I'm extending my quick script from a previous post on Automating Image Uploads to Cloudinary with Python. I figured — why stop there! Instead of just automating the upload flow and getting the url copied to my clipboard, I could have it generate the whole markdown file for me!

Form there, the steps that will be automated:

• Verifying it's an art post (based on which folder I select to save the image to)
• Creating a file with the post date in the filename (I usually put them together on Fridays, and share on Saturday)
• Writing details from the command line, such as title, alt text, and post body.

Here we go!

Reorganizing

First order of business is organizing the code. This was previously just a script, but now I'd like to encapsulate everything within a class.

In my index.py, I'll setup the class:

from dotenv import load_dotenv
load_dotenv()

import datetime
import cloudinary
import cloudinary.uploader
import cloudinary.api
import pyperclip

class ImageHandler():
    """
    Upload Images to cloudinary. Generate Blog page if applicable
    """
    def __init__(self, test: bool = False):
        self.test = test

        config = cloudinary.config(secure=True)
        print("****1. Set up and configure the SDK:****\nCredentials: ", config.cloud_name, config.api_key, "\n")

Beneath, I'll wrap the previous code in a method called "upload_image()"

I'll still call the file directly to run the main script, so at the bottom of the file I'll add the code to do so:

if __name__ == '__main__':
    ImageHandler().run()

So, on run, I want to handle the upload to Cloudinary. From the result, I want to get the image url back, and I want to know if I should start generating an art post:

def run(self):
    image_url, is_art_image = self.upload_image()
    print(is_art_image)
    if is_art_image:
        self.generate_blog_post(image_url=image_url)

To make that happen, I'm adding a bit of logic to image_upload to check if I'm storing the image in my art folder:

def upload_image(self):

    ... 
    
    options = [
        "/chrisdpadilla/blog/art",
        "/chrisdpadilla/blog/images",
        "/chrisdpadilla/albums",
    ]

    is_art_image = selected_number == 0
    
    ...
    
    selected_number = int(selected_number_input) - 1
    is_art_image = selected_number == 0
        
    ...
    
    return (res_url, is_art_image)

That tuple will then return string and boolean in a neat little package.

Now for the meat of it! I'll write out my generate_blog_post() method.

Starting with date getting. Using datetime and timedelta, I'll be checking to see when the next Saturday will be (ART_TARGET_WEEKDAY is a constant set to 6 for Saturday):

def generate_blog_post(self, image_url: str = '') -> bool:
    today = datetime.datetime.now()

    weekday = today.weekday()
    days_from_target = datetime.timedelta(days=ART_TARGET_WEEKDAY - weekday)
    target_date = today + days_from_target
    target_date_string = target_date.strftime('%Y-%m-%d')

Sometimes, I do this a week out, so I'll add some back and forth incase I want to manually set the date:

    date_ok = input(f"{target_date_string} date ok? (Y/n): ")
    if "n" in date_ok:
        print("Set date:")
        target_date_string = input()
        print(f"New date: {target_date_string}")
        input("Press enter to continue ")

From here, it's all command line inputs.

    
    title = input("Title: ")
    alt_text = input("Alt text: ")
    caption = input("Caption: ")

And then I'll use a with statement to do my file writing. The benefit of using the with keyword here is that it will handle file closing automatically.

    md_body = SKETCHES_POST_TEMPLATE.format(title, target_date_string, alt_text, image_url, caption)
    with open(BLOG_POST_DIR + f"/sketches-{target_date_string}.md", "w") as f:
        f.write(md_body)

    return True

Violà! Running through the command line now generates a little post! Lots of clicking and copy-and-pasting time saved! 🙌