parser/pageparser: Fix handling of commented out front matter

When the page parser was rewritten in 0.51, this was interpreted literally, but commented out front matter is used in the wild to "hide it from GitHub", e.g:

```
<!--
+++
title = "hello"
+++
-->
```

Fixes #5478

go.sum

@@ -68,6 +68,7 @@ github.com/magefile/mage v1.4.0 h1:RI7B1CgnPAuu2O9lWszwya61RLmfL0KCdo+QyyI/Bhk=
 github.com/magefile/mage v1.4.0/go.mod h1:IUDi13rsHje59lecXokTfGX0QIzO45uVPlXnJYsXepA=
 github.com/magiconair/properties v1.8.0 h1:LLgXmsheXeRoUOBOjtwPQCWIYqM/LU1ayDtDePerRcY=
 github.com/magiconair/properties v1.8.0/go.mod h1:PppfXfuXeibc/6YijjN8zIbojt8czPbwD3XqdrwzmxQ=
+github.com/markbates/inflect v0.0.0-20171215194931-a12c3aec81a6 h1:LZhVjIISSbj8qLf2qDPP0D8z0uvOWAW5C85ly5mJW6c=
 github.com/markbates/inflect v0.0.0-20171215194931-a12c3aec81a6/go.mod h1:oTeZL2KHA7CUX6X+fovmK9OvIOFuqu0TwdQrZjLTh88=
 github.com/matryer/try v0.0.0-20161228173917-9ac251b645a2/go.mod h1:0KeJpeMD6o+O4hW7qJOT7vyQPKrWmj26uf5wMc/IiIs=
 github.com/mattn/go-isatty v0.0.4 h1:bnP0vzxcAdeI1zdubAl5PjU6zsERjGZb7raWodagDYs=

hugolib/page_content.go

@@ -83,8 +83,6 @@ Loop:
 
 		switch {
 		case it.Type == pageparser.TypeIgnore:
-		case it.Type == pageparser.TypeHTMLComment:
-			// Ignore. This is only a leading Front matter comment.
 		case it.Type == pageparser.TypeHTMLStart:
 			// This is HTML without front matter. It can still have shortcodes.
 			p.renderable = false

hugolib/page_test.go

@@ -1586,6 +1586,33 @@ CONTENT:{{ .Content }}
 	)
 }
 
+// https://github.com/gohugoio/hugo/issues/5478
+func TestPageWithCommentedOutFrontMatter(t *testing.T) {
+	b := newTestSitesBuilder(t)
+	b.WithSimpleConfigFile()
+
+	b.WithContent("page.md", `<!--
++++
+title = "hello"
++++
+-->
+This is the content.
+`)
+
+	b.WithTemplatesAdded("layouts/_default/single.html", `
+Title: {{ .Title }}
+Content:{{ .Content }}
+`)
+
+	b.CreateSites().Build(BuildCfg{})
+
+	b.AssertFileContent("public/page/index.html",
+		"Title: hello",
+		"Content:<p>This is the content.</p>",
+	)
+
+}
+
 // TODO(bep) this may be useful for other tests.
 func compareObjects(a interface{}, b interface{}) bool {
 	aStr := strings.Split(fmt.Sprintf("%v", a), "")

parser/pageparser/item.go

@@ -108,7 +108,6 @@ const (
 
 	// page items
 	TypeHTMLStart          // document starting with < as first non-whitespace
-	TypeHTMLComment        // We ignore leading comments
 	TypeLeadSummaryDivider // <!--more-->,  # more
 	TypeFrontMatterYAML
 	TypeFrontMatterTOML

parser/pageparser/pagelexer.go

@@ -53,6 +53,8 @@ type pageLexer struct {
 	summaryDivider []byte
 	// Set when we have parsed any summary divider
 	summaryDividerChecked bool
+	// Whether we're in a HTML comment.
+	isInHTMLComment bool
 
 	lexerShortcodeState
 
@@ -120,7 +122,7 @@ var (
 	delimYAML         = []byte("---")
 	delimOrg          = []byte("#+")
 	htmlCommentStart  = []byte("<!--")
-	htmlCOmmentEnd    = []byte("-->")
+	htmlCommentEnd    = []byte("-->")
 )
 
 func (l *pageLexer) next() rune {
@@ -195,6 +197,15 @@ func (l *pageLexer) consumeCRLF() bool {
 	return consumed
 }
 
+func (l *pageLexer) consumeToNextLine() {
+	for {
+		r := l.next()
+		if r == eof || isEndOfLine(r) {
+			return
+		}
+	}
+}
+
 func (l *pageLexer) consumeSpace() {
 	for {
 		r := l.next()
@@ -206,6 +217,10 @@ func (l *pageLexer) consumeSpace() {
 }
 
 func lexMainSection(l *pageLexer) stateFunc {
+	if l.isInHTMLComment {
+		return lexEndFromtMatterHTMLComment
+	}
+
 	// Fast forward as far as possible.
 	var l1, l2 int
 
@@ -312,16 +327,15 @@ LOOP:
 		case r == byteOrderMark:
 			l.emit(TypeIgnore)
 		case !isSpace(r) && !isEndOfLine(r):
-			// No front matter.
 			if r == '<' {
 				l.backup()
 				if l.hasPrefix(htmlCommentStart) {
-					right := l.index(htmlCOmmentEnd)
-					if right == -1 {
-						return l.errorf("starting HTML comment with no end")
-					}
-					l.pos += right + len(htmlCOmmentEnd)
-					l.emit(TypeHTMLComment)
+					// This may be commented out front mattter, which should
+					// still be read.
+					l.consumeToNextLine()
+					l.isInHTMLComment = true
+					l.emit(TypeIgnore)
+					continue LOOP
 				} else {
 					if l.pos > l.start {
 						l.emit(tText)
@@ -341,6 +355,19 @@ LOOP:
 	return lexMainSection
 }
 
+func lexEndFromtMatterHTMLComment(l *pageLexer) stateFunc {
+	l.isInHTMLComment = false
+	right := l.index(htmlCommentEnd)
+	if right == -1 {
+		return l.errorf("starting HTML comment with no end")
+	}
+	l.pos += right + len(htmlCommentEnd)
+	l.emit(TypeIgnore)
+
+	// Now move on to the shortcodes.
+	return lexMainSection
+}
+
 func lexDone(l *pageLexer) stateFunc {
 
 	// Done!

parser/pageparser/pageparser_intro_test.go

@@ -60,7 +60,8 @@ var frontMatterTests = []lexerTest{
 	{"No front matter", "\nSome text.\n", []Item{tstSomeText, tstEOF}},
 	{"YAML front matter", "---\nfoo: \"bar\"\n---\n\nSome text.\n", []Item{tstFrontMatterYAML, tstSomeText, tstEOF}},
 	{"YAML empty front matter", "---\n---\n\nSome text.\n", []Item{nti(TypeFrontMatterYAML, ""), tstSomeText, tstEOF}},
-	{"YAML commented out front matter", "<!--\n---\nfoo: \"bar\"\n---\n-->\nSome text.\n", []Item{nti(TypeHTMLComment, "<!--\n---\nfoo: \"bar\"\n---\n-->"), tstSomeText, tstEOF}},
+	{"YAML commented out front matter", "<!--\n---\nfoo: \"bar\"\n---\n-->\nSome text.\n", []Item{nti(TypeIgnore, "<!--\n"), tstFrontMatterYAML, nti(TypeIgnore, "-->"), tstSomeText, tstEOF}},
+	{"YAML commented out front matter, no end", "<!--\n---\nfoo: \"bar\"\n---\nSome text.\n", []Item{nti(TypeIgnore, "<!--\n"), tstFrontMatterYAML, nti(tError, "starting HTML comment with no end")}},
 	// Note that we keep all bytes as they are, but we need to handle CRLF
 	{"YAML front matter CRLF", "---\r\nfoo: \"bar\"\r\n---\n\nSome text.\n", []Item{tstFrontMatterYAMLCRLF, tstSomeText, tstEOF}},
 	{"TOML front matter", "+++\nfoo = \"bar\"\n+++\n\nSome text.\n", []Item{tstFrontMatterTOML, tstSomeText, tstEOF}},
@@ -78,7 +79,6 @@ var frontMatterTests = []lexerTest{
 func TestFrontMatter(t *testing.T) {
 	t.Parallel()
 	for i, test := range frontMatterTests {
-
 		items := collect([]byte(test.input), false, lexIntroSection)
 		if !equal(items, test.items) {
 			got := crLfReplacer.Replace(fmt.Sprint(items))