Last active
July 2, 2021 06:00
-
-
Save owulveryck/510cad103240b345b7db56dfd6a333b9 to your computer and use it in GitHub Desktop.
Simple script to turn a webpage into an epub
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package main | |
import ( | |
"log" | |
"net/http" | |
"net/url" | |
"os" | |
"path/filepath" | |
"strings" | |
"github.com/bmaupin/go-epub" | |
"github.com/cixtor/readability" | |
"golang.org/x/net/html" | |
) | |
func main() { | |
if len(os.Args) != 3 { | |
log.Fatalf("usage: %v URL output.epub", os.Args[0]) | |
} | |
url := os.Args[1] | |
r := readability.New() | |
res, err := http.Get(url) | |
if err != nil { | |
log.Fatal(err) | |
} | |
defer res.Body.Close() | |
article, err := r.Parse(res.Body, url) | |
if err != nil { | |
log.Fatal(err) | |
} | |
d := newDocument() | |
err = d.setMeta(&article) | |
if err != nil { | |
log.Fatal(err) | |
} | |
err = d.replaceImages(article.Node) | |
if err != nil { | |
log.Fatal(err) | |
} | |
var body strings.Builder | |
err = html.Render(&body, article.Node) | |
if err != nil { | |
// handle error | |
log.Fatal(err) | |
} | |
// Create a new EPUB | |
d.AddSection(body.String(), "Content", "", "") | |
err = d.Write(os.Args[2]) | |
if err != nil { | |
log.Fatal(err) | |
} | |
} | |
type document struct { | |
*epub.Epub | |
buf strings.Builder | |
currSection string | |
} | |
func newDocument() *document { | |
return &document{ | |
Epub: epub.NewEpub(""), | |
buf: strings.Builder{}, | |
currSection: "Preamble", | |
} | |
} | |
func (d *document) setMeta(a *readability.Article) error { | |
d.SetTitle(a.Title) | |
d.SetDescription(a.Excerpt) | |
d.SetAuthor(a.Byline) | |
if a.Image != "" { | |
img, err := d.AddImage(a.Image, "") | |
if err != nil { | |
return err | |
} | |
d.SetCover(img, "") | |
} | |
return nil | |
} | |
func (d *document) replaceImages(n *html.Node) error { | |
if n.Type == html.ElementNode && n.Data == "img" { | |
for i, a := range n.Attr { | |
if a.Key == "src" { | |
// get the filname | |
u, err := url.Parse(a.Val) | |
if err != nil { | |
return err | |
} | |
f := filepath.Base(u.Path) | |
img, err := d.AddImage(a.Val, f) | |
if err != nil { | |
return err | |
} | |
n.Attr[i].Val = img | |
} | |
// remove the srcset | |
if a.Key == "srcset" { | |
n.Attr[i] = n.Attr[len(n.Attr)-1] // Copy last element to index i. | |
n.Attr[len(n.Attr)-1] = html.Attribute{} // Erase last element (write zero value). | |
n.Attr = n.Attr[:len(n.Attr)-1] // Truncate slice. | |
} | |
} | |
} | |
for c := n.FirstChild; c != nil; c = c.NextSibling { | |
err := d.replaceImages(c) | |
if err != nil { | |
return err | |
} | |
} | |
return nil | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment