fix: clean name respects special characters

parse/capitalize.go

@@ -109,6 +109,9 @@ func Capitalize(str string) string {
 	str = breakRegex.ReplaceAllStringFunc(str, func(word string) string {
 		return strings.Title(word)
 	})
+	str = abbreviationRegexp.ReplaceAllStringFunc(str, func(abbr string) string {
+		return strings.ToUpper(abbr)
+	})
 	str = strings.Replace(str, "'S", "'s", -1)
 	return str
 }

parse/capitalize_test.go

@@ -8,37 +8,7 @@ import (
 )
 
 func TestCapitalize(t *testing.T) {
-	for _, v := range []string{
-		"Gone with the Wind",
-		"The Shawshank Redemption",
-		"The Godfather: Part II",
-		"Schindler's List",
-		"The Lord of the Rings: The Return of the King",
-		"The Good, the Bad and the Ugly",
-		"12 Angry Men",
-		"Avengers: Infinity War",
-		"The Lord of the Rings: The Fellowship of the Ring",
-		"Star Wars: Episode V - The Empire Strikes Back",
-		"One Flew Over the Cuckoo's Nest",
-		"The Silence of the Lambs",
-		"Léon: The Professional",
-		"Se7en",
-		"Star Wars: Episode IV - A New Hope",
-		"City of God",
-		"Life Is Beautiful",
-		"Once Upon a Time in America",
-		"21 and Over",
-		"2001: A Space Odyssey",
-		"To Kill a Mockingbird",
-		"Monty Python and the Holy Grail",
-		"L.A. Confidential",
-		"Lock, Stock and Two Smoking Barrels",
-		"Mr. Smith Goes to Washington",
-		"V for Vendetta",
-		"Kill Bill: Vol. 1",
-		"Agents of S.H.I.E.L.D.",
-		"Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb",
-	} {
+	for _, v := range testMovieTitles {
 		assert.Equal(t, v, Capitalize(strings.ToLower(v)))
 	}
 }

parse/parse.go

@@ -64,8 +64,25 @@ func Filename(filename string) string {
 	return strings.TrimSuffix(f, filepath.Ext(f))
 }
 
-var abbreviationRegexp = regexp.MustCompile(`\s[A-Z]\s[A-Z](\s[A-Z])*\s`)
-var illegalcharsRegexp = regexp.MustCompile(`[^\p{L}0-9\s&'_\(\)-]`)
+var abbreviationList = []string{
+	"mr",
+	"mrs",
+	"dr",
+	"vol",
+}
+
+func isAbbreviation(str string) bool {
+	lower := strings.ToLower(str)
+	for _, v := range abbreviationList {
+		if lower == v {
+			return true
+		}
+	}
+	return false
+}
+
+var abbreviationRegexp = regexp.MustCompile(`\b[A-Za-z]([\s\.][A-Za-z])+\b`)
+var illegalcharsRegexp = regexp.MustCompile(`[^\p{L}0-9\s&'_\(\)\-,:]`)
 var spaceReplaceRegexp = regexp.MustCompile(`[\.\s_]+`)
 
 // CleanName returns the media name cleaned from punctuation
@@ -74,7 +91,14 @@ func CleanName(name string) string {
 	name = illegalcharsRegexp.ReplaceAllString(name, "")
 
 	name = abbreviationRegexp.ReplaceAllStringFunc(name, func(match string) string {
-		return " " + strings.Replace(match, " ", "", -1) + " "
+		return strings.Replace(match, " ", ".", -1) + "."
+	})
+
+	name = wordRegex.ReplaceAllStringFunc(name, func(match string) string {
+		if isAbbreviation(match) {
+			return match + "."
+		}
+		return match
 	})
 
 	name = Capitalize(name)

parse/parse_test.go

@@ -1,11 +1,56 @@
 package parse
 
 import (
+	"regexp"
+	"strings"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
 )
 
+var testMovieTitles = []string{
+	"Gone with the Wind",
+	"The Shawshank Redemption",
+	"The Godfather: Part II",
+	"Schindler's List",
+	"The Lord of the Rings: The Return of the King",
+	"The Good, the Bad and the Ugly",
+	"12 Angry Men",
+	"Avengers: Infinity War",
+	"The Lord of the Rings: The Fellowship of the Ring",
+	"Star Wars: Episode V - The Empire Strikes Back",
+	"One Flew Over the Cuckoo's Nest",
+	"The Silence of the Lambs",
+	"Léon: The Professional",
+	"Se7en",
+	"Star Wars: Episode IV - A New Hope",
+	"City of God",
+	"Life Is Beautiful",
+	"Once Upon a Time in America",
+	"21 and Over",
+	"2001: A Space Odyssey",
+	"To Kill a Mockingbird",
+	"Monty Python and the Holy Grail",
+	"L.A. Confidential",
+	"Lock, Stock and Two Smoking Barrels",
+	"Mr. Smith Goes to Washington",
+	"V for Vendetta",
+	"Kill Bill: Vol. 1",
+	"Agents of S.H.I.E.L.D.",
+	"Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb",
+	"X-Men Origins: Wolverine",
+	"Mr. & Mrs. Smith",
+}
+
+func TestCleanNameMovieTitles(t *testing.T) {
+	space := regexp.MustCompile(`[\s\.]+`)
+
+	for _, v := range testMovieTitles {
+		s := space.ReplaceAllString(v, ".")
+		assert.Equal(t, v, CleanName(strings.ToLower(s)))
+	}
+}
+
 func TestIdentity(t *testing.T) {
 	assert.Equal(t, "thisisatest", Identity("thìs is â tést"))
 	assert.Equal(t, "vyzkousejtetentoretezec", Identity("vyzkoušejte tento řetězec"))
@@ -15,9 +60,9 @@ func TestIdentity(t *testing.T) {
 
 func TestCleanName(t *testing.T) {
 	assert.Equal(t, "This Is a Test", CleanName("this.is.a.test"))
-	assert.Equal(t, "This Is a (Test)", CleanName("this?_=is#.,a_(test)"))
-	assert.Equal(t, "Abc ABC Abc", CleanName("abc.A.B.C.abc"))
-	assert.Equal(t, "Abc ABC Abc", CleanName("abc A B C abc"))
+	assert.Equal(t, "This Is a (Test)", CleanName("this?_=is#.a_(test)"))
+	assert.Equal(t, "Abc A.B.C. Abc", CleanName("abc.A.B.C.abc"))
+	assert.Equal(t, "Abc A.B.C. Abc", CleanName("abc A B C abc"))
 	assert.Equal(t, "A Good Day to Die Hard", CleanName("A.Good.Day.To.Die.Hard"))
 	assert.Equal(t, "This Is a Test", CleanName("This.Is.A.Test"))
 }

provider/subscene.go

@@ -38,6 +38,8 @@ const subsceneDelay = 500 * time.Millisecond
 
 var subsceneLock = new(sync.Mutex)
 
+var subsceneIllegal = regexp.MustCompile(`[^\p{L}0-9\s]`)
+
 // lockSubscene is used to limit the number of calls to subscene to prevent spamming
 func lockSubscene() {
 	subsceneLock.Lock()
@@ -59,12 +61,13 @@ func (s *subscene) ResolveSubtitle(l types.Linker) (types.Downloadable, error) {
 }
 
 func (s *subscene) searchTerm(m types.Media) string {
+	var term string
 	if movie, ok := m.TypeMovie(); ok {
-		return s.searchTermMovie(movie)
+		term = s.searchTermMovie(movie)
 	} else if episode, ok := m.TypeEpisode(); ok {
-		return s.searchTermEpisode(episode)
+		term = s.searchTermEpisode(episode)
 	}
-	return ""
+	return subsceneIllegal.ReplaceAllString(term, "")
 }
 
 func (s *subscene) searchTermMovie(movie types.Movie) string {