Hi,

I'm trying to make a small application which allows me to capture an image, process the image through the Microsoft Azure vision API and then give back the recognized text.

I'm struggeling to make my LiveCameraView to take a picture and save it.

My references:
AVFoundationPG - Media Capture
AVCapturePhotoOutput
Original code

from objc_util import *
import console
import ui

class LiveCameraView(ui.View):
   def __init__(self,device=0, *args, **kwargs):
      
      self.background_color = 'white'
      ui.View.__init__(self,*args,**kwargs)
      
      # Define a session, used to serve as a placeholder for the input and output + definition of quality
      self._session=ObjCClass('AVCaptureSession').alloc().init()
      
      # Set the quality of the capture
      self._session.setSessionPreset_('AVCaptureSessionPresetMedium');
      
      # Select an input device
      inputDevices=ObjCClass('AVCaptureDevice').devices()
      self._inputDevice=inputDevices[device]
      
      #self._inputDevice.unlockForConfiguration()
      #self._inputDevice.setFocusMode_(ns('AVCaptureFocusModeContinuousAutoFocus'))
      
      # Enable autofocus
      if self._inputDevice.isFocusModeSupported_(2):
        if self._inputDevice.lockForConfiguration_(None):
            self._inputDevice.focusMode = 2
            self._inputDevice.unlockForConfiguration()
      
      # Add the device to your session 
      deviceInput=ObjCClass('AVCaptureDeviceInput').deviceInputWithDevice_error_(self._inputDevice, None);
      
      # Configure device output
      
      deviceOutput = ObjCClass('AVCapturePhotoOutput').alloc().init()
      
      photoSettings = ObjCClass('AVCapturePhotoSettings')
      
      # FROM HERE ON, I'M LOST

      deviceConnection = ObjCClass('AVCaptureConnection')
      
      if self._session.canAddInput_(deviceInput):
         self._session.addInput_(deviceInput)
      
      if self._session.canAddOutput_(deviceOutput):
        self._session.addOutput_(deviceOutput)
      self._previewLayer=ObjCClass('AVCaptureVideoPreviewLayer').alloc().initWithSession_(self._session)
      self._previewLayer.setVideoGravity_( 
         'AVLayerVideoGravityResizeAspectFill')
      rootLayer=ObjCInstance(self).layer()
      rootLayer.setMasksToBounds_(True)
      self._previewLayer.setFrame_(
         CGRect(CGPoint(-70, 0), CGSize(self.height,100)))
      rootLayer.insertSublayer_atIndex_(self._previewLayer,0)
      self._session.startRunning()
      b = ui.Button(title = 'Scan code')
      b.action = self.snap
      b.center = (self.width * 0.5, self.height * 0.5)
      b.flex = 'LRTB'
      self.add_subview(b)
      
   def snap(self, sender):
      console.alert('snap','snap')
      
   def will_close(self):
      self._session.stopRunning()
   def layout(self):
      if not self._session.isRunning():
         self._session.startRunning()

rootview=LiveCameraView(frame=(0,0,200,500))
rootview.present('popover')

Can someone help me to capture the image and process it so I'm be able to send it using requests?

Thank you in advance,

Kind regards,

Gilles