re2 1.18.2 → 1.19.0

package/.prettierrc

@@ -1,5 +1,5 @@
 {
-  "printWidth": 120,
+  "printWidth": 80,
   "singleQuote": true,
   "bracketSpacing": false,
   "arrowParens": "avoid",

package/README.md

@@ -58,6 +58,7 @@ Supported properties:
 * [`re2.unicode`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/unicode)
   * `RE2` engine always works in the Unicode mode. See details below.
 * [`re2.sticky`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/sticky)
+* *Since 1.19.0*: [`re2.hasIndices`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/hasIndices)
 * [`re2.source`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/source)
 * [`re2.flags`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/flags)
 
@@ -352,6 +353,8 @@ console.log('re2_res : ' + re2_res); // prints: re2_res : abc,a,b,c
 
 ## Release history
 
+- 1.19.0 *Added `hasIndices` AKA the `d` flag. Thx, [teebu](https://github.com/teebu).*
+- 1.18.3 *Fixed bug with non-matched groups. Thx, [Dan Setterquist](https://github.com/dset).*
 - 1.18.2 *Reference to the binary module by its full name.*
 - 1.18.1 *Support for Node 16, 18, 20 + Darwin arm64 precompiled binaries.*
 - 1.18.0 *Modified TS bindings, added a type test (thx, [Kenichi Kamiya](https://github.com/kachick) and [Jamie Magee](https://github.com/JamieMagee)).*

package/lib/accessors.cc

@@ -39,9 +39,13 @@ NAN_GETTER(WrappedRE2::GetFlags)
 	auto re2 = Nan::ObjectWrap::Unwrap<WrappedRE2>(info.This());
 
 	std::string flags;
+	if (re2->hasIndices)
+	{
+		flags = "d";
+	}
 	if (re2->global)
 	{
-		flags = "g";
+		flags += "g";
 	}
 	if (re2->ignoreCase)
 	{
@@ -135,6 +139,18 @@ NAN_GETTER(WrappedRE2::GetSticky)
 	info.GetReturnValue().Set(re2->sticky);
 }
 
+NAN_GETTER(WrappedRE2::GetHasIndices)
+{
+	if (!WrappedRE2::HasInstance(info.This()))
+	{
+		info.GetReturnValue().SetUndefined();
+		return;
+	}
+
+	auto re2 = Nan::ObjectWrap::Unwrap<WrappedRE2>(info.This());
+	info.GetReturnValue().Set(re2->hasIndices);
+}
+
 NAN_GETTER(WrappedRE2::GetLastIndex)
 {
 	if (!WrappedRE2::HasInstance(info.This()))

package/lib/addon.cc

@@ -23,9 +23,9 @@ static NAN_METHOD(GetUtf16Length)
 	info.GetReturnValue().Set(-1);
 }
 
-static void cleanup(void* p)
+static void cleanup(void *p)
 {
-	v8::Isolate* isolate = static_cast<v8::Isolate*>(p);
+	v8::Isolate *isolate = static_cast<v8::Isolate *>(p);
 	auto p_tpl = Nan::GetIsolateData<Nan::Persistent<v8::FunctionTemplate>>(isolate);
 	delete p_tpl;
 }
@@ -71,6 +71,7 @@ v8::Local<v8::Function> WrappedRE2::Init()
 	Nan::SetAccessor(instanceTemplate, Nan::New("dotAll").ToLocalChecked(), GetDotAll);
 	Nan::SetAccessor(instanceTemplate, Nan::New("unicode").ToLocalChecked(), GetUnicode);
 	Nan::SetAccessor(instanceTemplate, Nan::New("sticky").ToLocalChecked(), GetSticky);
+	Nan::SetAccessor(instanceTemplate, Nan::New("hasIndices").ToLocalChecked(), GetHasIndices);
 	Nan::SetAccessor(instanceTemplate, Nan::New("lastIndex").ToLocalChecked(), GetLastIndex, SetLastIndex);
 	Nan::SetAccessor(instanceTemplate, Nan::New("internalSource").ToLocalChecked(), GetInternalSource);
 

package/lib/exec.cc

@@ -52,8 +52,9 @@ NAN_METHOD(WrappedRE2::Exec)
 			{
 				size_t s = getUtf8CharSize(str.data[lastIndex]);
 				lastIndex += s;
-				if (s == 4 && n >= 2) --n; // this utf8 character will take two utf16 characters
-				// the decrement above is protected to avoid an overflow of an unsigned integer
+				if (s == 4 && n >= 2)
+					--n; // this utf8 character will take two utf16 characters
+						 // the decrement above is protected to avoid an overflow of an unsigned integer
 			}
 		}
 	}
@@ -74,7 +75,7 @@ NAN_METHOD(WrappedRE2::Exec)
 
 	// form a result
 
-	auto result = Nan::New<v8::Array>();
+	auto result = Nan::New<v8::Array>(), indices = Nan::New<v8::Array>();
 	int indexOffset = re2->global || re2->sticky ? re2->lastIndex : 0;
 
 	if (str.isBuffer)
@@ -82,13 +83,24 @@ NAN_METHOD(WrappedRE2::Exec)
 		for (size_t i = 0, n = groups.size(); i < n; ++i)
 		{
 			const auto &item = groups[i];
-			if (item.data() != NULL)
+			const auto data = item.data();
+			if (data)
 			{
-				Nan::Set(result, i, Nan::CopyBuffer(item.data(), item.size()).ToLocalChecked());
+				Nan::Set(result, i, Nan::CopyBuffer(data, item.size()).ToLocalChecked());
+				if (re2->hasIndices) {
+					auto pair = Nan::New<v8::Array>();
+					auto offset = data - str.data;
+					Nan::Set(pair, 0, Nan::New<v8::Integer>(static_cast<int>(offset)));
+					Nan::Set(pair, 1, Nan::New<v8::Integer>(static_cast<int>(offset + item.size())));
+					Nan::Set(indices, i, pair);
+				}
 			}
 			else
 			{
 				Nan::Set(result, i, Nan::Undefined());
+				if (re2->hasIndices) {
+					Nan::Set(indices, i, Nan::Undefined());
+				}
 			}
 		}
 		Nan::Set(result, Nan::New("index").ToLocalChecked(), Nan::New<v8::Integer>(indexOffset + static_cast<int>(groups[0].data() - str.data)));
@@ -98,18 +110,35 @@ NAN_METHOD(WrappedRE2::Exec)
 		for (size_t i = 0, n = groups.size(); i < n; ++i)
 		{
 			const auto &item = groups[i];
-			if (item.data() != NULL)
+			const auto data = item.data();
+			if (data)
 			{
-				Nan::Set(result, i, Nan::New(item.data(), item.size()).ToLocalChecked());
+				Nan::Set(result, i, Nan::New(data, item.size()).ToLocalChecked());
+				if (re2->hasIndices) {
+					auto pair = Nan::New<v8::Array>();
+					auto offset = getUtf16Length(str.data + lastIndex, data);
+					auto length = getUtf16Length(data, data + item.size());
+					Nan::Set(pair, 0, Nan::New<v8::Integer>(static_cast<int>(offset)));
+					Nan::Set(pair, 1, Nan::New<v8::Integer>(static_cast<int>(offset + length)));
+					Nan::Set(indices, i, pair);
+				}
 			}
 			else
 			{
 				Nan::Set(result, i, Nan::Undefined());
+				if (re2->hasIndices) {
+					Nan::Set(indices, i, Nan::Undefined());
+				}
 			}
 		}
 		Nan::Set(result, Nan::New("index").ToLocalChecked(), Nan::New<v8::Integer>(indexOffset + static_cast<int>(getUtf16Length(str.data + lastIndex, groups[0].data()))));
 	}
 
+	if (re2->global || re2->sticky)
+	{
+		re2->lastIndex += str.isBuffer ? groups[0].data() - str.data + groups[0].size() - lastIndex : getUtf16Length(str.data + lastIndex, groups[0].data() + groups[0].size());
+	}
+
 	Nan::Set(result, Nan::New("input").ToLocalChecked(), info[0]);
 
 	const auto &groupNames = re2->regexp.CapturingGroupNames();
@@ -128,15 +157,33 @@ NAN_METHOD(WrappedRE2::Exec)
 		}
 
 		Nan::Set(result, Nan::New("groups").ToLocalChecked(), groups);
+
+		if (re2->hasIndices) {
+			auto indexGroups = Nan::New<v8::Object>();
+			Nan::SetPrototype(indexGroups, Nan::Null());
+
+			for (auto group : groupNames)
+			{
+				auto value = Nan::Get(indices, group.first);
+				if (!value.IsEmpty())
+				{
+					Nan::Set(indexGroups, Nan::New(group.second).ToLocalChecked(), value.ToLocalChecked());
+				}
+			}
+
+			Nan::Set(indices, Nan::New("groups").ToLocalChecked(), indexGroups);
+		}
 	}
 	else
 	{
 		Nan::Set(result, Nan::New("groups").ToLocalChecked(), Nan::Undefined());
+		if (re2->hasIndices) {
+			Nan::Set(indices, Nan::New("groups").ToLocalChecked(), Nan::Undefined());
+		}
 	}
 
-	if (re2->global || re2->sticky)
-	{
-		re2->lastIndex += str.isBuffer ? groups[0].data() - str.data + groups[0].size() - lastIndex : getUtf16Length(str.data + lastIndex, groups[0].data() + groups[0].size());
+	if (re2->hasIndices) {
+		Nan::Set(result, Nan::New("indices").ToLocalChecked(), indices);
 	}
 
 	info.GetReturnValue().Set(result);

package/lib/match.cc

@@ -61,8 +61,9 @@ NAN_METHOD(WrappedRE2::Match)
 			{
 				size_t s = getUtf8CharSize(a.data[lastIndex]);
 				lastIndex += s;
-				if (s == 4 && n >= 2) --n; // this utf8 character will take two utf16 characters
-				// the decrement above is protected to avoid an overflow of an unsigned integer
+				if (s == 4 && n >= 2)
+					--n; // this utf8 character will take two utf16 characters
+						 // the decrement above is protected to avoid an overflow of an unsigned integer
 			}
 			anchor = RE2::ANCHOR_START;
 		}
@@ -81,16 +82,34 @@ NAN_METHOD(WrappedRE2::Match)
 
 	// form a result
 
-	auto result = Nan::New<v8::Array>();
+	auto result = Nan::New<v8::Array>(), indices = Nan::New<v8::Array>();
 
 	if (a.isBuffer)
 	{
 		for (size_t i = 0, n = groups.size(); i < n; ++i)
 		{
 			const auto &item = groups[i];
-			if (item.data() != NULL)
+			const auto data = item.data();
+			if (data)
 			{
-				Nan::Set(result, i, Nan::CopyBuffer(item.data(), item.size()).ToLocalChecked());
+				Nan::Set(result, i, Nan::CopyBuffer(data, item.size()).ToLocalChecked());
+				if (!re2->global && re2->hasIndices)
+				{
+					auto pair = Nan::New<v8::Array>();
+					auto offset = getUtf16Length(a.data + lastIndex, data);
+					auto length = getUtf16Length(data, data + item.size());
+					Nan::Set(pair, 0, Nan::New<v8::Integer>(static_cast<int>(offset)));
+					Nan::Set(pair, 1, Nan::New<v8::Integer>(static_cast<int>(offset + length)));
+					Nan::Set(indices, i, pair);
+				}
+			}
+			else
+			{
+				Nan::Set(result, i, Nan::Undefined());
+				if (!re2->global && re2->hasIndices)
+				{
+					Nan::Set(indices, i, Nan::Undefined());
+				}
 			}
 		}
 		if (!re2->global)
@@ -104,9 +123,27 @@ NAN_METHOD(WrappedRE2::Match)
 		for (size_t i = 0, n = groups.size(); i < n; ++i)
 		{
 			const auto &item = groups[i];
-			if (item.data() != NULL)
+			const auto data = item.data();
+			if (data)
 			{
-				Nan::Set(result, i, Nan::New(item.data(), item.size()).ToLocalChecked());
+				Nan::Set(result, i, Nan::New(data, item.size()).ToLocalChecked());
+				if (!re2->global && re2->hasIndices)
+				{
+					auto pair = Nan::New<v8::Array>();
+					auto offset = getUtf16Length(a.data + lastIndex, data);
+					auto length = getUtf16Length(data, data + item.size());
+					Nan::Set(pair, 0, Nan::New<v8::Integer>(static_cast<int>(offset)));
+					Nan::Set(pair, 1, Nan::New<v8::Integer>(static_cast<int>(offset + length)));
+					Nan::Set(indices, i, pair);
+				}
+			}
+			else
+			{
+				Nan::Set(result, i, Nan::Undefined());
+				if (!re2->global && re2->hasIndices)
+				{
+					Nan::Set(indices, i, Nan::Undefined());
+				}
 			}
 		}
 		if (!re2->global)
@@ -143,12 +180,38 @@ NAN_METHOD(WrappedRE2::Match)
 			}
 
 			Nan::Set(result, Nan::New("groups").ToLocalChecked(), groups);
+
+			if (re2->hasIndices)
+			{
+				auto indexGroups = Nan::New<v8::Object>();
+				Nan::SetPrototype(indexGroups, Nan::Null());
+
+				for (auto group : groupNames)
+				{
+					auto value = Nan::Get(indices, group.first);
+					if (!value.IsEmpty())
+					{
+						Nan::Set(indexGroups, Nan::New(group.second).ToLocalChecked(), value.ToLocalChecked());
+					}
+				}
+
+				Nan::Set(indices, Nan::New("groups").ToLocalChecked(), indexGroups);
+			}
 		}
 		else
 		{
 			Nan::Set(result, Nan::New("groups").ToLocalChecked(), Nan::Undefined());
+			if (re2->hasIndices)
+			{
+				Nan::Set(indices, Nan::New("groups").ToLocalChecked(), Nan::Undefined());
+			}
 		}
 	}
 
+	if (re2->hasIndices)
+	{
+		Nan::Set(result, Nan::New("indices").ToLocalChecked(), indices);
+	}
+
 	info.GetReturnValue().Set(result);
 }

package/lib/new.cc

@@ -232,6 +232,7 @@ NAN_METHOD(WrappedRE2::New)
 	bool dotAll = false;
 	bool unicode = false;
 	bool sticky = false;
+	bool hasIndices = false;
 
 	auto context = Nan::GetCurrentContext();
 
@@ -272,6 +273,9 @@ NAN_METHOD(WrappedRE2::New)
 			case 'y':
 				sticky = true;
 				break;
+			case 'd':
+				hasIndices = true;
+				break;
 			}
 		}
 		size = 0;
@@ -306,6 +310,7 @@ NAN_METHOD(WrappedRE2::New)
 		dotAll = bool(flags & v8::RegExp::kDotAll);
 		unicode = bool(flags & v8::RegExp::kUnicode);
 		sticky = bool(flags & v8::RegExp::kSticky);
+		hasIndices = bool(flags & v8::RegExp::kHasIndices);
 	}
 	else if (info[0]->IsObject() && !info[0]->IsString())
 	{
@@ -332,6 +337,7 @@ NAN_METHOD(WrappedRE2::New)
 			dotAll = re2->dotAll;
 			unicode = true;
 			sticky = re2->sticky;
+			hasIndices = re2->hasIndices;
 		}
 	}
 	else if (info[0]->IsString())
@@ -385,7 +391,7 @@ NAN_METHOD(WrappedRE2::New)
 	options.set_dot_nl(dotAll);
 	options.set_log_errors(false); // inappropriate when embedding
 
-	std::unique_ptr<WrappedRE2> re2(new WrappedRE2(re2::StringPiece(data, size), options, source, global, ignoreCase, multiline, dotAll, sticky));
+	std::unique_ptr<WrappedRE2> re2(new WrappedRE2(re2::StringPiece(data, size), options, source, global, ignoreCase, multiline, dotAll, sticky, hasIndices));
 	if (!re2->regexp.ok())
 	{
 		return Nan::ThrowSyntaxError(re2->regexp.error().c_str());

package/lib/replace.cc

@@ -228,7 +228,8 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 				{
 					size_t s = getUtf8CharSize(data[lastIndex]);
 					lastIndex += s;
-					if (s == 4 && n >= 2) {
+					if (s == 4 && n >= 2)
+					{
 						--n; // this utf8 character will take two utf16 characters
 					}
 					// the decrement above is protected to avoid an overflow of an unsigned integer
@@ -299,7 +300,7 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 
 inline Nan::Maybe<std::string> replace(const Nan::Callback *replacer, const std::vector<re2::StringPiece> &groups, const re2::StringPiece &str, const v8::Local<v8::Value> &input, bool useBuffers, const std::map<std::string, int> &namedGroups)
 {
-	std::vector<v8::Local<v8::Value> > argv;
+	std::vector<v8::Local<v8::Value>> argv;
 
 	auto context = Nan::GetCurrentContext();
 
@@ -308,7 +309,15 @@ inline Nan::Maybe<std::string> replace(const Nan::Callback *replacer, const std:
 		for (size_t i = 0, n = groups.size(); i < n; ++i)
 		{
 			const auto &item = groups[i];
-			argv.push_back(Nan::CopyBuffer(item.data(), item.size()).ToLocalChecked());
+			const auto data = item.data();
+			if (data)
+			{
+				argv.push_back(Nan::CopyBuffer(data, item.size()).ToLocalChecked());
+			}
+			else
+			{
+				argv.push_back(Nan::Undefined());
+			}
 		}
 		argv.push_back(Nan::New(static_cast<int>(groups[0].data() - str.data())));
 	}
@@ -317,7 +326,15 @@ inline Nan::Maybe<std::string> replace(const Nan::Callback *replacer, const std:
 		for (size_t i = 0, n = groups.size(); i < n; ++i)
 		{
 			const auto &item = groups[i];
-			argv.push_back(Nan::New(item.data(), item.size()).ToLocalChecked());
+			const auto data = item.data();
+			if (data)
+			{
+				argv.push_back(Nan::New(data, item.size()).ToLocalChecked());
+			}
+			else
+			{
+				argv.push_back(Nan::Undefined());
+			}
 		}
 		argv.push_back(Nan::New(static_cast<int>(getUtf16Length(str.data(), groups[0].data()))));
 	}
@@ -381,7 +398,8 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 				{
 					size_t s = getUtf8CharSize(data[lastIndex]);
 					lastIndex += s;
-					if (s == 4 && n >= 2) {
+					if (s == 4 && n >= 2)
+					{
 						--n; // this utf8 character will take two utf16 characters
 					}
 					// the decrement above is protected to avoid an overflow of an unsigned integer

package/lib/wrapped_re2.h

@@ -11,8 +11,8 @@ class WrappedRE2 : public Nan::ObjectWrap
 {
 private:
 	WrappedRE2(const re2::StringPiece &pattern, const re2::RE2::Options &options, const std::string &src,
-			   const bool &g, const bool &i, const bool &m, const bool &s, const bool &y) : regexp(pattern, options),
-																							source(src), global(g), ignoreCase(i), multiline(m), dotAll(s), sticky(y), lastIndex(0) {}
+			   const bool &g, const bool &i, const bool &m, const bool &s, const bool &y, const bool &d) : regexp(pattern, options),
+																										   source(src), global(g), ignoreCase(i), multiline(m), dotAll(s), sticky(y), hasIndices(d), lastIndex(0) {}
 
 	static NAN_METHOD(New);
 	static NAN_METHOD(ToString);
@@ -25,6 +25,7 @@ private:
 	static NAN_GETTER(GetDotAll);
 	static NAN_GETTER(GetUnicode);
 	static NAN_GETTER(GetSticky);
+	static NAN_GETTER(GetHasIndices);
 	static NAN_GETTER(GetLastIndex);
 	static NAN_SETTER(SetLastIndex);
 	static NAN_GETTER(GetInternalSource);
@@ -70,6 +71,7 @@ public:
 	bool multiline;
 	bool dotAll;
 	bool sticky;
+	bool hasIndices;
 	size_t lastIndex;
 };
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "re2",
-  "version": "1.18.2",
+  "version": "1.19.0",
   "description": "Bindings for RE2: fast, safe alternative to backtracking regular expression engines.",
   "homepage": "https://github.com/uhop/node-re2",
   "bugs": "https://github.com/uhop/node-re2/issues",

package/tests/test_exec.js

@@ -21,9 +21,13 @@ unit.add(module, [
 
     var result = re.exec('The Quick Brown Fox Jumps Over The Lazy Dog');
 
-    eval(t.TEST("t.unify(result, ['Quick Brown Fox Jumps', 'Brown', 'Jumps'])"));
+    eval(
+      t.TEST("t.unify(result, ['Quick Brown Fox Jumps', 'Brown', 'Jumps'])")
+    );
     eval(t.TEST('result.index === 4'));
-    eval(t.TEST("result.input === 'The Quick Brown Fox Jumps Over The Lazy Dog'"));
+    eval(
+      t.TEST("result.input === 'The Quick Brown Fox Jumps Over The Lazy Dog'")
+    );
     eval(t.TEST('re.lastIndex === 25'));
   },
   function test_execSucc(t) {
@@ -148,12 +152,20 @@ unit.add(module, [
 
     var result = re.exec('Каждый Охотник Желает Знать Где Сидит Фазан');
 
-    eval(t.TEST("t.unify(result, ['Охотник Желает Знать Где', 'Желает', 'Где'])"));
+    eval(
+      t.TEST("t.unify(result, ['Охотник Желает Знать Где', 'Желает', 'Где'])")
+    );
     eval(t.TEST('result.index === 7'));
-    eval(t.TEST("result.input === 'Каждый Охотник Желает Знать Где Сидит Фазан'"));
+    eval(
+      t.TEST("result.input === 'Каждый Охотник Желает Знать Где Сидит Фазан'")
+    );
     eval(t.TEST('re.lastIndex === 31'));
 
-    eval(t.TEST("result.input.substr(result.index) === 'Охотник Желает Знать Где Сидит Фазан'"));
+    eval(
+      t.TEST(
+        "result.input.substr(result.index) === 'Охотник Желает Знать Где Сидит Фазан'"
+      )
+    );
     eval(t.TEST("result.input.substr(re.lastIndex) === ' Сидит Фазан'"));
   },
   function test_execUnicodeSubsequent(t) {
@@ -248,11 +260,21 @@ unit.add(module, [
 
     eval(t.TEST('result.index === 13'));
     eval(t.TEST('result.input instanceof Buffer'));
-    eval(t.TEST("result.input.toString() === 'Каждый Охотник Желает Знать Где Сидит Фазан'"));
+    eval(
+      t.TEST(
+        "result.input.toString() === 'Каждый Охотник Желает Знать Где Сидит Фазан'"
+      )
+    );
     eval(t.TEST('re.lastIndex === 58'));
 
-    eval(t.TEST("result.input.toString('utf8', result.index) === 'Охотник Желает Знать Где Сидит Фазан'"));
-    eval(t.TEST("result.input.toString('utf8', re.lastIndex) === ' Сидит Фазан'"));
+    eval(
+      t.TEST(
+        "result.input.toString('utf8', result.index) === 'Охотник Желает Знать Где Сидит Фазан'"
+      )
+    );
+    eval(
+      t.TEST("result.input.toString('utf8', re.lastIndex) === ' Сидит Фазан'")
+    );
   },
 
   // Sticky tests
@@ -317,7 +339,9 @@ xy2 (at start of line)
     eval(t.TEST("result[0] === 'xy'"));
     eval(t.TEST('result.index > 3'));
     eval(t.TEST('result.index < pattern.length - 4'));
-    eval(t.TEST('result[0] === pattern.substr(result.index, result[0].length)'));
+    eval(
+      t.TEST('result[0] === pattern.substr(result.index, result[0].length)')
+    );
   },
 
   // dotAll tests
@@ -332,5 +356,63 @@ xy2 (at start of line)
     eval(t.TEST("new RE2('a.c', 's').test('abc')"));
     eval(t.TEST("new RE2(/a.c/s).test('a c')"));
     eval(t.TEST("new RE2(/a.c/s).test('a\\nc')"));
+  },
+
+  // hasIndices tests
+
+  function test_execHasIndices(t) {
+    'use strict';
+
+    eval(t.TEST("!new RE2('1').hasIndices"));
+    eval(t.TEST('!new RE2(/1/).hasIndices'));
+
+    var re = new RE2('(aa)(?<b>b)?(?<c>ccc)', 'd');
+
+    eval(t.TEST('re.hasIndices'));
+
+    var result = re.exec('1aabccc2');
+
+    eval(t.TEST('result.length === 4'));
+    eval(t.TEST("result.input === '1aabccc2'"));
+    eval(t.TEST('result.index === 1'));
+    eval(t.TEST('Object.keys(result.groups).length === 2'));
+    eval(t.TEST("result.groups.b === 'b'"));
+    eval(t.TEST("result.groups.c === 'ccc'"));
+    eval(t.TEST("result[0] === 'aabccc'"));
+    eval(t.TEST("result[1] === 'aa'"));
+    eval(t.TEST("result[2] === 'b'"));
+    eval(t.TEST("result[3] === 'ccc'"));
+    eval(t.TEST('result.indices.length === 4'));
+    eval(t.TEST('t.unify(result.indices, [[1, 7], [1, 3], [3, 4], [4, 7]])'));
+    eval(t.TEST('Object.keys(result.indices.groups).length === 2'));
+    eval(t.TEST('t.unify(result.indices.groups.b, [3, 4])'));
+    eval(t.TEST('t.unify(result.indices.groups.c, [4, 7])'));
+
+    result = re.exec('1aaccc2');
+
+    eval(t.TEST('result.length === 4'));
+    eval(t.TEST("result.input === '1aaccc2'"));
+    eval(t.TEST('result.index === 1'));
+    eval(t.TEST('Object.keys(result.groups).length === 2'));
+    eval(t.TEST('result.groups.b === undefined'));
+    eval(t.TEST("result.groups.c === 'ccc'"));
+    eval(t.TEST("result[0] === 'aaccc'"));
+    eval(t.TEST("result[1] === 'aa'"));
+    eval(t.TEST('result[2] === undefined'));
+    eval(t.TEST("result[3] === 'ccc'"));
+    eval(t.TEST('result.indices.length === 4'));
+    eval(
+      t.TEST('t.unify(result.indices, [[1, 6], [1, 3], undefined, [3, 6]])')
+    );
+    eval(t.TEST('Object.keys(result.indices.groups).length === 2'));
+    eval(t.TEST('t.unify(result.indices.groups.b, undefined)'));
+    eval(t.TEST('t.unify(result.indices.groups.c, [3, 6])'));
+
+    try {
+      re = new RE2(new RegExp('1', 'd'));
+      eval(t.TEST('re.hasIndices'));
+    } catch (e) {
+      // squelch
+    }
   }
 ]);

package/tests/test_match.js

@@ -1,137 +1,161 @@
-"use strict";
-
-
-var unit = require("heya-unit");
-var RE2  = require("../re2");
+'use strict';
 
+var unit = require('heya-unit');
+var RE2 = require('../re2');
 
 // tests
 
 unit.add(module, [
+  // These tests are copied from MDN:
+  // https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/match
+
+  function test_match(t) {
+    'use strict';
+
+    var str = 'For more information, see Chapter 3.4.5.1';
+
+    var re = new RE2(/(chapter \d+(\.\d)*)/i);
+    var result = re.match(str);
+
+    eval(t.TEST('result.input === str'));
+    eval(t.TEST('result.index === 26'));
+    eval(t.TEST('result.length === 3'));
+    eval(t.TEST("result[0] === 'Chapter 3.4.5.1'"));
+    eval(t.TEST("result[1] === 'Chapter 3.4.5.1'"));
+    eval(t.TEST("result[2] === '.1'"));
+  },
+  function test_matchGlobal(t) {
+    'use strict';
 
-	// These tests are copied from MDN:
-	// https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/match
-
-	function test_match(t) {
-		"use strict";
+    var re = new RE2(/[A-E]/gi);
+    var result = re.match(
+      'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz'
+    );
 
-		var str = "For more information, see Chapter 3.4.5.1";
+    eval(
+      t.TEST(
+        "t.unify(result, ['A', 'B', 'C', 'D', 'E', 'a', 'b', 'c', 'd', 'e'])"
+      )
+    );
+  },
+  function test_matchFail(t) {
+    'use strict';
 
-		var re = new RE2(/(chapter \d+(\.\d)*)/i);
-		var result = re.match(str);
+    var re = new RE2('(a+)?(b+)?');
+    var result = re.match('aaabb');
 
-		eval(t.TEST("result.input === str"));
-		eval(t.TEST("result.index === 26"));
-		eval(t.TEST("result.length === 3"));
-		eval(t.TEST("result[0] === 'Chapter 3.4.5.1'"));
-		eval(t.TEST("result[1] === 'Chapter 3.4.5.1'"));
-		eval(t.TEST("result[2] === '.1'"));
-	},
-	function test_matchGlobal(t) {
-		"use strict";
+    eval(t.TEST("result[1] === 'aaa'"));
+    eval(t.TEST("result[2] === 'bb'"));
 
-		var re = new RE2(/[A-E]/gi);
-		var result = re.match("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz");
+    result = re.match('aaacbb');
 
-		eval(t.TEST("t.unify(result, ['A', 'B', 'C', 'D', 'E', 'a', 'b', 'c', 'd', 'e'])"));
-	},
-	function test_matchFail(t) {
-		"use strict";
+    eval(t.TEST("result[1] === 'aaa'"));
+    eval(t.TEST('result[2] === undefined'));
+  },
+  function test_matchInvalid(t) {
+    'use strict';
 
-		var re = new RE2("(a+)?(b+)?");
-		var result = re.match("aaabb");
+    var re = RE2('');
 
-		eval(t.TEST("result[1] === 'aaa'"));
-		eval(t.TEST("result[2] === 'bb'"));
+    try {
+      re.match({
+        toString() {
+          throw 'corner';
+        }
+      });
+      t.test(false); // shouldn't be here
+    } catch (e) {
+      eval(t.TEST("e === 'corner'"));
+    }
+  },
 
-		result = re.match("aaacbb");
+  // Unicode tests
 
-		eval(t.TEST("result[1] === 'aaa'"));
-		eval(t.TEST("result[2] === undefined"));
-	},
-	function test_matchInvalid(t) {
-		"use strict";
+  function test_matchUnicode(t) {
+    'use strict';
 
-		var re = RE2('');
+    var str = 'Это ГЛАВА 3.4.5.1';
 
-		try {
-			re.match({ toString() { throw "corner"; } });
-			t.test(false); // shouldn't be here
-		} catch(e) {
-			eval(t.TEST("e === 'corner'"));
-		}
-	},
+    var re = new RE2(/(глава \d+(\.\d)*)/i);
+    var result = re.match(str);
 
-	// Unicode tests
+    eval(t.TEST('result.input === str'));
+    eval(t.TEST('result.index === 4'));
+    eval(t.TEST('result.length === 3'));
+    eval(t.TEST("result[0] === 'ГЛАВА 3.4.5.1'"));
+    eval(t.TEST("result[1] === 'ГЛАВА 3.4.5.1'"));
+    eval(t.TEST("result[2] === '.1'"));
+  },
 
-	function test_matchUnicode(t) {
-		"use strict";
+  // Buffer tests
 
-		var str = "Это ГЛАВА 3.4.5.1";
+  function test_matchBuffer(t) {
+    'use strict';
 
-		var re = new RE2(/(глава \d+(\.\d)*)/i);
-		var result = re.match(str);
+    var buf = new Buffer('Это ГЛАВА 3.4.5.1');
 
-		eval(t.TEST("result.input === str"));
-		eval(t.TEST("result.index === 4"));
-		eval(t.TEST("result.length === 3"));
-		eval(t.TEST("result[0] === 'ГЛАВА 3.4.5.1'"));
-		eval(t.TEST("result[1] === 'ГЛАВА 3.4.5.1'"));
-		eval(t.TEST("result[2] === '.1'"));
-	},
+    var re = new RE2(/(глава \d+(\.\d)*)/i);
+    var result = re.match(buf);
 
-	// Buffer tests
+    eval(t.TEST('result.input instanceof Buffer'));
+    eval(t.TEST('result.length === 3'));
+    eval(t.TEST('result[0] instanceof Buffer'));
+    eval(t.TEST('result[1] instanceof Buffer'));
+    eval(t.TEST('result[2] instanceof Buffer'));
 
-	function test_matchBuffer(t) {
-		"use strict";
+    eval(t.TEST('result.input === buf'));
+    eval(t.TEST('result.index === 7'));
+    eval(
+      t.TEST("result.input.toString('utf8', result.index) === 'ГЛАВА 3.4.5.1'")
+    );
+    eval(t.TEST("result[0].toString() === 'ГЛАВА 3.4.5.1'"));
+    eval(t.TEST("result[1].toString() === 'ГЛАВА 3.4.5.1'"));
+    eval(t.TEST("result[2].toString() === '.1'"));
+  },
 
-		var buf = new Buffer("Это ГЛАВА 3.4.5.1");
+  // Sticky tests
 
-		var re = new RE2(/(глава \d+(\.\d)*)/i);
-		var result = re.match(buf);
+  function test_matchSticky(t) {
+    'use strict';
 
-		eval(t.TEST("result.input instanceof Buffer"));
-		eval(t.TEST("result.length === 3"));
-		eval(t.TEST("result[0] instanceof Buffer"));
-		eval(t.TEST("result[1] instanceof Buffer"));
-		eval(t.TEST("result[2] instanceof Buffer"));
+    var re = new RE2('\\s+', 'y');
 
-		eval(t.TEST("result.input === buf"));
-		eval(t.TEST("result.index === 7"));
-		eval(t.TEST("result.input.toString('utf8', result.index) === 'ГЛАВА 3.4.5.1'"));
-		eval(t.TEST("result[0].toString() === 'ГЛАВА 3.4.5.1'"));
-		eval(t.TEST("result[1].toString() === 'ГЛАВА 3.4.5.1'"));
-		eval(t.TEST("result[2].toString() === '.1'"));
-	},
+    eval(t.TEST("re.match('Hello world, how are you?') === null"));
 
-	// Sticky tests
+    re.lastIndex = 5;
 
-	function test_matchSticky(t) {
-		"use strict";
+    var result = re.match('Hello world, how are you?');
 
-		var re = new RE2("\\s+", "y");
+    eval(t.TEST("t.unify(result, [' '])"));
+    eval(t.TEST('result.index === 5'));
+    eval(t.TEST('re.lastIndex === 6'));
 
-		eval(t.TEST("re.match('Hello world, how are you?') === null"));
+    var re2 = new RE2('\\s+', 'gy');
 
-		re.lastIndex = 5;
+    eval(t.TEST("re2.match('Hello world, how are you?') === null"));
 
-		var result = re.match("Hello world, how are you?");
+    re2.lastIndex = 5;
 
-		eval(t.TEST("t.unify(result, [' '])"));
-		eval(t.TEST("result.index === 5"));
-		eval(t.TEST("re.lastIndex === 6"));
+    eval(t.TEST("re2.match('Hello world, how are you?') === null"));
 
-		var re2 = new RE2("\\s+", "gy");
+    var re3 = new RE2(/[A-E]/giy);
+    var result3 = re3.match(
+      'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz'
+    );
 
-		eval(t.TEST("re2.match('Hello world, how are you?') === null"));
+    eval(t.TEST("t.unify(result3, ['A', 'B', 'C', 'D', 'E'])"));
+  },
 
-		re2.lastIndex = 5;
+  // hasIndices tests
 
-		eval(t.TEST("re2.match('Hello world, how are you?') === null"));
+  function test_matchHasIndices(t) {
+    'use strict';
 
-		var re3 = new RE2(/[A-E]/giy);
-		var result3 = re3.match("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz");
+    var re = new RE2('(aa)(?<b>b)?(?<c>ccc)', 'd'),
+      str1 = '1aabccc2',
+      str2 = '1aaccc2';
 
-		eval(t.TEST("t.unify(result3, ['A', 'B', 'C', 'D', 'E'])"));
-	}
+    eval(t.TEST("t.unify(str1.match(re), re.exec(str1))"));
+    eval(t.TEST("t.unify(str2.match(re), re.exec(str2))"));
+  }
 ]);

package/tests/test_replace.js

@@ -13,7 +13,10 @@ unit.add(module, [
     'use strict';
 
     var re = new RE2(/apples/gi);
-    var result = re.replace('Apples are round, and apples are juicy.', 'oranges');
+    var result = re.replace(
+      'Apples are round, and apples are juicy.',
+      'oranges'
+    );
     eval(t.TEST("result === 'oranges are round, and oranges are juicy.'"));
 
     re = new RE2(/xmas/i);
@@ -199,15 +202,24 @@ unit.add(module, [
     'use strict';
 
     var re = new RE2(/яблоки/gi);
-    var result = re.replace(new Buffer('Яблоки красны, яблоки сочны.'), 'апельсины');
+    var result = re.replace(
+      new Buffer('Яблоки красны, яблоки сочны.'),
+      'апельсины'
+    );
     eval(t.TEST('result instanceof Buffer'));
     eval(t.TEST("result.toString() === 'апельсины красны, апельсины сочны.'"));
 
-    result = re.replace(new Buffer('Яблоки красны, яблоки сочны.'), new Buffer('апельсины'));
+    result = re.replace(
+      new Buffer('Яблоки красны, яблоки сочны.'),
+      new Buffer('апельсины')
+    );
     eval(t.TEST('result instanceof Buffer'));
     eval(t.TEST("result.toString() === 'апельсины красны, апельсины сочны.'"));
 
-    result = re.replace('Яблоки красны, яблоки сочны.', new Buffer('апельсины'));
+    result = re.replace(
+      'Яблоки красны, яблоки сочны.',
+      new Buffer('апельсины')
+    );
     eval(t.TEST("typeof result == 'string'"));
     eval(t.TEST("result === 'апельсины красны, апельсины сочны.'"));
   },
@@ -280,5 +292,47 @@ unit.add(module, [
 
     eval(t.TEST("re2.replace('ABCDEFABCDEF', '!') === '!!!!!FABCDEF'"));
     eval(t.TEST('re2.lastIndex === 0'));
+  },
+
+  // Non-matches
+
+  function test_replaceOneNonMatch(t) {
+    'use strict';
+
+    function replacer(match, capture, offset, string) {
+      t.test(typeof offset == 'number');
+      t.test(typeof match == 'string');
+      t.test(typeof string == 'string');
+      t.test(typeof capture == 'undefined');
+      t.test(offset === 0);
+      t.test(string === 'hello ');
+      return '';
+    }
+
+    var re = new RE2(/hello (world)?/);
+    re.replace('hello ', replacer);
+  },
+  function test_replaceTwoNonMatches(t) {
+    'use strict';
+
+    function replacer(match, capture1, capture2, offset, string, groups) {
+      t.test(typeof offset == 'number');
+      t.test(typeof match == 'string');
+      t.test(typeof string == 'string');
+      t.test(typeof capture1 == 'undefined');
+      t.test(typeof capture2 == 'undefined');
+      t.test(offset === 1);
+      t.test(match === 'b & y');
+      t.test(string === 'ab & yz');
+      t.test(typeof groups == 'object');
+      t.test(Object.keys(groups).length == 2);
+      t.test(groups.a === undefined);
+      t.test(groups.b == undefined);
+      return '';
+    }
+
+    var re = new RE2(/b(?<a>1)? & (?<b>2)?y/);
+    var result = re.replace('ab & yz', replacer);
+    eval(t.TEST("result === 'az'"));
   }
 ]);