html: handle end tags in strange places

Pass tests1.dat, test 111: </strong></b></em></i></u></strike></s></blink></tt></pre></big></small></font></select></h1></h2></h3></h4></h5></h6></body></br></a></img></title></span></style></script></table></th></td></tr></frame></area></link></param></hr></input></col></base></meta></basefont></bgsound></embed></spacer></p></dd></dt></caption></colgroup></tbody></tfoot></thead></address></blockquote></center></dir></div></dl></fieldset></listing></menu></ol></ul></li></nobr></wbr></form></button></marquee></object></html></frameset></head></iframe></image></isindex></noembed></noframes></noscript></optgroup></option></plaintext></textarea> | <html> | <head> | <body> | <br> | <p> Also pass all the remaining tests in tests1.dat. R=nigeltao CC=golang-dev https://golang.org/cl/5372066

html: handle end tags in strange places
Pass tests1.dat, test 111: </strong></b></em></i></u></strike></s></blink></tt></pre></big></small></font></select></h1></h2></h3></h4></h5></h6></body></br></a></img></title></span></style></script></table></th></td></tr></frame></area></link></param></hr></input></col></base></meta></basefont></bgsound></embed></spacer></p></dd></dt></caption></colgroup></tbody></tfoot></thead></address></blockquote></center></dir></div></dl></fieldset></listing></menu></ol></ul></li></nobr></wbr></form></button></marquee></object></html></frameset></head></iframe></image></isindex></noembed></noframes></noscript></optgroup></option></plaintext></textarea> | <html> | <head> | <body> | <br> | <p> Also pass all the remaining tests in tests1.dat. R=nigeltao CC=golang-dev https://golang.org/cl/5372066
3df05124 · Andrew Balholm · Nigel Tao · e8188c16 · 3df05124 · 3df05124
Commit 3df05124 authored Nov 12, 2011 by Andrew Balholm Committed by Nigel Tao Nov 12, 2011
Hide whitespace changes
Inline Side-by-side

Showing with 18 additions and 31 deletions

parse.go src/pkg/html/parse.go +17 -30

parse_test.go src/pkg/html/parse_test.go +1 -1

No files found.
--- a/src/pkg/html/parse.go
+++ b/src/pkg/html/parse.go
@@ -352,30 +352,19 @@ func initialIM(p *parser) (insertionMode, bool) {
 // Section 11.2.5.4.2.
 func beforeHTMLIM(p *parser) (insertionMode, bool) {
-	var (
-		add     bool
-		attr    []Attribute
-		implied bool
-	)
 	switch p.tok.Type {
-	case ErrorToken:
-		implied = true
-	case TextToken:
-		// TODO: distinguish whitespace text from others.
-		implied = true
 	case StartTagToken:
 		if p.tok.Data == "html" {
-			add = true
+			p.addElement(p.tok.Data, p.tok.Attr)
-			attr = p.tok.Attr
+			return beforeHeadIM, true
-		} else {
-			implied = true
 		}
 	case EndTagToken:
 		switch p.tok.Data {
 		case "head", "body", "html", "br":
-			implied = true
+			// Drop down to creating an implied <html> tag.
 		default:
 			// Ignore the token.
+			return beforeHTMLIM, true
 		}
 	case CommentToken:
 		p.doc.Add(&Node{
@@ -384,10 +373,9 @@ func beforeHTMLIM(p *parser) (insertionMode, bool) {
 		})
 		return beforeHTMLIM, true
 	}
-	if add || implied {
+	// Create an implied <html> tag.
-		p.addElement("html", attr)
+	p.addElement("html", nil)
-	}
+	return beforeHeadIM, false
-	return beforeHeadIM, !implied
 }
 // Section 11.2.5.4.3.
@@ -691,6 +679,9 @@ func inBodyIM(p *parser) (insertionMode, bool) {
 			if p.popUntil(defaultScopeStopTags, p.tok.Data) {
 				p.clearActiveFormattingElements()
 			}
+		case "br":
+			p.tok.Type = StartTagToken
+			return inBodyIM, false
 		default:
 			p.inBodyEndTagOther(p.tok.Data)
 		}
@@ -1192,18 +1183,15 @@ func inSelectIM(p *parser) (insertionMode, bool) {
 func afterBodyIM(p *parser) (insertionMode, bool) {
 	switch p.tok.Type {
 	case ErrorToken:
-		// TODO.
+		// Stop parsing.
-	case TextToken:
+		return nil, true
-		// TODO.
 	case StartTagToken:
-		// TODO.
+		if p.tok.Data == "html" {
+			return useTheRulesFor(p, afterBodyIM, inBodyIM)
+		}
 	case EndTagToken:
-		switch p.tok.Data {
+		if p.tok.Data == "html" {
-		case "html":
-			// TODO: autoclose the stack of open elements.
 			return afterAfterBodyIM, true
-		default:
-			// TODO.
 		}
 	case CommentToken:
 		// The comment is attached to the <html> element.
@@ -1216,8 +1204,7 @@ func afterBodyIM(p *parser) (insertionMode, bool) {
 		})
 		return afterBodyIM, true
 	}
-	// TODO: should this be "return inBodyIM, true"?
+	return inBodyIM, false
-	return afterBodyIM, true
 }
 // Section 11.2.5.4.19.

--- a/src/pkg/html/parse_test.go
+++ b/src/pkg/html/parse_test.go
@@ -133,7 +133,7 @@ func TestParser(t *testing.T) {
 		n int
 	}{
 		// TODO(nigeltao): Process all the test cases from all the .dat files.
-		{"tests1.dat", 111},
+		{"tests1.dat", -1},
 		{"tests2.dat", 0},
 		{"tests3.dat", 0},
 	}