Notes

class DataExtractionView(APIView):
authentication_classes = [TokenAuthentication]
permission_classes = [IsAuthenticated]

def post(self, request, id):
try:
instance = TicketField.objects.get(id=id)
print(instance.id, "Instance")
image_link = instance.tickets_jpeg
instance_id = instance.id
print(image_link)
except TicketField.DoesNotExist:
return Response({"message": "Image does not exist"}, status=status.HTTP_404_NOT_FOUND)

response = requests.get(image_link)
if response.status_code == 200:
# Download the image and save into directory
# image_name = os.path.basename(image_link)
# image_path = os.path.join(images_directory, image_name)
# print(image_path,"!!!!!!!!!!!!!!!!!")
match = re.search(r'([^/]+).jpg', image_link)
filename = f"{match.group(1)}.jpg"
print(filename)
image_path = f"{settings.IMAGES_ROOT}/{instance_id}"
print("IMAGES PATHN",image_path)
os.makedirs(image_path, exist_ok=True)
image_file_path = f"{image_path}/{filename}"
print("IMAGES PATH",image_file_path)
with open(image_file_path, 'wb') as f:
f.write(response.content)

cropped_image_path, cropped_image_dir = get_cropped_images(image_file_path, instance_id)
print('IMAGES PATH', cropped_image_path)

bol_text = ''
outbound_text = ''
inbound_text = ''

# Initialize AWS Textract client
textract_client = boto3.client('textract', region_name=settings.REGION_NAME,
aws_access_key_id=settings.AWS_ACCESS_KEY_ID,
aws_secret_access_key=settings.AWS_SECRET_ACCESS_KEY)

try:
for image in cropped_image_path:
print('IMAGE', image)
# Read the image
with open(image, 'rb') as file:
file_bytes = file.read()

# Call Textract API to analyze the image
response = textract_client.analyze_document(
Document={'Bytes': file_bytes},
FeatureTypes=['TABLES']
)

# Extract raw text from the response
text = ""
for item in response["Blocks"]:
if item["BlockType"] == "LINE":
text = text + " " + item["Text"]

# Clean the text by removing ambiguity
cleaned_text = text.replace('"', '').replace('\', '')
# print("CLEANED TEXT", cleaned_text)

# Determine which type of image it is based on the image name
image_name = image.split('/')[-1] # Assuming image path contains '/'
if "inbound" in image_name:
inbound_text += cleaned_text
print("INBOUND", inbound_text)
elif "lading" in image_name:
bol_text += cleaned_text
print("LADING", bol_text)
elif "outbound" in image_name:
outbound_text += cleaned_text
print("OUTBOUND", outbound_text)

# Constructing the dictionary
result = {
'BOL': bol_text.strip(),
'OUTBOUND': outbound_text.strip(),
'INBOUND': inbound_text.strip()
}
print("RESULT DICT", result)
print('BOL RAW TEXT', result['BOL'])
print("FORT RAW TEXT", result['INBOUND'])
for key in result:
if 'BOL' in key:
if result['BOL'] != '':
bol_raw_text = result['BOL']
# Calling BOL function to do further processing.
bol_mapped_data = bol(bol_raw_text, instance)
else:
# If bol_raw_text is empty, store None or appropriate value
bol_mapped_data = {}

elif 'OUTBOUND' in key:
if result['OUTBOUND'] != '':
outbound_raw_text = result['OUTBOUND']
# Calling OUTBOUND function to do further processing.
outbound_mapped_data = outbound(outbound_raw_text, instance)
else:
# If outbound_raw_text is empty, store None or appropriate value
outbound_mapped_data = {}

elif 'INBOUND' in key:
if result['INBOUND'] != '':
inbound_raw_text = result['INBOUND']
# Calling INBOUND function to do further processing.
inbound_mapped_data = inbound(inbound_raw_text, instance)
print('INBOUND DATA', inbound_mapped_data)
else:
# If inbound_raw_text is empty, store None or appropriate value
inbound_mapped_data = {}

# Mapping the data of BOL_OUT_IN
bol_out_in = bol_mapped_data.get('bol_number', 'NA') + ', ' + outbound_mapped_data.get(
'outbound_ticket_number', 'NA') + ', ' + inbound_mapped_data.get('inbound_ticket_number', 'NA')
instance.bol_out_in = bol_out_in
instance.save()

context = {
'bol_number': bol_mapped_data.get('bol_number', ''),
'outbound_ticket_number': outbound_mapped_data.get('outbound_ticket_number', ''),
'inbound_ticket_number': inbound_mapped_data.get('inbound_ticket_number', ''),
'freight_rate': '',
'fields': '',
'bol_contract_number': bol_mapped_data.get('bol_contract_number', ''),
'bol_driver_name': bol_mapped_data.get('bol_driver_name', ''),
'bol_truck_number': bol_mapped_data.get('bol_truck_number', ''),
'outbound_fds_load_number': outbound_mapped_data.get('outbound_fds_load_number', '')
}

# Delete the Image Directory
shutil.rmtree(image_path)

# Delete the Cropped Image Directory
shutil.rmtree(cropped_image_dir)
return Response({"message": "Data Extracted Successfully!!", "context": context}, status=status.HTTP_200_OK)

except ClientError as e:
return Response({"message": f"AWS Textract error: {e}"},
status=status.HTTP_500_INTERNAL_SERVER_ERROR)
else:
return Response({"message": "Failed to fetch the image"}, status=response.status_code)