Quick & Dirty example of parsing a bank statement pdf (of a german bank) and converting it to csv

bank_statement_pdf_parser.js

const getCategory = function(addresseeClient, eventReference, postingText){

  if (/Auszahlung|Geldautomat|GAA /i.test(addresseeClient)){
    return "cash";
  }

  return "none";
}

pdfs.forEach( function (pdfFilename) {

  pdfParser.pdf2json(pdfFilename, function (error, pdf) {
      if(error != null){
          console.log(error);
      }else{

        pdf["pages"].forEach(function(page){
          let data = initDataStruct();

          page["texts"].forEach(function(text){
            var posX = text["left"] ;
            var textBlock =  text["text"];
            let type = undefined;
            //console.log(posX + " " +textBlock);

            if (posX >= 50 && posX < 80 &&  /\d\d\.\d\d\.\d{4}/.test(textBlock)){
              type = "col1";
            }

            if (posX >= 80 && posX < 150){
              type = "col2";
            }

            if (posX>= 150 && posX < 300){
              type = "col3";
            }

            if (posX >= 300 && posX < 510){
              type = "col4";
            }

            if (type){
              data[type].push(textBlock);
            }

            if (posX > 510){
              type = "col5";

              if (data["col1"][0]){
                let postingDate  = data["col1"][0];
                let valueDate = data["col1"][1] || "";
                let eventReference = '"' +  data["col2"].join(" ") + '"';
                let addresseeClient = '"'+data["col3"].join(" ") + '"';
                let postingText ='"' + data["col4"].join(" ") + '"';
                let balance = textBlock.replace("+", '').replace('.', '').replace(",", '.');

                let csvCols = [postingDate, valueDate , getCategory(addresseeClient, eventReference, postingText) ,eventReference , addresseeClient, postingText,  balance];

                console.log(csvCols.join(';'));
              }
              //console.log(data);
              data = initDataStruct();
            }

          });
        });
      }
  });
});

package.json

{
  "name": "xxx",
  "version": "1.0.0",
  "main": "bank_statement_pdf_parser.js",
  "license": "MIT",
  "scripts": {
    
  },
  "dependencies": {
    "glob": "^7.1.2",
    "pdf-parser": "^1.0.4"
  }
}