html: parse DOCTYPE into name and public and system identifiers

Pass tests2.dat, test 59: <!DOCTYPE <!DOCTYPE HTML>>--> | <!DOCTYPE <!doctype> | <html> | <head> | <body> | ">" |  | "-->" Pass all the tests in doctype01.dat. Also pass tests2.dat, test 60: <!doctype html><div><form></form><div></div></div> R=nigeltao CC=golang-dev https://golang.org/cl/5437045

html: parse DOCTYPE into name and public and system identifiers
77b0ad1e · Andrew Balholm · Nigel Tao · b3923a27 · 77b0ad1e · 77b0ad1e
Commit 77b0ad1e authored Nov 24, 2011 by Andrew Balholm Committed by Nigel Tao Nov 24, 2011
Hide whitespace changes
Inline Side-by-side

Showing with 129 additions and 6 deletions

src/pkg/html/parse.go src/pkg/html/parse.go +54 -4

src/pkg/html/parse_test.go src/pkg/html/parse_test.go +19 -2

src/pkg/html/render.go src/pkg/html/render.go +56 -0

No files found.
--- a/src/pkg/html/parse.go
+++ b/src/pkg/html/parse.go
@@ -321,6 +321,59 @@ func (p *parser) resetInsertionMode() {

 const whitespace = " \t\r\n\f"

+// parseDoctype parses the data from a DoctypeToken into a name,
+// public identifier, and system identifier. It returns a Node whose Type 
+// is DoctypeNode, whose Data is the name, and which has attributes
+// named "system" and "public" for the two identifiers if they were present.
+func parseDoctype(s string) *Node {
+	n := &Node{Type: DoctypeNode}
+
+	// Find the name.
+	space := strings.IndexAny(s, whitespace)
+	if space == -1 {
+		space = len(s)
+	}
+	n.Data = strings.ToLower(s[:space])
+	s = strings.TrimLeft(s[space:], whitespace)
+
+	if len(s) < 6 {
+		// It can't start with "PUBLIC" or "SYSTEM".
+		// Ignore the rest of the string.
+		return n
+	}
+
+	key := strings.ToLower(s[:6])
+	s = s[6:]
+	for key == "public" || key == "system" {
+		s = strings.TrimLeft(s, whitespace)
+		if s == "" {
+			break
+		}
+		quote := s[0]
+		if quote != '"' && quote != '\'' {
+			break
+		}
+		s = s[1:]
+		q := strings.IndexRune(s, rune(quote))
+		var id string
+		if q == -1 {
+			id = s
+			s = ""
+		} else {
+			id = s[:q]
+			s = s[q+1:]
+		}
+		n.Attr = append(n.Attr, Attribute{Key: key, Val: id})
+		if key == "public" {
+			key = "system"
+		} else {
+			key = ""
+		}
+	}
+
+	return n
+}
+
 // Section 11.2.5.4.1.
 func initialIM(p *parser) bool {
 	switch p.tok.Type {
@@ -337,10 +390,7 @@ func initialIM(p *parser) bool {
 		})
 		return true
 	case DoctypeToken:
-		p.doc.Add(&Node{
-			Type: DoctypeNode,
-			Data: p.tok.Data,
-		})
+		p.doc.Add(parseDoctype(p.tok.Data))
 		p.im = beforeHTMLIM
 		return true
 	}

--- a/src/pkg/html/parse_test.go
+++ b/src/pkg/html/parse_test.go
@@ -97,7 +97,23 @@ func dumpLevel(w io.Writer, n *Node, level int) error {
 	case CommentNode:
 		fmt.Fprintf(w, "<!-- %s -->", n.Data)
 	case DoctypeNode:
-		fmt.Fprintf(w, "<!DOCTYPE %s>", n.Data)
+		fmt.Fprintf(w, "<!DOCTYPE %s", n.Data)
+		if n.Attr != nil {
+			var p, s string
+			for _, a := range n.Attr {
+				switch a.Key {
+				case "public":
+					p = a.Val
+				case "system":
+					s = a.Val
+				}
+			}
+			if p != "" || s != "" {
+				fmt.Fprintf(w, ` "%s"`, p)
+				fmt.Fprintf(w, ` "%s"`, s)
+			}
+		}
+		io.WriteString(w, ">")
 	case scopeMarkerNode:
 		return errors.New("unexpected scopeMarkerNode")
 	default:
@@ -133,8 +149,9 @@ func TestParser(t *testing.T) {
 		n int
 	}{
 		// TODO(nigeltao): Process all the test cases from all the .dat files.
+		{"doctype01.dat", -1},
 		{"tests1.dat", -1},
-		{"tests2.dat", 59},
+		{"tests2.dat", -1},
 		{"tests3.dat", 0},
 	}
 	for _, tf := range testFiles {

--- a/src/pkg/html/render.go
+++ b/src/pkg/html/render.go
@@ -9,6 +9,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"strings"
 )

 type writer interface {
@@ -98,6 +99,40 @@ func render1(w writer, n *Node) error {
 		if _, err := w.WriteString(n.Data); err != nil {
 			return err
 		}
+		if n.Attr != nil {
+			var p, s string
+			for _, a := range n.Attr {
+				switch a.Key {
+				case "public":
+					p = a.Val
+				case "system":
+					s = a.Val
+				}
+			}
+			if p != "" {
+				if _, err := w.WriteString(" PUBLIC "); err != nil {
+					return err
+				}
+				if err := writeQuoted(w, p); err != nil {
+					return err
+				}
+				if s != "" {
+					if err := w.WriteByte(' '); err != nil {
+						return err
+					}
+					if err := writeQuoted(w, s); err != nil {
+						return err
+					}
+				}
+			} else if s != "" {
+				if _, err := w.WriteString(" SYSTEM "); err != nil {
+					return err
+				}
+				if err := writeQuoted(w, s); err != nil {
+					return err
+				}
+			}
+		}
 		return w.WriteByte('>')
 	default:
 		return errors.New("html: unknown node type")
@@ -181,6 +216,27 @@ func render1(w writer, n *Node) error {
 	return w.WriteByte('>')
 }

+// writeQuoted writes s to w surrounded by quotes. Normally it will use double
+// quotes, but if s contains a double quote, it will use single quotes.
+// It is used for writing the identifiers in a doctype declaration.
+// In valid HTML, they can't contain both types of quotes.
+func writeQuoted(w writer, s string) error {
+	var q byte = '"'
+	if strings.Contains(s, `"`) {
+		q = '\''
+	}
+	if err := w.WriteByte(q); err != nil {
+		return err
+	}
+	if _, err := w.WriteString(s); err != nil {
+		return err
+	}
+	if err := w.WriteByte(q); err != nil {
+		return err
+	}
+	return nil
+}
+
 // Section 13.1.2, "Elements", gives this list of void elements. Void elements
 // are those that can't have any contents.
 var voidElements = map[string]bool{