from numpy import array
from pandas import to_datetime, DataFrame
import csv
import os
from sys import getsizeof
from versuche.helper import normalice_header


def detect_tabnum(filename, tabstr,encoding='utf-8'):
    filename = os.path.normpath(filename)

    tabstr = tabstr.lower()

    #Einlesen
    with open(filename,'r', encoding=encoding) as inFile:
        reader = csv.reader(inFile, delimiter='\t')
        counter = 0
        for row in reader:

            row = [r.lower() for r in row]
            if any(tabstr in mystring for mystring in row):
                if 'plain' in row:
                    return row[1]

            counter += 1

            if counter>100:
                return False



def str2float(str):
    try:
        str = str.replace(',','.')
        return float(str)
    except:
        return None


def read_data(filename,
              table,
              pkdata = '001',
              encoding='utf-8',
              to_si = False,
              debug=False):
    '''

    :param filename: File-Name
    :param table: Table-Number
    :param pkdata: Table-Number of speciment definitions, default: 1
    :param encoding: Encoding, default: utf-8
    :param debug: debug-mode
    :return:

    '''

    #print('start read GEOSYS')
    filename = os.path.normpath(filename)

    try:
        dictOut = {}
        dictOut['durch'] = 0
        dictOut['hoehe'] = 0

        #---------------------------------------------------------------------
        #Daten einlesen und umwandeln
        #---------------------------------------------------------------------


        data = []
        zuordnung = []

        #Einlesen
        with open(filename,'r', encoding=encoding) as inFile:
            reader = csv.reader(inFile, delimiter='\t')
            for row in reader:
                if len(row) > 2:
                    data.append(row)

        if debug:
            print('Anz. Datensätze: ', str(len(data)), getsizeof(data))
        #aufräumen
        ##Datenstruktur anlegen


        data_clean = {}
        data_clean['head'] = []
        data_clean['data'] = []

        for idx, d in enumerate(data):
            try:
                v = d[0][0:3]
                if v in pkdata: data_clean['head'].append(d)
                if v in table: data_clean['data'].append(d)

            except:
                pass

        # aufräumen
        data = data_clean
        del(data_clean)
        if debug:
            print('data_clean fin')

        ## Header aufbereiten

        for idx, row in enumerate(data['head']):
            #print(idx,row)
            if idx == 0:
                id_durchmesser = None
                id_hoehe = None
                id_name = None

                for idx_name, name in enumerate(row):
                    if name in [r'Probekörberdurchmesser',r'Diameter of specimen', 'PK-Durchmesser', 'Probekörper-Durchmesser']:
                        id_durchmesser = idx_name
                    elif name in [r'Probekörperbezeichnung']:
                        id_name = idx_name
                    elif name in ['Probekörperhöhe','Gap length','PK-Höhe']:
                        id_hoehe = idx_name

                if debug:
                    print(id_durchmesser, id_hoehe, id_name)
            elif idx == 1:
                unit_durch = None
                unit_hoehe = None

                try:
                    unit_durch = row[id_durchmesser]
                    unit_hoehe = row[id_hoehe]
                except:
                    pass

            elif idx == 2:
                durchmesser = None
                hoehe = None
                name = None
                try:
                    durchmesser = str2float(row[id_durchmesser])
                    hoehe = str2float(row[id_hoehe])
                    name = row[id_name]

                except:
                    pass

        header = {'d': durchmesser,
                  'h': hoehe,
                  'name': name,
                  'unit_h': unit_hoehe,
                  'unit_d': unit_durch}

        if debug:
            print('header\n', header)

        ## Daten in Pandas DataFrame umwandeln
        if debug:
            print('daten umwandel')

        temp = []
        for idx, row in enumerate(data['data']):
            if idx == 0:
                if debug:
                    print('head')
                data_head = []
                for idx_name, name in enumerate(row):
                    if idx_name <= 1: continue
                    data_head.append(name)
            elif idx == 1:
                data_units = []
                for idx_name, name in enumerate(row):
                    if idx_name <= 1: continue
                    data_units.append(name)
            else:
                t = []
                for idx_col, value in enumerate(row):
                    if idx_col <= 1:
                        continue
                    else:
                        t.append(str2float(value))

                temp.append(t)

        data = array(temp)

        if debug:
            print(data_head,data_units)

        ## Bezeichnungen der Daten normalisieren
        data_head = normalice_header(data_head)

        # Pandas DataFrame erstellen
        data = DataFrame(data=data, columns=data_head)
        if debug:
            print(data.head())

        data = data.set_index('t')

        #data._units = data_units

        # Zykelnzähler anpassen
        if 'N' in data.columns:
            data['N'] = data['N'].astype(int)

        # Daten sortieren
        data.sort_index()

        # Index normieren
        data.index = data.index - data.index[0]

        return header, data


    except:
        print('Fehler beim lesen')
        raise