Are you annoyed by the extra space after your name [name|]?

The extra space is no more! All fixed. Here's a little backstory that I hope inspires you to build more tools for yourself.

Back in 2018, I built TechLetter.App as part of a 24 hour challenge to build and launch an app. I wanted an easy to way to turn markdown into shippable newsletters and thought others might too.

They did not. TechLetter.App turned into the perfect example of Robin Sloane's An app can be a home-cooked meal. Gets about 30 visits per month. Mostly from me.

I can't imagine living without this tool and ConvertKit's banner that they're sunsetting the old editor with a Paste HTML option fills me with dread.

May have to change email providers over this 😅

How it works

TechLetter.App was one of the case studies peppered into Serverless Handbook.

It takes markdown, uses the unist ecosystem to parse the input, pass it through a chain of custom plugins, and spit out a bunch of React components. The Export button then takes the raw HTML from the parent node of the rendered document.

Plugins do the hard work. Some make calls to serverless functions that take screenshots, others transform text to other text.

The core business logic looks like this:

remark()
    .use(urlThumbnail, {
        domains: [
            "m.twitter.com",
            "twitter.com",
            "youtube.com",
            "youtu.be",
            "instagram.com",
            "codesandbox.io",
        ],
    })
    .use(twitterUserLinks)
    .use(codeScreenshot)
    .use(githubLinks)
    .use(remarkGiphySearch)
    .use(remarkNameOrFriend)
    .use(remarkSparkJoy)
    .use(remark2react, {
	    // ...

A chain of syntax tree transformations. Each .use() accepts a Remark plugin that transforms the syntax tree.

The space-after-name bug

Yes many readers complained about the extra space since I added the feature in August 2020. About one per quarter. It never quite made the priority list 😇

I use a custom bit of markdown syntax – [name|something else] – to inject your name into the email. If the system knows your name, it uses that, otherwise the something else part.

That happens via liquid syntax that ConvertKit and most other emailing systems understand. My custom Remark plugin finds the weird markdown syntax and turns it into liquid syntax.

Like this:

const nameOrFriend = /^\s*name\s*\|(.*)$/

export function remarkNameOrFriend() {
  return (tree) => {
    visit(tree, "linkReference", (node, index, parent) => {
      if (node.label.match(nameOrFriend)) {
        node.type = "text"

        node.value = node.label.replace(
          nameOrFriend,
          '{% if subscriber.first_name != blank %} {{ subscriber.first_name | truncatewords: 1, "" | capitalize }} {% else %} $1 {% endif %}'
        )
        node.children = null
      }
    })
  }
}

Can you spot the bug?

It's right here:

    '{% if subscriber.first_name != blank %} {{ subscriber.first_name | truncatewords: 1, "" | capitalize }} {% else %} $1 {% endif %}'

That has spaces between the liquid tags 🤣

Needs to be:

    '{% if subscriber.first_name != blank %}{{ subscriber.first_name | truncatewords: 1, "" | capitalize }}{% else %}$1{% endif %}'

And that's the bug that took me 919 days to fix.

Cheers,
~Swizec

PS: getting this deployed was an ordeal because software breaks when you change it. But that's a post for another day