Extras

Unfortunately there is a bunch more to running a website besides building the application. Fortunately they are MOSTLY boilerplate.

​

SEO

For Search Engine Optimization the most imortant part (along with actually good content) are the meta tags. Open _base.html, you will see the most important meta tags are created and filled programmatically. Now open app.py and go down to our context manager (it inserts variables into ALL of our templates):

@app.context_processor
def injectProcessor():
    ...
    return dict(
        user_logged_in=user_logged_in, 
        user_payed=user_payed,
        html_lang="en",
        locale="en_US",
        description=os.getenv("DESCRIPTION", ""),
        title=os.getenv("TITLE", ""),
        url=request.host_url,
        author=os.getenv("PERSONAL_NAME", ""),
        name=os.getenv("APPLICATION_NAME", ""),
        type="website",
        keywords=os.getenv("KEYWORDS", ""),
        og_img_path="./static/assets/hero.png"
    )

You can see that all of the personalized meta information is pulled from .env, fill it accordingly. Also change the image path to the one you want to be displayed, when you share your website on social media.

Tip: when setting the title and description, you can follow the length guidelines in this tool.

​

Terms and Conditions / Privacy Agreement

In the footer of the page you have links to these 2 things.

They are written by ChatGPT. Use this as a placeholder in development until you create a correct one for you. I do NOT take any liability for these documents.

Just like the meta tags, these use information from .env.

def process_markdown(filepath):
    with open(filepath, "r", encoding="utf-8") as f:
        md = markdown.markdown(
            f.read()
            .replace("COMPANY NAME", os.getenv("COMPANY_NAME", "-"))
            .replace("APPLICATION NAME", os.getenv("APPLICATION_NAME", "-"))
            .replace("WEBSITE URL", request.host_url)
            .replace("CURRENCY", os.getenv("CURRENCY", ""))
        )
    
    return md


@app.get("/toc")
def tocRoute():
    return render_template("message.html", heading="Terms and Conditions", body=process_markdown("./static/texts/toc.md"))

@app.get("/privacy")
def privacyRoute():
    return render_template("message.html", heading="Privacy Agreement", body=process_markdown("./static/texts/privacy.md"))

​

Favicon

The favicon is the little icon that gets displayed in your browser. Interestingly it is more than that. Once you create a favicon from a normal image in this tool, you will get a bunch of different files, I will just let ChatGPT explain:

Simply extract the metioned files into static/assets/favicon

​

Security

To cover most security considerations, we use an amazing Plugin called “Flask Talisman” We simply wrap our application and pass any custom settings. I highly recommend to check out their ReadMe to understand what it is you are actually doing under the hood.

One of these customizations, is the “Content Security Policy” header. It defines, from where our HTMl can load content from. By default, it is our server. When later deploying the application, we will serve our static content through a CDN, so we whitelist it as a valid origin. If you have any other sources you want to allow content to be loaded from (for example Plausible Analytics in my case), add them here as well.

csp = {
    'default-src': [
        '\'self\'',
        'plausible.io'
    ],
}
if os.getenv("CDN_DOMAIN", None):
    csp["default-src"].append(os.getenv("CDN_DOMAIN"))

Talisman(app, content_security_policy=csp)

​

Writing an actual page

You can check out this Hollywood-level demo on how to quickly add a new page to your application

​

Congrats!

You now know how to develop quickly using ShipPil. But that is not the end. Continue reading the deployment chapter to learn how you can server your newly built application to the World Wide Web.