fixes dlclark#24 changes $ behavior in singleline when using RE2 and …

…ECMAScript modes

PCRE and .NET have a different definition of $ than RE2 and ECMAScript
engines in singleline mode.  PCRE defines it as "$ asserts position at the end
of the string, or before the line terminator right at the end of the string (if any)."
This means that a pattern of "^ac$\n" is valid and can match "ac\n" OR "ac".

This behavior is different in RE2 and ECMAScript engines.  For these engines the
pattern "^ac$\n" won't match any inputs in singleline mode because the $ demands the
string ends but the pattern requires an extra \n so they both cannot be true.

The PCRE/.NET behavior feels wrong, but for this project I maintain compatibility with
them in "default" mode.  The other, less suprising behavior is enabled by using either
the RE2 option or the ECMAScript option.

regexp_re2_test.go

@@ -96,3 +96,26 @@ func TestRE2NamedAscii_Concat(t *testing.T) {
 		t.Fatal("Expected match")
 	}
 }
+
+func TestRE2Dollar_Singleline(t *testing.T) {
+	// PCRE allows for \n after the $ and RE2 doesn't
+	r := MustCompile(`^ac$\n`, RE2)
+	if m, _ := r.MatchString("ac"); m {
+		t.Fatal("Expected no match")
+	}
+	if m, _ := r.MatchString("ac\n"); m {
+		t.Fatal("Expected no match")
+	}
+}
+
+func TestRE2Dollar_Multiline(t *testing.T) {
+	r := MustCompile(`^ac$\n`, RE2|Multiline)
+	if m, _ := r.MatchString("ac"); m {
+		t.Fatal("Expected no match")
+	}
+	if m, err := r.MatchString("ac\n"); err != nil {
+		t.Fatal(err)
+	} else if !m {
+		t.Fatal("Expected match")
+	}
+}

regexp_test.go

@@ -758,6 +758,25 @@ func TestECMANegateRange(t *testing.T) {
 	}
 }
 
+func TestDollar(t *testing.T) {
+	// PCRE/C# allow \n to match to $ at end-of-string in singleline mode...
+	// a weird edge-case kept for compatibility, ECMAScript/RE2 mode don't allow it
+	re := MustCompile(`ac$`, 0)
+	if m, err := re.MatchString("ac\n"); err != nil {
+		t.Fatal(err)
+	} else if !m {
+		t.Fatal("Expected match")
+	}
+}
+func TestECMADollar(t *testing.T) {
+	re := MustCompile(`ac$`, ECMAScript)
+	if m, err := re.MatchString("ac\n"); err != nil {
+		t.Fatal(err)
+	} else if m {
+		t.Fatal("Expected no match")
+	}
+}
+
 func TestThreeByteUnicode_InputOnly(t *testing.T) {
 	// confirm the bmprefix properly ignores 3-byte unicode in the input value
 	// this used to panic

runner.go

@@ -566,9 +566,22 @@ func (r *runner) execute() error {
 			continue
 
 		case syntax.EndZ:
-			if r.rightchars() > 1 || r.rightchars() == 1 && r.charAt(r.textPos()) != '\n' {
+			rchars := r.rightchars()
+			if rchars > 1 {
 				break
 			}
+			// RE2 and EcmaScript define $ as "asserts position at the end of the string"
+			// PCRE/.NET adds "or before the line terminator right at the end of the string (if any)"
+			if (r.re.options & (RE2 | ECMAScript)) != 0 {
+				// RE2/Ecmascript mode
+				if rchars > 0 {
+					break
+				}
+			} else if rchars == 1 && r.charAt(r.textPos()) != '\n' {
+				// "regular" mode
+				break
+			}
+
 			r.advance(0)
 			continue