import csv
import os
from io import BytesIO
from sys import getsizeof

from numpy import array
from pandas import DataFrame


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_geosys(buffer: BytesIO,
                table,
                pkdata='001',
                metadata_ids=['003', '015'],
                encoding='utf-8',
                to_si=False,
                debug=False):
    '''

    :param buffer: Bytes IO Object
    :param table: Table-Number
    :param pkdata: Table-Number of speciment definitions, default: 1
    :param encoding: Encoding, default: utf-8
    :param debug: debug-mode
    :return:

    '''

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

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


        #Einlesen
        buffer.seek(0)

        lines = buffer.readlines()
        
        data = []
        
        for line in lines:
            try:
                line = line.decode(encoding)
                line = line.split('\t')

                if len(line) > 2:   
                    v = line[0][0:3]
                    if len(v) == 3:
                        if (table == v) or (pkdata == v) or (v in metadata_ids):
                            data.append(line)
            except:
                pass
        
        if debug:
            print('Anz. Datensätze: ', str(len(data)), getsizeof(data))
            
        #aufräumen
        ##Datenstruktur anlegen

        data_processed = {}
        data_processed['head'] = []
        data_processed['metadata'] = {}
        data_processed['data'] = []

        for i in metadata_ids:
            data_processed['metadata'][i] = []

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

            except:
                pass

        # replace object
        data = data_processed
        
        assert len(data['data']) != 0

        if debug:
            print('data_clean fin')
            
        ## Header aufbereiten

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

                for idx_name, name in enumerate(row):
                    name_lower = name.lower()

                    if any(map(name_lower.__contains__, ['durchmesser'])):
                        id_durchmesser = idx_name

                    elif any(map(name_lower.__contains__, ['bezeichnung'])):
                        id_name = idx_name

                    elif any(map(name_lower.__contains__, ['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 = {
            'speciment_diameter': durchmesser,
            'speciment_height': hoehe,
            'name': name,
            'unit_h': unit_hoehe,
            'unit_d': unit_durch
        }

        meta = data['metadata']
        
        for key in meta.keys():
            sel = meta[key]

            assert len(sel[0]) == len(sel[2])
            
            if len(sel) <= 3:
                d = { sel[0][i]: sel[2][i].strip() for i in range(len(sel[0])) }
            # Fix: In Geosys gibt es den Parameter Oberspannung zweimal. Erster entfernen
            else:    
                d = { sel[0][i]: sel[3][i].strip() for i in range(len(sel[0])) }
                
            header_append = d

            header.update(header_append)
            
        #Fix Frequenz: Ich muss dies in den Eingangsdaten der TUD anpassen
        try:
            l = 'Versuchsart\r\n'
            
            header['Frequenz'] = float(header[l].split('Hz')[0].split('Steifigkeit')[1].strip().replace(',','.'))
            
        except:
            pass

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

        # add metadata to 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('convert 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
        # Pandas DataFrame erstellen
        data = DataFrame(data=data, columns=data_head)
        if debug:
            print(data.head())

        return header, data

    except:
        print('Fehler beim lesen')
        raise