tempFilePath);
|
|
await initializeSalesforceConnection();
|
|
await attachPdfToAccount(sfdc, resultPdf);
|
|
}
|
|
catch (err) {
|
|
console.error('An error occurred in main: ', err);
|
|
}
|
|
finally {
|
|
cleanUp(tempFilePath);
|
|
}
|
|
}
|
|
|
|
main(); |