lang code parsing

errors.go

@@ -4,5 +4,8 @@ import "git.milar.in/milarin/adverr"
 
 var (
 	ErrTorrentParserInsufficientData           = adverr.NewErrTmpl("ErrTorrentParserInsufficientData", "regex '%s' must at least provide title and episode")
+	ErrTorrentParserInsufficientLanguageData   = adverr.NewErrTmpl("ErrTorrentParserInsufficientLanguageData", "no language reference in regex and no default language set")
+	ErrTorrentParserInsufficientSubtitleData   = adverr.NewErrTmpl("ErrTorrentParserInsufficientSubtitleData", "no subtitle reference in regex and no default subtitle set")
+	ErrTorrentParserInsufficientResolutionData = adverr.NewErrTmpl("ErrTorrentParserInsufficientResolutionData", "no resolution reference in regex and no default resolution set")
 	ErrTorrentParserInvalidGroupReference      = adverr.NewErrTmpl("ErrTorrentParserInvalidGroupReference", "options references group %d but regex only has %d groups")
 )

go.mod

@@ -3,7 +3,7 @@ module git.milar.in/animan/parsers
 go 1.18
 
 require (
-	git.milar.in/animan/model v0.0.0-20220804154959-f8b66e8c9eda
+	git.milar.in/animan/model v0.0.0-20220815093549-fe9b565a9dd2
 	git.milar.in/milarin/adverr v0.2.1
 )
 

go.sum

@@ -1,5 +1,7 @@
 git.milar.in/animan/model v0.0.0-20220804154959-f8b66e8c9eda h1:9XG7Dn+UtyLwwNcuol10EdHcBQjYD+1WzmnJbRUJQfs=
 git.milar.in/animan/model v0.0.0-20220804154959-f8b66e8c9eda/go.mod h1:oXDr3slnzXoccIrci2wotX0cWwMVuAQ9dSvmh1buE4c=
+git.milar.in/animan/model v0.0.0-20220815093549-fe9b565a9dd2 h1:vNbYNp8VJQ6FBENgZMHA1IFOfPkQEA+TIaN8PFF8m48=
+git.milar.in/animan/model v0.0.0-20220815093549-fe9b565a9dd2/go.mod h1:oXDr3slnzXoccIrci2wotX0cWwMVuAQ9dSvmh1buE4c=
 git.milar.in/milarin/adverr v0.2.1 h1:eyXFGC+Ui/kcNt2+NqP3HiAplwxzqeNr9DfitsUb3c4=
 git.milar.in/milarin/adverr v0.2.1/go.mod h1:wwfglcey4R3vqjNL/d8mbnvFJGzETRXzAEolIHZY32w=
 git.milar.in/milarin/anilist v1.5.0 h1:fSiAXY/topNk4ISEp2QtcG9HHKLJfMc8w05iqc+Paf0=

lang_codes.go

@@ -0,0 +1,144 @@
+package parsers
+
+import "strings"
+
+// langSynonyms converts all irregular lang codes to ISO 639-1
+var langSynonyms = map[string]string{
+	// english
+	"ENG": "en", // Erai-Raws
+	"US":  "en", // Erai-Raws
+
+	// portuguese
+	"POR-BR": "pt", // Erai-Raws
+	"POR":    "pt", // Erai-Raws
+	"BR":     "pt", // Erai-Raws
+
+	// spanish
+	"SPA-LA": "es", // Erai-Raws
+	"SPA":    "es", // Erai-Raws
+	"MX":     "es", // Erai-Raws
+
+	// arabic
+	"ARA": "ar", // Erai-Raws
+	"SA":  "ar", // Erai-Raws
+
+	// french
+	"FRE": "fr", // Erai-Raws
+
+	// german
+	"GER": "de", // Erai-Raws
+
+	// italian
+	"ITA": "it", // Erai-Raws
+
+	// russian
+	"RUS": "ru", // Erai-Raws
+
+	// japanese
+	"JPN": "ja", // Erai-Raws
+	"JP":  "ja", // Erai-Raws
+
+	// polish
+	"POL": "pl", // Erai-Raws
+
+	// dutch
+	"DUT": "nl", // Erai-Raws
+
+	// norwegian
+	"NOB": "no", // Erai-Raws
+
+	// turkish
+	"TUR": "tr", // Erai-Raws
+
+	// swedish
+	"SWE": "sv", // Erai-Raws
+	"SE":  "sv", // Erai-Raws
+
+	// greek
+	"GRE": "el", // Erai-Raws
+	"GR":  "el", // Erai-Raws
+
+	// hebrew
+	"HEB": "he", // Erai-Raws
+	"IL":  "he", // Erai-Raws
+
+	// romanian
+	"RUM": "ro", // Erai-Raws
+
+	// indonesian
+	"IND": "id", // Erai-Raws
+
+	// thai
+	"THA": "th", // Erai-Raws
+
+	// korean
+	"KOR": "ko", // Erai-Raws
+	"KR":  "ko", // Erai-Raws
+
+	// danish
+	"DAN": "da", // Erai-Raws
+	"DK":  "da", // Erai-Raws
+
+	// chinese (simplified & traditional)
+	"CHI": "zh", // Erai-Raws
+	"CN":  "zh", // Erai-Raws
+
+	// bulgarian
+	"BUL": "bg", // Erai-Raws
+
+	// vietnamese
+	"VIE": "vi", // Erai-Raws
+	"VN":  "vi", // Erai-Raws
+
+	// hindi
+	"HIN": "hi", // Erai-Raws
+	"IN":  "hi", // Erai-Raws
+
+	// tamil
+	"TEL": "ta", // Erai-Raws
+	"LK":  "ta", // Erai-Raws
+
+	// ukrainian
+	"UKR": "uk", // Erai-Raws
+	"UA":  "uk", // Erai-Raws
+
+	// hungarian
+	"HUN": "hu", // Erai-Raws
+
+	// czech
+	"CES": "cs", // Erai-Raws
+	"CZ":  "cs", // Erai-Raws
+
+	// croatian
+	"HRV": "hr", // Erai-Raws
+
+	// malaysian
+	"MAY": "ms", // Erai-Raws
+	"MY":  "ms", // Erai-Raws
+
+	// slovakian
+	"SLK": "sk", // Erai-Raws
+
+	// filipino
+	"FIL": "tl", // Erai-Raws
+	"PH":  "tl", // Erai-Raws
+}
+
+// ParseLanguage converts irregular language codes to ISO 639-1
+func ParseLanguage(str string) string {
+	if code, ok := langSynonyms[str]; ok {
+		return code
+	}
+
+	return strings.ToLower(str)
+}
+
+// ParseLanguages converts multiple irregular language codes to ISO 639-1.
+// It simply calls ParseLanguage for each language code
+func ParseLanguages(langCodes []string) []string {
+	codes := make([]string, 0, len(langCodes))
+	for _, irregularLangCode := range langCodes {
+		codes = append(codes, ParseLanguage(irregularLangCode))
+	}
+	return codes
+}

parsers.go

@@ -22,8 +22,9 @@ var Parsers = []model.Parser{
 				Name:             1,
 				Episode:          2,
 				Resolution:       3,
-				Languages:      4,
+				Subtitles:        4,
-				LanguageParser: SquareBracketsLanguageParser,
+				SubtitleParser:   SquareBracketsLanguageParser,
+				DefaultLanguages: []string{"ja"},
 			},
 		),
 		FileParser: nil,

torrent_parsing.go

@@ -9,22 +9,38 @@ import (
 // TorrentParseOptions holds the subgroup index in which information can be found in a given regex
 // as well as some parser specific functions
 type TorrentParseOptions struct {
+	// regex group references
 	Name       int
 	Episode    int
 	Languages  int
 	Subtitles  int
 	Resolution int
 
+	// language parsers
 	LanguageParser LanguageParserFunc
+	SubtitleParser LanguageParserFunc
+
+	// default values used when group reference is 0
+	DefaultLanguages  []string
+	DefaultSubtitles  []string
+	DefaultResolution model.Resolution
 }
 
 func regexTorrentParser(regex string, options TorrentParseOptions) model.TorrentParserFunc {
 	pattern := regexp.MustCompile(regex)
 
+	// handle faulty regexes
 	if options.Name == 0 || options.Episode == 0 {
 		panic(ErrTorrentParserInsufficientData.New(regex))
+	} else if options.Languages == 0 && options.DefaultLanguages == nil {
+		panic(ErrTorrentParserInsufficientLanguageData.New(regex))
+	} else if options.Subtitles == 0 && options.DefaultSubtitles == nil {
+		panic(ErrTorrentParserInsufficientSubtitleData.New(regex))
+	} else if options.Resolution == 0 && options.DefaultResolution == 0 {
+		panic(ErrTorrentParserInsufficientResolutionData.New(regex))
 	}
 
+	// handle faulty group references
 	for _, g := range []int{options.Name, options.Episode, options.Languages, options.Subtitles, options.Resolution} {
 		if g > pattern.NumSubexp() {
 			panic(ErrTorrentParserInvalidGroupReference.New(g, pattern.NumSubexp()))
@@ -32,6 +48,8 @@ func regexTorrentParser(regex string, options TorrentParseOptions) model.Torrent
 	}
 
 	return func(parser *model.Parser, torrent *model.Torrent) (ParsedTorrent *model.ParsedTorrent, ok bool) {
+		var err error
+
 		matches := pattern.FindStringSubmatch(torrent.Title)
 
 		if matches == nil {
@@ -43,17 +61,31 @@ func regexTorrentParser(regex string, options TorrentParseOptions) model.Torrent
 			return nil, false
 		}
 
-		resolution, err := model.ParseResolution(matches[options.Resolution])
+		resolution := options.DefaultResolution
+		if options.Resolution != 0 {
+			resolution, err = model.ParseResolution(matches[options.Resolution])
 			if err != nil {
 				return nil, false
 			}
+		}
+
+		languages := options.DefaultLanguages
+		if options.Languages != 0 {
+			languages = options.LanguageParser(matches[options.Languages])
+		}
+
+		subtitles := options.DefaultSubtitles
+		if options.Subtitles != 0 {
+			subtitles = options.SubtitleParser(matches[options.Subtitles])
+		}
 
 		return &model.ParsedTorrent{
-			Title:      matches[options.Name],
+			OriginalAnimeTitle: matches[options.Name],
 			Episode:            episode,
 			Resolution:         resolution,
 			Parser:             parser,
-			Languages:  options.LanguageParser(matches[options.Languages]),
+			Languages:          ParseLanguages(languages),
+			Subtitles:          ParseLanguages(subtitles),
 
 			Torrent: torrent,
 		}, true