clean_medium_html.go

package main

import (
	"bytes"
	"io/ioutil"
	"log"
	"os"
	"strings"

	"golang.org/x/net/html"
)

func main() {
	files := os.Args[2:]
	output := os.Args[1]
	all := &bytes.Buffer{}
	for _, f := range files {
		b, _ := ioutil.ReadFile(f)
		doc, _ := html.Parse(bytes.NewBuffer(b))
		removeUselessTags(doc)
		temp := &bytes.Buffer{}
		html.Render(temp, doc)
		if _, err := all.Write(temp.Bytes()); err != nil {
			log.Fatal(err)
		}
		if _, err := all.WriteString("\n\n---\n\n"); err != nil {
			log.Fatal(err)
		}
	}
	ioutil.WriteFile(output, all.Bytes(), 0644)
}

func removeUselessTags(n *html.Node) {
	for c := n.FirstChild; c != nil; c = c.NextSibling {
		if n.Type == html.ElementNode {
			switch n.Data {
			case "script", "aside", "style", "div":
				if n.Data == "div" {
					for _, a := range n.Attr {
						if a.Key == "class" && strings.Contains(a.Val, "postActions js-postActionsFooter") {
							n.RemoveChild(c)
						}
					}
				} else {
					n.RemoveChild(c)
				}
			}
		}
		removeUselessTags(c)
	}
}