# ----------------------------------- # Imports des librairies # ----------------------------------- try: from PIL import Image from uuid import uuid4 import json except ImportError: import Image import pytesseract from pytesseract import Output import cv2 import re from pdf2image import convert_from_path from TSLibOCR import * import time import os import numpy as np from TS_OCR import * import MySQLdb from datetime import datetime from collections import Counter debug = False """ Retourne le temps passé depuis le lancement du programme """ def getTime(): return "" class bcolors: HEADER = '\033[95m' OKBLUE = '\033[94m' OKCYAN = '\033[96m' OKGREEN = '\033[92m' WARNING = '\033[93m' FAIL = '\033[91m' ENDC = '\033[0m' BOLD = '\033[1m' UNDERLINE = '\033[4m' class Charal(TS_OCR): sqlUser = 'toosmart' sqlHost = '' sqlPasswd = 'i5Hf7MSHKkPN1hy' sqlDb = '' idAbattoir = 27 linesPoids = [] linesLots = [] lineMerged = {} dateLivraison = "" typeLivraison = "" enAnomalie = 0 poidsTotal = 0.000 cleCoche = "TRUIE LBC VPF" ocrDpi = 350 ocrFolder = '/home/serge/www/toosmart_digicoche/data/tmp/' outputFileName = '' motsASurligner = [] dateTuerie = "" poidsTotalCalcule = 0.000 poidsTotalDetecte = 0 qttCocheCalcule = 0 qttCocheDetecte = 0 nonIntegrable = False mergedImage = [] def __init__(self, pdf, outputFileName,ocrDirectory,dbName,sqlHost): try: self.sqlDb = dbName self.sqlHost = sqlHost """ Connexion à la base de données MySQL de Digicoche (Eteko)""" self.conn = MySQLdb.connect(host=self.sqlHost, user=self.sqlUser, passwd=self.sqlPasswd, db=self.sqlDb) self.cursor = self.conn.cursor() self.pdf = pdf self.outputFileName = outputFileName self.ocrFolder = ocrDirectory self.printDebug("Connexion à la base de données OK "+self.sqlDb) self.printDebug("outputFileName = " + str(outputFileName)) except: self.printDebug("Erreur de connexion à la base de données "+self.sqlDb+ " "+self.sqlHost) def printDebug(self,pTxt): self._debugOCR = self._debugOCR + '
' + pTxt if debug: print(pTxt) # Analyse du BL de l'abattoir def read(self): # conversion en fichier jpeg try: imageAmelioreeSurlignee = [] # lecture pagesPdf = convert_from_path(self.pdf, self.ocrDpi) iPage = 0 ocrReturnArray = [] except Exception as err: self.printDebug(f"Unexpected {err=}, {type(err)=}") self.printDebug("ERROR") """ On parcours les pages du pdf """ for page in pagesPdf: try: # conversion de l'image en fichier jpeg image_name = "Page_" + str(iPage) + ".jpg" page.save(self.ocrFolder + 'Page_' + str(iPage) + '.jpg', "JPEG") # --------------------------------- # Optimisation de l'image img = cv2.imread(self.ocrFolder + 'Page_' + str(iPage) + '.jpg') result = img.copy() # On supprime les lignes horizontales gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) thresh = cv2.threshold(gray, 6, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1] horizontal_kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (40, 1)) remove_horizontal = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, horizontal_kernel, iterations=1) cnts = cv2.findContours(remove_horizontal, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE) cnts = cnts[0] if len(cnts) == 2 else cnts[1] for c in cnts: cv2.drawContours(result, [c], -1, (255, 255, 255), 1) # --------------------------------- # On lance la reconaissance OCR sur l'image df = pytesseract.image_to_string(np.array(result)) cv2.imwrite(self.ocrFolder + 'Page_' + str(iPage) + '_result.jpg', result) ocrReturnArray.append(df) # --------------------------------- imageAmelioreeSurlignee.append(Image.open(self.ocrFolder + 'Page_' + str(iPage) + '_result.jpg')) iPage = iPage + 1 except Exception as err: self.printDebug(f"Unexpected {err=}, {type(err)=}") self.printDebug("ERROR") self.printDebug('1 : ' + ocrReturnArray[0]) # Recherche du numéro du BL try: m = re.findall(r"BL[ ]?N°[ ]?([0-9]+)", ocrReturnArray[0], re.MULTILINE) if len(m) > 0: for element in m: self._numBl = element except Exception as err: self._numBl = "" self.printDebug(f"Unexpected1 {err=}, {type(err)=}") self.printDebug("ERROR1") if self._numBl != "": if self.blExist(): # Le BL Existe déjà en base de donneés print('{"blEnDoublon":1}') self.nonIntegrable = True return False # --------------------------------- # Recherche de la date de livraison m = re.findall(r"Livraison du ([0-9]{2}/[0-9]{1,2}/[0-9]{2})", ocrReturnArray[0]) if len(m) > 0: self.dateLivraison = str(m[0]) # --------------------------------- # Recherche du type de livraison m = re.search(self.cleCoche, ocrReturnArray[0]) if m is not None: self.typeLivraison = m.group() else: # BL Non intégrable car la clef n'est pas correcte # On retourne dans le bash l'information puis on quitte le script # (NE PAS SUPPRIMER CETTE LIGNE !!!) print('{"blNonIntegrable":1}') self.nonIntegrable = True return False # --------------------------------- # Recherche du nombre de coches livrées try: m = re.findall(r"^([0-9]+)\D?CR\D?([0-9,]+)$", ocrReturnArray[0],re.MULTILINE) if len(m) > 0: for element in m: self.qttCocheDetecte = round(int(element[0]) / 2) self.poidsTotalDetecte = element[1] except Exception as err: self.qttCocheDetecte = 0 self.printDebug(f"Unexpected1 {err=}, {type(err)=}") self.printDebug("ERROR1") # --------------------------------- j = 0 """ On parcours les pages du pdf """ for Sfile in ocrReturnArray: # Recherche du nombre de coches, du poids et du lot if self.typeLivraison != "": m = re.findall(r"(0000[0-9]{4})", Sfile) if len(m) > 0: self.qttCocheCalcule = len(m) self.linesLots.extend(m) m = re.findall(r"[0-9]{8}.*([0-9]{2}/[0-9]{1,2}/[0-9]{4})", Sfile) if len(m) > 0: self.dateTuerie = str(m[0]) # --------------------------------- # Gestion de l'image pour surlignage des zones try: tesseract_output = pytesseract.image_to_data(imageAmelioreeSurlignee[j],output_type=pytesseract.Output.DICT) self.surlignerZones(tesseract_output, j) except: self.printDebug("ERROR") j = j + 1 # --------------------------------- # Merge des fichiers PDF try: im = Image.open(self.ocrFolder + self.outputFileName + '-color-0.jpg') im.save(self.ocrFolder+self.outputFileName+"-BL.pdf", save_all=True, append_images=self.mergedImage) except: self.printDebug("ERROR") # --------------------------------- """ Surligne les zones détectées par l'OCR dans le fichier """ def surlignerZones(self, tesseract_output,iPage): try: imageASurligner = cv2.imread(self.ocrFolder + 'Page_'+str(iPage)+'_result.jpg') blockASurligner = [] except Exception as err: print(f"UnexpectedC {err=}, {type(err)=}") print("ERRORC") for i, level_idx in enumerate(tesseract_output['level']): if level_idx == 5: # if tesseract_output['text'][i] in self.motsASurligner: bbox = { 'x': 1 * tesseract_output['left'][i], 'y': 1 * tesseract_output['top'][i], 'width': 1 * tesseract_output['width'][i], 'height': 1 * tesseract_output['height'][i], 'rotation': 0, 'confidence': tesseract_output['conf'][i], 'text': tesseract_output['text'][i] } if int(tesseract_output['conf'][i]) >= 85: color = (42, 219, 151) elif int(tesseract_output['conf'][i]) >= 50: color = (0, 165, 255) else: color = (0, 0, 255) try: overlay = imageASurligner.copy() cv2.rectangle(overlay, (bbox['x'], bbox['y']), (bbox['x'] + bbox['width'], bbox['y'] + bbox['height']), color, cv2.FILLED) alpha = 0.5 imageASurligner = cv2.addWeighted(overlay, alpha, imageASurligner, 1 - alpha, 0) blockASurligner.append(bbox) except Exception as err: print(f"UnexpectedD {err=}, {type(err)=}") print("ERRORD") # On sauvegarde le fichier try: imageASurligner = cv2.cvtColor(imageASurligner, cv2.COLOR_BGR2RGB) img = Image.fromarray(imageASurligner, "RGB") width, height = img.size new_size = (width // 3, height // 3) resized_image = img.resize(new_size) if iPage > 0: self.mergedImage.append(resized_image) resized_image.save(self.ocrFolder + self.outputFileName + '-color-'+str(iPage)+'.jpg') except Exception as err: print("erreur") print(f"UnexpectedE {err=}, {type(err)=}") return True def mergeData(self): if self.nonIntegrable: return False try: # On stocke la date de la tuerue dans le format YYYYMMDD self.dateTuerie = datetime.strptime(self.dateTuerie, "%d/%m/%Y").strftime("%Y%m%d") except: self.dateTuerie = "" try: # On stocke la date de la tuerue dans le format YYYYMMDD self.dateLivraison = datetime.strptime(self.dateLivraison, "%d/%m/%y").strftime("%Y%m%d") except: self.dateLivraison = "" try: # On créé l'entête du BL dans eteko # Date de création de la ligne currentDate = datetime.now().strftime('%Y%m%d%H%M') self.poidsTotalCalcule = self.poidsTotalDetecte sql = ("INSERT INTO tier_att(id_base, id_type, id_tier,attdatcre, id_work, actif, newobj, att1, att2, att3,att6,att7,att8,att9,attf100,att17,att90) " "VALUES (1, 10, NULL,'"+str(currentDate)+"', 16, 1, 0, "+str(self.idAbattoir)+", '") + self.typeLivraison + "', '" + str( self.poidsTotalCalcule) + "','" + str(self.poidsTotalDetecte) + "','" + str( self.poidsTotalCalcule) + "','" + str(self.qttCocheDetecte) + "','" + str( self.qttCocheCalcule) + "', '" + self.dateLivraison + "','"+ self._numBl+"','"+ self._debugOCR.replace("'", "\\'") +"');" try: self.cursor.execute(sql) except Exception as err: print(f"UnexpectedA {err=}, {type(err)=}") print("ERROR A2") # On récupère l'id du BL créé dans Eteko blId = self.cursor.lastrowid # On retourne dans le bash l'id du BL créé # (NE PAS SUPPRIMER CETTE LIGNE !!!) print('{"blId":' + str(blId) + "}") # ------------------------------------------ for key in self.linesLots: numeroTuerie = str(key)[-4:] sql = ( "INSERT INTO prod_att(id_base, id_type, id_prod, id_work, actif, newobj, fid_prod, type, name, revision, `desc`, att1, att2, att3, att4, attf100, attf101,attf104) VALUES (1, 11, NULL, 11, 1, 0, null, null, null, null, null, '" + str( blId) + "','', '" + str(numeroTuerie) + "', '', '" + str(self.dateTuerie) + "', '0', '" + self.dateLivraison + "');") self.cursor.execute(sql) # Fermeture de la connexion à MySQL self.conn.close() except Exception as err: self.printDebug(f"Unexpected {err=}, {type(err)=}") self.printDebug("ERROR")