Chomp Unicode BOM if present

Useful if using or sharing files with users that use editors that
append a unicode byte order marker header (like Windows notepad).

This will still assume files are UTF-8 encoded.

Closes #2075

1 files changed, 19 insertions, 0 deletions

diff --git a/parser/page.go b/parser/page.go
index 949bfdac6..5092caddc 100644
--- a/parser/page.go
+++ b/parser/page.go
@@ -50,6 +50,8 @@ const (
 	HTMLCommentStart = "<!--"
 	// HTMLCommentEnd identifies the end of HTML comment.
 	HTMLCommentEnd = "-->"
+	// BOM Unicode byte order marker
+	BOM = '\ufeff'
 )
 
 var (
@@ -101,6 +103,10 @@ func (p *page) Metadata() (meta interface{}, err error) {
 func ReadFrom(r io.Reader) (p Page, err error) {
 	reader := bufio.NewReader(r)
 
+	// chomp BOM and assume UTF-8
+	if err = chompBOM(reader); err != nil && err != io.EOF {
+		return
+	}
 	if err = chompWhitespace(reader); err != nil && err != io.EOF {
 		return
 	}
@@ -135,6 +141,19 @@ func ReadFrom(r io.Reader) (p Page, err error) {
 	return newp, nil
 }
 
+func chompBOM(r io.RuneScanner) (err error) {
+	for {
+		c, _, err := r.ReadRune()
+		if err != nil {
+			return err
+		}
+		if c != BOM {
+			r.UnreadRune()
+			return nil
+		}
+	}
+}
+
 func chompWhitespace(r io.RuneScanner) (err error) {
 	for {
 		c, _, err := r.ReadRune()