Add support to extract voicename (#29)

* Add support to extract voicename

* Add more tests and Add Custom voice tag parser logic

* Fix comments

* Fix test fail due to incorrect end tag

* Improve test coverage

* Fix names

* Add test

* Use html tokenizer for parsing voice tags

* Unexport the ParseTextWebVTT and use internal test

go.mod

@@ -6,5 +6,6 @@ require (
 	github.com/asticode/go-astikit v0.8.0
 	github.com/asticode/go-astits v1.3.0
 	github.com/stretchr/testify v1.4.0
+	golang.org/x/net v0.0.0-20200904194848-62affa334b73
 	golang.org/x/text v0.3.2
 )

go.sum

@@ -9,9 +9,19 @@ github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSs
 github.com/pkg/profile v1.4.0/go.mod h1:NWz/XGvpEW1FyYQ7fCx4dqYBLlfTcE+A9FLAkNKqjFE=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/stretchr/objx v0.1.0 h1:4G4v2dO3VZwixGIRoQ5Lfboy6nUhCyYzaqnIAPPhYs4=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/testify v1.4.0 h1:2E4SXV/wtOkTonXsotYi4li6zVWxYlZuYNCXe9XRJyk=
 github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
+golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
+golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
+golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/net v0.0.0-20200904194848-62affa334b73 h1:MXfv8rhZWmFeqX3GNZRsd6vOLoaCHjYEX3qkRo3YBUA=
+golang.org/x/net v0.0.0-20200904194848-62affa334b73/go.mod h1:/O7V0waA8r7cgGh81Ro3o1hOxt32SMVPicZroKQ2sZA=
+golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20190412213103-97732733099d/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20200323222414-85ca7c5b95cd/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.2 h1:tW2bmiBqwgJj/UpqtC8EpXEZVYOwU0yG4iWbprSVAcs=
 golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=

webvtt.go

@@ -4,10 +4,13 @@ import (
 	"bufio"
 	"fmt"
 	"io"
+	"regexp"
 	"sort"
 	"strconv"
 	"strings"
 	"time"
+
+	"golang.org/x/net/html"
 )
 
 // https://www.w3.org/TR/webvtt1/
@@ -24,6 +27,7 @@ const (
 // Vars
 var (
 	bytesWebVTTTimeBoundariesSeparator = []byte(webvttTimeBoundariesSeparator)
+	webVTTRegexpStartTag               = regexp.MustCompile(`(<v([\.\w]*)([\s\w]+)+>)`)
 )
 
 // parseDurationWebVTT parses a .vtt duration
@@ -34,7 +38,6 @@ func parseDurationWebVTT(i string) (time.Duration, error) {
 // ReadFromWebVTT parses a .vtt content
 // TODO Tags (u, i, b)
 // TODO Class
-// TODO Speaker name
 func ReadFromWebVTT(i io.Reader) (o *Subtitles, err error) {
 	// Init
 	o = NewSubtitles()
@@ -193,7 +196,8 @@ func ReadFromWebVTT(i io.Reader) (o *Subtitles, err error) {
 			case webvttBlockNameStyle:
 				// TODO Do something with the style
 			case webvttBlockNameText:
-				item.Lines = append(item.Lines, Line{Items: []LineItem{{Text: line}}})
+				// Voice Tag to extract Speaker Name
+				item.Lines = append(item.Lines, parseTextWebVTT(line))
 			default:
 				// This is the ID
 				index, _ = strconv.Atoi(line)
@@ -203,6 +207,48 @@ func ReadFromWebVTT(i io.Reader) (o *Subtitles, err error) {
 	return
 }
 
+// parseTextWebVTT parses the input line to fill the Line
+func parseTextWebVTT(i string) (o Line) {
+	tr := html.NewTokenizer(strings.NewReader(i))
+	var voiceName, endTag, text string
+	for {
+		t := tr.Next()
+		if err := tr.Err(); err != nil {
+			break
+		}
+
+		switch t {
+		case html.StartTagToken:
+			matches := webVTTRegexpStartTag.FindStringSubmatch(string(tr.Raw()))
+			if matches == nil {
+				return
+			}
+
+			if len(matches[1]) == 0 {
+				return
+			}
+
+			voiceName = matches[3]
+
+		case html.EndTagToken:
+			endTag = string(tr.Raw())
+
+		case html.TextToken:
+			text = string(tr.Raw())
+		}
+	}
+
+	// should be either no tag, or match voice tag
+	if endTag != "" && endTag != "</v>" {
+		return
+	}
+
+	o.VoiceName = strings.TrimSpace(voiceName)
+	o.Items = []LineItem{{Text: strings.TrimSpace(text)}}
+
+	return
+}
+
 // formatDurationWebVTT formats a .vtt duration
 func formatDurationWebVTT(i time.Duration) string {
 	return formatDuration(i, ".", 3)

webvtt_internal_test.go

@@ -0,0 +1,36 @@
+package astisub
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestParseTextWebVTT(t *testing.T) {
+
+	t.Run("When both voice tags are available", func(t *testing.T) {
+		testData := `<v Bob>Correct tag</v>`
+
+		s := parseTextWebVTT(testData)
+		assert.Equal(t, "Bob", s.VoiceName)
+		assert.Equal(t, 1, len(s.Items))
+		assert.Equal(t, "Correct tag", s.Items[0].Text)
+	})
+
+	t.Run("When there is no end tag", func(t *testing.T) {
+		testData := `<v Bob> Text without end tag`
+
+		s := parseTextWebVTT(testData)
+		assert.Equal(t, "Bob", s.VoiceName)
+		assert.Equal(t, 1, len(s.Items))
+		assert.Equal(t, "Text without end tag", s.Items[0].Text)
+	})
+
+	t.Run("When the end tag is correct", func(t *testing.T) {
+		testData := `<v Bob>Incorrect end tag</vi>`
+
+		s := parseTextWebVTT(testData)
+		assert.Equal(t, "", s.VoiceName)
+		assert.Equal(t, 0, len(s.Items))
+	})
+}

webvtt_test.go

@@ -3,6 +3,7 @@ package astisub_test
 import (
 	"bytes"
 	"io/ioutil"
+	"strings"
 	"testing"
 
 	"github.com/asticode/go-astisub"
@@ -44,3 +45,37 @@ func TestBroken1WebVTT(t *testing.T) {
 	_, err := astisub.OpenFile("./testdata/broken-1-in.vtt")
 	assert.Nil(t, err)
 }
+
+func TestWebVTTWithVoiceName(t *testing.T) {
+	voiceName1 := "Roger Bingham"
+	voiceName2 := "Bingham"
+	voiceName3 := "Lee"
+	testData := `WEBVTT
+
+	NOTE this a example with voicename
+
+	1
+	00:02:34.00 --> 00:02:35.00
+	<v.first.local Roger Bingham> I'm the fist speaker
+
+	2
+	00:02:34.00 --> 00:02:35.00
+	<v Bingham> I'm the second speaker
+
+	3
+	00:00:04.000 --> 00:00:08.000
+	<v Lee>What are you doing here?</v>
+
+	4
+	00:00:04.000 --> 00:00:08.000
+	<v Bob>Incorrect tag?</vi>`
+
+	s, err := astisub.ReadFromWebVTT(strings.NewReader(testData))
+	assert.NoError(t, err)
+
+	assert.Len(t, s.Items, 4)
+	assert.Equal(t, voiceName1, s.Items[0].Lines[0].VoiceName)
+	assert.Equal(t, voiceName2, s.Items[1].Lines[0].VoiceName)
+	assert.Equal(t, voiceName3, s.Items[2].Lines[0].VoiceName)
+	assert.Equal(t, "", s.Items[3].Lines[0].VoiceName)
+}