import calendar
import os.path
from pypdf import PdfReader
from pypdf.errors import PdfReadError
import sys
import logging
from datetime import datetime
import csv

logger = logging.getLogger(__name__)
ch = logging.StreamHandler()
ch.setLevel(logging.DEBUG)
stream_formatter = logging.Formatter('%(levelname)s - %(message)s')
ch.setFormatter(stream_formatter)
logger.addHandler(ch)
logging.basicConfig(filename='log.txt', encoding='utf-8', format='%(asctime)s:%(levelname)s:%(message)s',
                    datefmt='%m/%d/%y %H:%M:%S', level=logging.DEBUG)


# class paycheck:
#
#     def __init__(self):
#         # TODO - add variables to be returned or obtained in code
#         pass
#
#     pass


def loadfiles(pointer, folder):
    # Take a file or a folder as an input and load all applicable filenames into a list
    # return a list
    list_files = []
    if folder:  # input is a folder
        logger.debug("Loading folder: %s", os.path.basename(pointer))
        for file in os.listdir(pointer):
            list_files.append(os.path.join(pointer, file))
    elif not folder:  # input is a file
        list_files.append(pointer)
        logger.debug("Loading file: %s", os.path.basename(pointer))

    return list_files


def checkargs(args):
    list_file = []
    list_folder = []
    for arg in args[1:]:
        if os.path.isfile(arg):
            list_file.append(arg)
        elif os.path.isdir(arg):
            list_folder.append(arg)
        if arg == "-debug":
            global debug
            debug = True

    # print(dict_file["folder"])
    if not len(list_file) == 1 and not len(list_folder) == 1:
        raise Exception("Error parsing filename or folder name.")
    if len(list_file) == 1 and len(list_folder) == 1:
        raise Exception("Both filename and folder provided.")

    if list_file:
        files = loadfiles(list_file[0], False)
    elif list_folder:
        files = loadfiles(list_folder[0], True)
    else:
        raise Exception("Error sending file or folder to loadfiles.")

    logging.debug('checkargs complete with no errors')

    return (files)


def paycheckv2_parser(lines, filename):
    ###  This will parse paychecks using the post Jul 2022 formar
    keywords = ['Current', 'BANK OF AMERICA', 'Earnings', 'TFP']
    dict_result = {}
    dict_tfp = {}

    dict_result['Filename'] = filename
    dict_result["Check Date"] = datetime.strptime(lines[2].split()[8], "%m/%d/%Y")
    dict_result["ParserVersion"] = "V2"

    for line in lines:
        # print(line)
        dict_result["month"] = dict_result["Check Date"].strftime('%m')
        if line.find(keywords[0]) == 0:
            dict_result["Gross Pay"] = line.split()[1]
        if line.find(keywords[1]) >= 0:
            dict_result["BoA Deposit"] = line.split()[6]
        if line.find(keywords[2]) == 0 and len(line.split()) > 1:
            dict_result["Earnings"] = line.split()[1]
        if line.find(keywords[3]) >= 0 and not line.find('TFP/Hrs') >= 0:
            ## Only displays for second paycheck of the period
            ## check if third from last entry is a number of some sort.  This shows it is a current pay item
            tfp_line = line.split()
            if len(tfp_line) >= 5:
                if tfp_line[-3].find('.') >= 0 and not tfp_line[-1] == "Taxes":
                    desc = ' '.join(tfp_line[:-3])
                    amt = tfp_line[-3][:-6]
                    dict_tfp[desc] = amt
                    dict_result["TFP"] = dict_tfp
            if tfp_line[-1] == "Taxes" and tfp_line[-4].find('.') >= 0:
                desc = tfp_line[0]
                amt = tfp_line[-4][:-6]
                dict_tfp[desc] = amt
                dict_result["TFP"] = dict_tfp

    total_tfp = float(0)
    for tfp in dict_tfp.items():
        total_tfp += float(tfp[1])
    dict_result["Total TFP"] = round(total_tfp, 2)

    # print(dict_result)

    return dict_result


def paycheckv1_parser(lines, filename, linetwo):
    # Check Date
    # Gross Pay
    # BoA Deposit
    # earnings
    # TFP

    dict_result = {}
    dict_tfp = {}

    dateline_num = 2
    grosspayline_num = 4

    if linetwo == "45":
        dateline_num += 1
        grosspayline_num += 1

    # print(lines[3])
    dateline = datetime.strptime(list(filter(None, lines[dateline_num].split(' ')))[5], '%m/%d/%Y').month
    # print(dateline)
    earningsline = list(filter(None, lines[grosspayline_num].split(' ')))
    # print(earningsline)

    # print(filename, earningsline[earningsline.index('=')+1])
    dict_result['Filename'] = filename
    dict_result["Check Date"] = datetime.strptime(list(filter(None, lines[dateline_num].split(' ')))[5], "%m/%d/%Y")
    dict_result['month'] = datetime.strptime(list(filter(None, lines[dateline_num].split(' ')))[5], "%m/%d/%Y").month
    dict_result["ParserVersion"] = "V1"
    dict_result["Gross Pay"] = list(filter(None, lines[grosspayline_num].split(' ')))[0]
    dict_result["Earnings"] = earningsline[earningsline.index('=') + 1]
    dict_result["BoA Deposit"] = float(0.0)

    start = 0
    end = 0

    for i, line in enumerate(lines):
        # print(line)
        if line.find("BANK OF AMERICA") >= 0:
            dict_result["BoA Deposit"] = list(filter(None, line.split(' ')))[-1]
        if line.find("Gross") >= 0 and line.find("Hrs") >= 0:
            # print("START")
            start = i
        if line.find('Total Gross Pay') >= 0:
            # print("END")
            end = i

    for line in lines[start + 1:end - 1]:
        split_line = line.split(' ')
        desc = (line[:26].strip())
        # print(desc)
        dataline = line[26:].split(' ')
        tfp = 0
        if len(dataline[0]) > 0 and not desc == "Pilot Per Diem DAY":
            tfp = float(dataline[0])

        # print(desc, tfp)

        if tfp > 0:
            dict_tfp[desc] = tfp

    dict_result["TFP"] = dict_tfp

    total_tfp = float(0)
    for tfp in dict_tfp.items():
        total_tfp += float(tfp[1])
    dict_result["Total TFP"] = round(total_tfp, 2)

    # print(dict_result)

    return dict_result


def calculate_result(list_result):
    if not type(list_result) is list:
        raise Exception("calculate_result requires a list of dictionaries, even if one item.")

    if len(list_result) > 1:
        list_result = consolodateMonths(list_result)

    result = []

    logging.debug('Length of calculate_result results:' + str(len(result)))

    for resulty in list_result:
        # print(resulty)
        if resulty['Total BoA'] > 0 and resulty['Total Earnings'] > 0:
            try:
                resulty['BoA to Earnings'] = resulty['Total BoA'] / resulty['Total Earnings']
                resulty['BoA to TFP'] = resulty['Total BoA'] / resulty['Total TFP']
                resulty['Total Deductions'] = resulty['Gross Pay'] - resulty['Total Earnings']
            except ZeroDivisionError:
                logging.warning("Zero Division Error")
            except:
                logging.warning('Unable to calculate totals.  Something went wrong.')
        result.append(resulty)

    logging.debug('calculate_result complete')

    return(result)


def consolodateMonths(list_results):
    # Takes list with all data and builds a new list with all paychecks for each month combined and totaled
    list_monthdata = []

    # print(list_results)
    # print(datetime.strftime(list_results[0]['Check Date'], '%Y'))

    for month in range(1, 13):
        monthdata = {}
        monthdata['Gross Pay'] = 0
        monthdata['Total TFP'] = 0
        monthdata['Total BoA'] = 0
        monthdata['Total Earnings'] = 0
        monthdata['Month'] = int(month)
        monthdata['Year'] = datetime.strftime(list_results[0]['Check Date'], '%Y')
        for i, check in enumerate(list_results):
            if int(check['Check Date'].strftime('%m').lstrip('0')) == month:
                # print(check)
                monthdata['Gross Pay'] += float(check['Gross Pay'].replace(',', ''))
                monthdata['Total TFP'] += float(check['Total TFP'])
                if check['BoA Deposit'] != float(0.0):
                    monthdata['Total BoA'] += float(check['BoA Deposit'].replace(',', ''))
                monthdata['Total Earnings'] += float(check['Earnings'].replace(',', ''))

        list_monthdata.append(monthdata)

    # print(list_monthdata)

    return list_monthdata


def parse_pdf(filename):  # Takes only one fielname, not a list!

    if type(filename) == list:
        raise Exception("parse_pdf accepts only one input, not a list")

    list_result = []

    # for file in filename:
    templines = []
    #print(file)
    #Test if file is valid pdf file

    try:
        reader = PdfReader(filename)
    except PdfReadError:
        raise Exception(filename, "is not a valid PDF file.")

    pages = reader.pages
    # print(pages)

    for i in range(len(pages)):
        templines.append(pages[i].extract_text().split('\n'))
    if len(pages) == 1:
        lines = templines[0]
    elif len(pages) == 2:
        lines = templines[0] + templines[1]
    else:
        raise Exception('Paycheck', i, ' appears to be greater than 2 pages.  Unable to Parse')


    # If lines[1] == 81, then its post July 2022 (new format)
    # If lines[1] == about 121, then is pre July 2022 (old format)
    # print(os.path.basename(filename))
    # print(len(lines[1]), lines[1])
    #
    # Check for signs of paycheck versioning
    if len(lines[1]) == 81:
        result = paycheckv2_parser(lines, filename)
        logging.debug('Using Parser v2.')
    elif len(lines[1]) == 122:
        result = paycheckv1_parser(lines, filename, "122")
        logging.debug('Using Parser v1 with line 2 length 122.')
    elif len(lines[1]) == 45:
        result = paycheckv1_parser(lines, filename, "45")
        logging.debug('Using Parser v1 with line 2 lenght 45')
    else:
        logger.warning("Unable to determine paycheck format for %s. Check logs for details.  Line 2 length: %s" % (os.path.basename(filename), str(len(lines[1]))))
        raise Exception("Unable to determine paycheck format for %s. Check logs for details.  Line 2 length: %s" % (os.path.basename(filename), str(len(lines[1]))))

    logging.debug('ParsePDF Complete')

    return result


# Begin Application

logging.warning('***Starting Application***')	

files = checkargs(sys.argv)  # Get Arguments from Command Line, check user's request, and return a list of files to

result = []

for i, file in enumerate(files):
    if os.path.isfile(file):  # Ignore folders such as W2 or other statements
      print(i, file)
      result.append(parse_pdf(file))

logging.debug('results: ' +str( len(result)))

full_result = calculate_result(result)

for i in full_result:
    print(i)

file_name = full_result[0]['Year']

with open(f'{file_name}.csv', 'w',  newline='') as f:
    w = csv.writer(f)
    w.writerow(full_result[0].keys())
    for row in full_result:
        w.writerow(row.values())
    f.close