feat(subtitles): Add interactive subtitle track selection

This change introduces a new feature that allows users to interactively select a subtitle track from a list of available tracks in an MKV file.

Previously, the application would automatically select the best English subtitle track. Now, it displays a list of all available subtitle tracks with their languages and prompts the user to choose one. If no selection is made, it falls back to the original behavior of selecting the best English track.

This provides more flexibility for users who want to extract subtitles in languages other than English or choose from multiple subtitle tracks for the same language.

Author: luispater

internal/video/mkv.go

@@ -6,12 +6,15 @@ import (
 	"io"
 	"os"
 	"path/filepath"
+	"strconv"
 	"strings"
 	"time"
 
 	"github.com/luispater/matroska-go"
 
+	"github.com/luispater/gemini-srt-translator-go/internal/logger"
 	"github.com/luispater/gemini-srt-translator-go/pkg/errors"
+	"github.com/luispater/gemini-srt-translator-go/pkg/languages"
 	"github.com/luispater/gemini-srt-translator-go/pkg/srt"
 )
 
@@ -265,18 +268,56 @@ func ExtractSubtitlesFromMKV(mkvPath string) (string, error) {
 		return "", err
 	}
 
-	// Select the best English subtitle track
-	track, err := parser.SelectBestEnglishTrack()
-	if err != nil {
-		return "", err
+	tracks := parser.GetSubtitleTracks()
+	if len(tracks) == 0 {
+		return "", errors.NewValidationError("no subtitle tracks found in MKV", nil)
 	}
 
-	// Generate output path
+	logger.Info("Available subtitle tracks:")
+	for i, tr := range tracks {
+		lines := len(tr.Entries)
+		name := strings.TrimSpace(tr.Name)
+		lang := tr.Language
+		bcp := languages.BCP47FromMKV(lang)
+		if bcp != "" {
+			if name != "" {
+				lang = fmt.Sprintf("%s (%s)", bcp, name)
+			} else {
+				lang = fmt.Sprintf("%s", bcp)
+			}
+		}
+		logger.Info(fmt.Sprintf("[%d] Language: %s, Lines:%d", i+1, lang, lines))
+	}
+
+	var selectedIdx int
+	for {
+		input := logger.InputPrompt("Select track number to extract: ")
+		input = strings.TrimSpace(input)
+		if input == "" {
+			best, errSel := parser.SelectBestEnglishTrack()
+			if errSel == nil && best != nil {
+				for i := range tracks {
+					if tracks[i].Number == best.Number {
+						selectedIdx = i
+						break
+					}
+				}
+				break
+			}
+		}
+		if n, errAtoi := strconv.Atoi(input); errAtoi == nil && n >= 1 && n <= len(tracks) {
+			selectedIdx = n - 1
+			break
+		}
+		logger.Warning("Invalid selection. Enter a valid number.")
+	}
+
+	selected := tracks[selectedIdx]
+
 	baseName := strings.TrimSuffix(filepath.Base(mkvPath), filepath.Ext(mkvPath))
 	outputPath := filepath.Join(filepath.Dir(mkvPath), baseName+"_extracted.srt")
 
-	// Extract to SRT
-	if err = parser.ExtractToSRT(track, outputPath); err != nil {
+	if err := parser.ExtractToSRT(&selected, outputPath); err != nil {
 		return "", err
 	}
 

pkg/languages/languages.go

@@ -1,6 +1,5 @@
 package languages
 
-// LanguageMap contains mappings from language names to language codes
 var LanguageMap = map[string]string{
 	// Major World Languages
 	"arabic":     "ar",
@@ -154,3 +153,125 @@ func GetLanguageCode(languageName string) (string, bool) {
 	code, exists := LanguageMap[languageName]
 	return code, exists
 }
+
+func BCP47FromMKV(code string) string {
+	c := code
+	if c == "" {
+		return "Undetermined"
+	}
+	c = normalize(c)
+	switch c {
+	case "en", "eng":
+		return "English"
+	case "es", "spa":
+		return "Spanish"
+	case "fr", "fre", "fra":
+		return "French"
+	case "de", "ger", "deu":
+		return "German"
+	case "pt", "por":
+		return "Portuguese"
+	case "zh", "zho", "chi", "cmn":
+		return "Chinese"
+	case "ja", "jpn":
+		return "Japanese"
+	case "ko", "kor":
+		return "Korean"
+	case "it", "ita":
+		return "Italian"
+	case "ru", "rus":
+		return "Russian"
+	case "ar", "ara":
+		return "Arabic"
+	case "hi", "hin":
+		return "Hindi"
+	case "nl", "dut", "nld":
+		return "Dutch"
+	case "sv", "swe":
+		return "Swedish"
+	case "pl", "pol":
+		return "Polish"
+	case "tr", "tur":
+		return "Turkish"
+	case "he", "heb":
+		return "Hebrew"
+	case "uk", "ukr":
+		return "Ukrainian"
+	case "cs", "cze", "ces":
+		return "Czech"
+	case "ro", "rum", "ron":
+		return "Romanian"
+	case "vi", "vie":
+		return "Vietnamese"
+	case "th", "tha":
+		return "Thai"
+	case "id", "ind":
+		return "Indonesian"
+	case "fa", "per", "fas":
+		return "Persian"
+	case "el", "gre", "ell":
+		return "Greek"
+	case "hu", "hun":
+		return "Hungarian"
+	case "fi", "fin":
+		return "Finnish"
+	case "no", "nor":
+		return "Norwegian"
+	case "da", "dan":
+		return "Danish"
+	case "bg", "bul":
+		return "Bulgarian"
+	case "sr", "srp":
+		return "Serbian"
+	case "hr", "hrv":
+		return "Croatian"
+	case "sk", "slk", "slo":
+		return "Slovak"
+	case "sl", "slv":
+		return "Slovenian"
+	case "lt", "lit":
+		return "Lithuanian"
+	case "lv", "lav":
+		return "Latvian"
+	case "et", "est":
+		return "Estonian"
+	case "ms", "msa", "may":
+		return "Malay"
+	case "bn", "ben":
+		return "Bengali"
+	case "ur", "urd":
+		return "Urdu"
+	case "ta", "tam":
+		return "Tamil"
+	case "te", "tel":
+		return "Telugu"
+	case "ml", "mal":
+		return "Malayalam"
+	case "mr", "mar":
+		return "Marathi"
+	case "ne", "nep":
+		return "Nepali"
+	case "km", "khm":
+		return "Khmer"
+	case "my", "bur", "mya":
+		return "Burmese"
+	case "bo", "tib":
+		return "Tibetan"
+	case "id-id":
+		return "Indonesian"
+	}
+	return c
+}
+
+func normalize(s string) string {
+	b := make([]byte, 0, len(s))
+	for i := 0; i < len(s); i++ {
+		ch := s[i]
+		if ch >= 'A' && ch <= 'Z' {
+			b = append(b, ch+32)
+			continue
+		}
+		b = append(b, ch)
+	}
+	return string(b)
+}