Object oriented architecture with an image processing code

This code is python 3X code (in django).
It takes an instance of user entry, and process images: it creates a resized version of original image, 2 thumbs, saves all, and update instance fields to point to correct files so it's correctly saved.
Then it returns the updated instance.

I'm not sure this code object oriented architecture is correct. It seems lots of line of redundant code. Could I have made it smarter?
I tried to break code on several methods that doe only one thing.
Then I process it with the method call 'execute' which is used elsewhere in the script when this code is used.

Also what do you think of encapsulation of this code?

```
class ImageProcessing:
def __init__(self, instance, image):
self.instance = instance
self.image = image

# Settings
self.sizes = OrderedDict()
self.sizes['image_main'] = (800, 800)
self.sizes['thumbsize_big'] = (200, 200)
self.sizes['thumbsize_small'] = (100, 100)
self.png_compress = 6
self.jpg_compress = 80
self.has_iterated = 0
# Variables
self.slug = slugify(instance.myField, allow_unicode=True)
self.saving_path = instance.savePath
self.base_dir = settings.BASE_DIR + self.saving_path
self.image_format = self.image.image.format.lower()
has_iterated = 0

def resize_image(self, image, size):
myimage = Image.open(image)
myimage = myimage.resize(size, Image.ANTIALIAS)
return myimage

def save_image(self, image, save_dir, png_compress, jpg_compress):
if self.image_format == 'png':
image.save(save_dir + '.png', compress_level=png_compress, format='PNG')
if self.image_format == 'jpg' or 'jpeg':
image.save(save_dir + '.jpg', quality=jpg_compress, format='JPEG')

def execute(self):
for key, value in self.sizes.items():
save_dir = self.base_dir + self.slug + '_' + key

i

Reducing repetition

self.sizes

self.sizes being repeated 4 times is not necessary:

self.sizes                    = OrderedDict()
    self.sizes['image_main']      = (800, 800)
    self.sizes['thumbsize_big']   = (200, 200)
    self.sizes['thumbsize_small'] = (100, 100)

You may give an initial argument to OrderedDict to avoid it:

self.sizes = OrderedDict( (
        ('image_main', (800, 800)),
        ('thumbsize_big', (200, 200)),
        ('thumbsize_small', (100, 100)) 
    ))

myimage

def resize_image(self, image, size):
    myimage = Image.open(image)
    myimage = myimage.resize(size, Image.ANTIALIAS)
    return myimage

In fact the myimage variable may be avoided completely, just chaining the method calls is much simpler:

def resize_image(self, image, size):
    return Image.open(image).resize(size, Image.ANTIALIAS)

Whole block repetition

if self.image_format == 'png':
            # Save images on disk
            self.save_image(self.resize_image(self.image, value),save_dir, self.png_compress, self.jpg_compress)
            # Save images in fields
            if self.has_iterated == 0: self.instance.image_main = self.saving_path + self.slug + '_' + key + '.png'
            if self.has_iterated == 1: self.instance.image_thumbsize_big = self.saving_path + self.slug + '_' + key + '.png'
            if self.has_iterated == 2: self.instance.image_thumbsize_small = self.saving_path + self.slug + '_' + key + '.png'
            self.has_iterated += 1
        if self.image_format == 'jpg' or 'jpeg':
            self.save_image(self.resize_image(self.image, value),save_dir, self.png_compress, self.jpg_compress)
            if self.has_iterated == 0: self.instance.image_main = self.saving_path + self.slug + '_' + key + '.jpg'
            if self.has_iterated == 1: self.instance.image_thumbsize_big = self.saving_path + self.slug + '_' + key + '.jpg'
            if self.has_iterated == 2: self.instance.image_thumbsize_small = self.saving_path + self.slug + '_' + key + '.jpg'
            self.has_iterated += 1

The blocks are very similar just the file-extension changes.

extension = 'png' if self.image_format == 'png' else 'jpg'

And then you can delete the conditional branches that follow and leave only one path.

Bug on the use of or

if self.image_format == 'jpg' or 'jpeg':

Does not work as you intended. or returns the first truthy value, so you wrote the same as:

if self.image_format == 'jpg':

StackExchange Code Review Q#117253, answer score: 2