A Python-Flask Image Factory

This article is a indirect successor to JavaScript-lazyload. You might want to read this first. To recap: We have designed a JavaScript routine, which will load images based on the view-ports size and position. Previously we have pre-generated those images and returned specific resolutions as static content, now we want to automatically scale images according to URL arguments.

Flask location block

In this guide we assume a media URL will look like this:

URL Example Colored

First we need a flask route, note that if you are using the default static location, you cannot send files from outside the ‘static’-directory.

The <path:path> maps the path after “picture” to the path argument. In addition we evaluate the URL-arguments. The function generateImage will be explained in the next section.

# unset default static directory if you want to
app = flask.Flask("NAME", static_folder=None)
def sendImage(path):
    scaleY = flask.request.args.get("scaley")
    scaleX = flask.request.args.get("scalex")
    encoding = flask.request.args.get("encoding")
    scaleY = round(float(scaleY))
    scaleX = round(float(scaleX))
    pathToGeneratedImage = generateImage(path, scaleX, scaleY, encoding)
    response = flask.send_from_directory(...)
    return response

Re-encoding, resizing and caching

Now we create a function, which will generate an image with the requested specifications from the original and save it to a caching path, which the function then will return.

We use the python module pillow (aka PIL) to resize and re-encode the images, specifically PIL.Image.thumbnail.

CACHE_DIR = "someotherdir/"

import PIL.Image
def generateImage(pathToOrig, scaleX, scaleY, encoding):
    '''Generate image with the requested scales and encoding'''

    # create a cache dir if it doesn't already exist #    
    if os.path.isfile(CACHE_DIR):
        raise OSError("Picture cache dir name is occupied by a file!")
    if not os.path.isdir(CACHE_DIR):

    # use same encoding if none is given # 
    filename, extension = os.path.splitext(os.path.basename(pathToOrig))
    if not encoding:
        encoding = extension.strip(".")

    # fix some extensions PILLOW doesn't like #
    if encoding.lower() == "jpg":
        encoding = "jpeg"

    # open image #
    image = PIL.Image.open(os.path.join(PICTURES_DIR, pathToOrig))

    # ensure sizes are valid #
    x, y = image.size
    if not scaleY:
        scaleY = y
    scaleX = min(x, scaleX)
    scaleY = min(y, scaleY)

    # generate new paths #
    FILE_FORMAT = "x-{x}-y-{y}-{fname}.{ext}"
    newFile = FILE_FORMAT.format(x=scaleX, y=scaleY, fname=filename, ext=encoding)
    newPath = os.path.join(CACHE_DIR, newFile)

    # save image with new size and encoding #
    image.thumbnail((scaleX, scaleY), PIL.Image.ANTIALIAS)
    image.save(newPath, encoding)

    # return the new path # 
    return newPath

Server: generate new image If you want to add additional headers to the response, you may add this extra step before returning the response.

raw = flask.send_from_directory(PICTURES_DIR, pathToGeneratedImage, cache_timeout=3600)
response = flask.make_response(raw)

And that’s it, now our picture-route supports scaling and encoding! As I said in the beginning, you might want to read JavaScript-lazyload for an idea on how to deploy this to your website.

by Yannik Schmidt
Tags: Web-Development, Fullstack, Python, SEO, JavaScript