ST1018: don't flag ZWJ that are part of emoji sequences

Closes gh-1113

stylecheck/lint.go

@@ -690,14 +690,21 @@ func CheckInvisibleCharacters(pass *analysis.Pass) (interface{}, error) {
 		var invalids []invalid
 		hasFormat := false
 		hasControl := false
+		prev := rune(-1)
+		const zwj = '\u200d'
 		for off, r := range lit.Value {
 			if unicode.Is(unicode.Cf, r) {
-				invalids = append(invalids, invalid{r, off})
-				hasFormat = true
+				// Don't flag joined emojis. These are multiple emojis joined with ZWJ, which some platform render as single composite emojis.
+				// For the purpose of this check, we consider all symbols, including all symbol modifiers, emoji.
+				if r != zwj || (r == zwj && !unicode.Is(unicode.S, prev)) {
+					invalids = append(invalids, invalid{r, off})
+					hasFormat = true
+				}
 			} else if unicode.Is(unicode.Cc, r) && r != '\n' && r != '\t' && r != '\r' {
 				invalids = append(invalids, invalid{r, off})
 				hasControl = true
 			}
+			prev = r
 		}
 
 		switch len(invalids) {

stylecheck/testdata/src/CheckInvisibleCharacters/CheckInvisibleCharacters.go

@@ -9,4 +9,6 @@ var (
 est`
 	e = `Zero​Width` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬" //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
 )

stylecheck/testdata/src/CheckInvisibleCharacters/CheckInvisibleCharacters.go.golden

@@ -10,6 +10,8 @@ var (
 est`
 	e = `Zero​Width` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬\a" //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
 )
 
 -- delete control character U+0007 --
@@ -24,6 +26,8 @@ var (
 est`
 	e = `Zero​Width` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬"   //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
 )
 
 -- delete format character U+200B --
@@ -38,6 +42,8 @@ var (
 est`
 	e = `ZeroWidth` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬"  //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
 )
 
 -- replace format character U+200B with '\u200b' --
@@ -52,6 +58,8 @@ var (
 est`
 	e = `Zero\u200bWidth` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬"  //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
 )
 
 -- delete all control characters --
@@ -66,6 +74,8 @@ var (
 est`
 	e = `Zero​Width` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬"  //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
 )
 
 -- replace all control characters with escape sequences --
@@ -80,4 +90,39 @@ var (
 est`
 	e = `Zero​Width` // want `Unicode format character U\+200B`
 	f = "\u200b"
+	g = "👩🏽‍🔬"  //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬​" // want `Unicode format and control characters`
+)
+
+
+-- delete all format and control characters --
+// Package pkg ...
+package pkg
+
+var (
+	a = ""  // want `Unicode control character U\+0007`
+	b = "" // want `Unicode control characters`
+	c = "Test	test"
+	d = `T
+est`
+	e = `Zero​Width` // want `Unicode format character U\+200B`
+	f = "\u200b"
+	g = "👩🏽‍🔬" //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬"  // want `Unicode format and control characters`
+)
+
+-- replace all format and control characters with escape sequences --
+// Package pkg ...
+package pkg
+
+var (
+	a = ""  // want `Unicode control character U\+0007`
+	b = "" // want `Unicode control characters`
+	c = "Test	test"
+	d = `T
+est`
+	e = `Zero​Width` // want `Unicode format character U\+200B`
+	f = "\u200b"
+	g = "👩🏽‍🔬"        //  want `Unicode control character U\+0007`
+	h = "👩🏽‍🔬\a\u200b" // want `Unicode format and control characters`
 )