testing.py

import json
import math
from openpyxl import Workbook
import os


column_dict = ['registrationno.B','regn.noB',
    'regndt:C','reg.dt:C','regdt:C',
    'chassisno.F','ch.noF','chnoF','chasisno.F',
    'engineno.E','enoE','engineE',
    'name&addressA','owner\'snameA','nameA','name.A',
    'mfg.dt.D','monthandyearofmfg.D','month/yrofD','mfgdtD','month/yearofmanufactureD']

def distance(x1 , y1 , x2 , y2): 
    return math.sqrt(math.pow(x2 - x1, 2) +
                math.pow(y2 - y1, 2) * 1.0) 


def lineFromPoints(P,Q): 
    a = Q[1] - P[1] 
    b = P[0] - Q[0]  
    c = -(a*(P[0]) + b*(P[1]) )    
    return a,b,c

def shortest_distance(x1, y1, a, b, c):  
       
    d = abs((a * x1 + b * y1 + c)) / (math.sqrt(a * a + b * b)) 
    return d


def get_line(P,Q,word_size,text_annotations,word_set,the_word):
    
    lines = []
    a,b,c = lineFromPoints(P,Q)
    for word_desc in text_annotations:
        try:
            x1 = word_desc['bounding_poly']['vertices'][0]['x']
            y1 = word_desc['bounding_poly']['vertices'][0]['y']
            x2 = word_desc['bounding_poly']['vertices'][3]['x']
            y2 = word_desc['bounding_poly']['vertices'][3]['y']
            x3 = word_desc['bounding_poly']['vertices'][1]['x']
            x2 = (x1+x2)/2
            y2 = (y1+y2)/2

            # if(the_word == 'E' and word_desc['description'] == 'COLOUR'):
            #     print(shortest_distance(x2,y2,a,b,c))
            if shortest_distance(x2,y2,a,b,c)<=(word_size):
                word_set.add(word_desc['description'].lower()+str(x1)+str(y1))
                lines.append([x1,word_desc['description'],x3])
        except:
            pass


    lines = sorted(lines, key=lambda x: x[0])
    line=[]
    for i,dei in enumerate(lines):
        try:
            line.append([dei[1],lines[i+1][0]-dei[2]])
        except:
            line.append([dei[1],0])
    return line



def parse_it(line,i,cell_col):    
    
    if(i>=len(line)):
        return '-1'
    if line[i][0]==':':
        i=i+1
    if(cell_col=='C' or cell_col=='D'):
        return line[i][0]
    retstr =''
    while True:
        retstr+=line[i][0]+' '    
        if i+1>=len(line) or line[i][1]>10:
            break
        
        i+=1
    if(cell_col == 'B' and len(retstr)<9):
        retstr+=line[i+1][0]
    return retstr


def parse_month(line,i):
    
    while(i<len(line) and line[i][0][0].isdigit()==False):      
        i+=1
    if(i==len(line)):
        return'-1'
    
    return line[i][0]


def check_match(line,i,col_name):
    cell_col = col_name[-1]
    col_name = col_name[:-1]
    for it in range(i,len(line)):
        if(line[it][0].lower() == 'regn'):
            print((line[it][0] + line[it+1][0]).lower(),col_name)
        if(col_name == line[it][0].lower()):
                 
            return parse_it(line,it+1,cell_col)
        elif it+1<len(line) and  col_name == (line[it][0] + line[it+1][0]).lower():            
            return parse_it(line,it+2,cell_col)
        elif it+2<len(line) and  col_name == (line[it][0] + line[it+1][0] + line[it+2][0]).lower():
            print(col_name)   
            return parse_it(line,it+3,cell_col)
        elif (line[it][0].lower() == 'month' or line[it][0].lower()=='month/') and col_name.startswith('month'):
            return parse_month(line,it+1)
    return '-1'

def save_to_sheet(jsons):
    book = Workbook()
    sheet  = book.active
    row_number=0
    lines_to_ret=[]
    current_line=['Name','Registration No','Registration Date','Manufacturing Date','Engine No','Chassis No.']
    lines_to_ret.append(current_line)
    for img_dict in jsons:    
            row_number+=1            
            word_set = set()
            text_annotations = img_dict['text_annotations'][1:]

            i=0
            lines=[]
            for word_desc in text_annotations:    ## Find the lines     
                try:
                    x1 = word_desc['bounding_poly']['vertices'][0]['x']
                    y1 = word_desc['bounding_poly']['vertices'][0]['y']        
                    if word_desc['description'].lower()+str(x1)+str(y1) not in word_set:
                            the_word = word_desc['description']              
                            x2 = word_desc['bounding_poly']['vertices'][1]['x']
                            y2 = word_desc['bounding_poly']['vertices'][1]['y'] 
                            x3 = word_desc['bounding_poly']['vertices'][3]['x'] 
                            y3 = word_desc['bounding_poly']['vertices'][3]['y']
                            x4 = word_desc['bounding_poly']['vertices'][2]['x'] 
                            y4 = word_desc['bounding_poly']['vertices'][2]['y'] 
                            x1 = (x1+x3)/2
                            y1 = (y1+y3)/2
                            x2 = (x2+x4)/2
                            y2 = (y2+y4)/2
                            word_size = distance(x1,y1,x3,y3)                        
                            lines.append(get_line([x1,y1],[x2,y2],word_size,text_annotations,word_set,the_word))
                            
                except:
                    pass
            current_line=[None,None,None,None,None,None]
            for line in lines: #Saving to excel
                
                for i,word in enumerate(line):        
                    for col_name in column_dict:
                        cell_insert = check_match(line,i,col_name)
                        if(cell_insert == '-1'):
                            continue
                        else:                    
                            
                            ins_col = ord(col_name[-1]) - ord('A') + 1
                            current_line[ins_col-1] = cell_insert
                            if(sheet.cell(column=ins_col,row=row_number).value == None):
                                sheet['{}{}'.format(col_name[-1],row_number)] = cell_insert
                            break            
                    
            lines_to_ret.append(current_line)
    return lines_to_ret