Created
November 2, 2012 13:25
-
-
Save nf/4001362 to your computer and use it in GitHub Desktop.
script to convert go tour html to present file format
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 ( | |
"bytes" | |
"exp/html" | |
"fmt" | |
"io/ioutil" | |
"log" | |
"os" | |
"regexp" | |
"strings" | |
) | |
func main() { | |
root, err := html.Parse(os.Stdin) | |
if err != nil { | |
log.Fatal(err) | |
} | |
f, err := os.Create("tour.article") | |
if err != nil { | |
log.Fatal(err) | |
} | |
defer f.Close() | |
fmt.Fprintln(f, header) | |
for n, s := range findAll(root, hasClass("slide")) { | |
title := text(find(s, isTag("h2"))) | |
fmt.Fprintf(f, "* %v\n\n", title) | |
fmt.Fprint(f, slideText(s)) | |
source := text(find(s, hasClass("source"))) | |
if source != "" { | |
fn := "tour/" + filename(n, title) | |
ioutil.WriteFile(fn, []byte(source), 0666) | |
fmt.Fprintf(f, ".play %s\n\n", fn) | |
} | |
} | |
} | |
const header = ` | |
A Tour of Go | |
The Go Authors | |
http://golang.org | |
` | |
var nonSafe = regexp.MustCompile(`[^a-z ]`) | |
func filename(n int, title string) string { | |
title = strings.ToLower(title) | |
title = nonSafe.ReplaceAllString(title, "") | |
title = strings.TrimSpace(title) | |
title = strings.Replace(title, " ", "-", -1) | |
return fmt.Sprintf("%03d-%s.go", n, title) | |
} | |
func slideText(s *html.Node) string { | |
var buf bytes.Buffer | |
walk(s, func(n *html.Node) bool { | |
switch { | |
case isTag("p")(n): | |
buf.WriteString(wrap(text(n))) | |
case isTag("pre")(n) && !hasClass("source")(n): | |
buf.WriteString(indent(text(n))) | |
buf.WriteByte('\n') | |
default: | |
return true | |
} | |
return false | |
}) | |
return buf.String() | |
} | |
func indent(s string) string { | |
var buf bytes.Buffer | |
for _, l := range strings.Split(s, "\n") { | |
buf.WriteByte('\t') | |
buf.WriteString(l) | |
buf.WriteByte('\n') | |
} | |
return buf.String() | |
} | |
func wrap(s string) string { | |
var buf bytes.Buffer | |
var cont bool | |
for _, l := range strings.Split(s, "\n") { | |
l = strings.TrimSpace(l) | |
if len(l) == 0 { | |
if cont { | |
buf.WriteByte('\n') | |
buf.WriteByte('\n') | |
} | |
cont = false | |
} else { | |
if cont { | |
buf.WriteByte(' ') | |
} | |
buf.WriteString(l) | |
cont = true | |
} | |
} | |
return buf.String() | |
} | |
func text(n *html.Node) string { | |
var buf bytes.Buffer | |
walk(n, func(n *html.Node) bool { | |
if hasClass("appengineMode")(n) { | |
return false | |
} | |
if n.Type == html.TextNode { | |
buf.WriteString(n.Data) | |
} | |
return true | |
}) | |
return buf.String() | |
} | |
type selector func(*html.Node) bool | |
func isTag(name string) selector { | |
return func(n *html.Node) bool { | |
return n.DataAtom.String() == name | |
} | |
} | |
func hasClass(name string) selector { | |
return func(n *html.Node) bool { | |
for _, a := range n.Attr { | |
if a.Key == "class" { | |
for _, c := range strings.Fields(a.Val) { | |
if c == name { | |
return true | |
} | |
} | |
} | |
} | |
return false | |
} | |
} | |
func findAll(node *html.Node, fn selector) (nodes []*html.Node) { | |
walk(node, func(n *html.Node) bool { | |
if fn(n) { | |
nodes = append(nodes, n) | |
} | |
return true | |
}) | |
return | |
} | |
func find(n *html.Node, fn selector) *html.Node { | |
var result *html.Node | |
walk(n, func(n *html.Node) bool { | |
if result != nil { | |
return false | |
} | |
if fn(n) { | |
result = n | |
return false | |
} | |
return true | |
}) | |
return result | |
} | |
func walk(n *html.Node, fn selector) { | |
var w func(*html.Node, bool) | |
w = func(n *html.Node, first bool) { | |
if n == nil { | |
return | |
} | |
if fn(n) { | |
w(n.FirstChild, false) | |
} | |
if !first { | |
w(n.NextSibling, false) | |
} | |
return | |
} | |
w(n, true) | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment