JDK中Lambda表達(dá)式的序列化與SerializedLambda的巧妙使用

前提

筆者在下班空余時(shí)間想以Javassist為核心基于JDBC寫一套摒棄反射調(diào)用的輕量級(jí)的ORM框架，過(guò)程中有研讀mybatis、tk-mapper、mybatis-plus和spring-boot-starter-jdbc的源代碼，其中發(fā)現(xiàn)了mybatis-plus中的LambdaQueryWrapper可以獲取當(dāng)前調(diào)用的Lambda表達(dá)式中的方法信息（實(shí)際上是CallSite的信息），這里做一個(gè)完整的記錄。本文基于JDK11編寫，其他版本的JDK不一定合適。

神奇的Lambda表達(dá)式序列化

之前在看Lambda表達(dá)式源碼實(shí)現(xiàn)的時(shí)候沒(méi)有細(xì)看LambdaMetafactory的注釋，這個(gè)類頂部大量注釋中其中有一段如下：

簡(jiǎn)單翻譯一下就是：可序列化特性。一般情況下，生成的函數(shù)對(duì)象（這里應(yīng)該是特指基于Lambda表達(dá)式實(shí)現(xiàn)的特殊函數(shù)對(duì)象）不需要支持序列化特性。如果需要支持該特性，FLAG_SERIALIZABLE（LambdaMetafactory的一個(gè)靜態(tài)整型屬性，值為1 << 0）可以用來(lái)表示函數(shù)對(duì)象是序列化的。一旦使用了支持序列化特性的函數(shù)對(duì)象，那么它們以SerializedLambda類的形式序列化，這些SerializedLambda實(shí)例需要額外的"捕獲類"的協(xié)助（捕獲類，如MethodHandles.Lookup的caller參數(shù)所描述），詳細(xì)信息參閱SerializedLambda。

在LambdaMetafactory的注釋中再搜索一下FLAG_SERIALIZABLE，可以看到這段注釋：

大意為：設(shè)置了FLAG_SERIALIZABLE標(biāo)記后生成的函數(shù)對(duì)象實(shí)例會(huì)實(shí)現(xiàn)Serializable接口，并且會(huì)存在一個(gè)名字為writeReplace的方法，該方法的返回值類型為SerializedLambda。調(diào)用這些函數(shù)對(duì)象的方法（前面提到的"捕獲類"）的調(diào)用者必須存在一個(gè)名字為$deserializeLambda$的方法，如SerializedLambda類所描述。

最后看SerializedLambda的描述，注釋有四大段，這里貼出并且每小段提取核心信息：

各個(gè)段落大意如下：

• 段落一：SerializedLambda是Lambda表達(dá)式的序列化形式，這類存儲(chǔ)了Lambda表達(dá)式的運(yùn)行時(shí)信息
• 段落二：為了確保Lambda表達(dá)式的序列化實(shí)現(xiàn)正確性，編譯器或者語(yǔ)言類庫(kù)可以選用的一種方式是確保writeReplace方法返回一個(gè)SerializedLambda實(shí)例
• 段落三：SerializedLambda提供一個(gè)readResolve方法，其職能類似于調(diào)用"捕獲類"中靜態(tài)方法$deserializeLambda$(SerializedLambda)并且把自身實(shí)例作為入?yún)?，該過(guò)程理解為反序列化過(guò)程
• 段落四：序列化和反序列化產(chǎn)生的函數(shù)對(duì)象的身份敏感操作的標(biāo)識(shí)形式（如System.identityHashCode()、對(duì)象鎖定等等）是不可預(yù)測(cè)的

最終的結(jié)論就是：如果一個(gè)函數(shù)式接口實(shí)現(xiàn)了Serializable接口，那么它的實(shí)例就會(huì)自動(dòng)生成了一個(gè)返回SerializedLambda實(shí)例的writeReplace方法，可以從SerializedLambda實(shí)例中獲取到這個(gè)函數(shù)式接口的運(yùn)行時(shí)信息。這些運(yùn)行時(shí)信息就是SerializedLambda的屬性：

舉個(gè)實(shí)際的例子，定義一個(gè)實(shí)現(xiàn)了Serializable的函數(shù)式接口并且調(diào)用它：

public?class?App?{

????@FunctionalInterface
????public?interface?CustomerFunction<S,?T>?extends?Serializable?{

????????T?convert(S?source);
????}

????public?static?void?main(String[]?args)?throws?Exception?{
????????CustomerFunction?function?=?Long::parseLong;
????????Long?result?=?function.convert("123");
????????System.out.println(result);
????????Method?method?=?function.getClass().getDeclaredMethod("writeReplace");
????????method.setAccessible(true);
????????SerializedLambda?serializedLambda?=?(SerializedLambda)method.invoke(function);
????????System.out.println(serializedLambda.getCapturingClass());
????}
}

執(zhí)行的DEBUG信息如下：

這樣就能獲取到函數(shù)式接口實(shí)例在調(diào)用方法時(shí)候的調(diào)用點(diǎn)運(yùn)行時(shí)信息，甚至連泛型參數(shù)擦除前的類型都能拿到，那么就可以衍生出很多技巧。例如：

public?class?ConditionApp?{

????@FunctionalInterface
????public?interface?CustomerFunction<S,?T>?extends?Serializable?{

????????T?convert(S?source);
????}

????@Data
????public?static?class?User?{

????????private?String?name;
????????private?String?site;
????}

????public?static?void?main(String[]?args)?throws?Exception?{
????????Condition?c1?=?addCondition(User::getName,?"=",?"throwable");
????????System.out.println("c1?=?"?+?c1);
????????Condition?c2?=?addCondition(User::getSite,?"IN",?"('throwx.cn','vlts.cn')");
????????System.out.println("c1?=?"?+?c2);
????}

????private?static??Condition?addCondition(CustomerFunction?function,
??????????????????????????????????????????????String?operation,
??????????????????????????????????????????????Object?value)?throws?Exception?{
????????Condition?condition?=?new?Condition();
????????Method?method?=?function.getClass().getDeclaredMethod("writeReplace");
????????method.setAccessible(true);
????????SerializedLambda?serializedLambda?=?(SerializedLambda)?method.invoke(function);
????????String?implMethodName?=?serializedLambda.getImplMethodName();
????????int?idx;
????????if?((idx?=?implMethodName.lastIndexOf("get"))?>=?0)?{
????????????condition.setField(Character.toLowerCase(implMethodName.charAt(idx?+?3))?+?implMethodName.substring(idx?+?4));
????????}
????????condition.setEntityKlass(Class.forName(serializedLambda.getImplClass().replace("/",?".")));
????????condition.setOperation(operation);
????????condition.setValue(value);
????????return?condition;
????}

????@Data
????private?static?class?Condition?{

????????private?Class?entityKlass;
????????private?String?field;
????????private?String?operation;
????????private?Object?value;
????}
}

//?執(zhí)行結(jié)果
c1?=?ConditionApp.Condition(entityKlass=class?club.throwable.lambda.ConditionApp$User,?field=name,?operation==,?value=throwable)
c1?=?ConditionApp.Condition(entityKlass=class?club.throwable.lambda.ConditionApp$User,?field=site,?operation=IN,?value=('throwx.cn','vlts.cn'))

?

很多人會(huì)擔(dān)心反射調(diào)用的性能，其實(shí)在高版本的JDK，反射性能已經(jīng)大幅度優(yōu)化，十分逼近直接調(diào)用的性能，更何況有些場(chǎng)景是少量反射調(diào)用場(chǎng)景，可以放心使用。

?

前面花大量篇幅展示了SerializedLambda的功能和使用，接著看Lambda表達(dá)式的序列化與反序列化：

public?class?SerializedLambdaApp?{

????@FunctionalInterface
????public?interface?CustomRunnable?extends?Serializable?{

????????void?run();
????}

????public?static?void?main(String[]?args)?throws?Exception?{
????????invoke(()?->?{
????????});
????}

????private?static?void?invoke(CustomRunnable?customRunnable)?throws?Exception?{
????????ByteArrayOutputStream?baos?=?new?ByteArrayOutputStream();
????????ObjectOutputStream?oos?=?new?ObjectOutputStream(baos);
????????oos.writeObject(customRunnable);
????????oos.close();
????????ObjectInputStream?ois?=?new?ObjectInputStream(new?ByteArrayInputStream(baos.toByteArray()));
????????Object?target?=?ois.readObject();
????????System.out.println(target);
????}
}

結(jié)果如下圖：

Lambda表達(dá)式序列化原理

關(guān)于Lambda表達(dá)式序列化的原理，可以直接參考ObjectStreamClass、ObjectOutputStream和ObjectInputStream的源碼，這里直接說(shuō)結(jié)論：

• 前提條件：待序列化對(duì)象需要實(shí)現(xiàn)Serializable接口
• 待序列化對(duì)象中如果存在writeReplace方法，則直接基于傳入的實(shí)例反射調(diào)用此方法得到的返回值類型作為序列化的目標(biāo)類型，對(duì)于Lambda表達(dá)式就是SerializedLambda類型
• 反序列化的過(guò)程剛好是逆轉(zhuǎn)的過(guò)程，調(diào)用的方法為readResolve，剛好前面提到SerializedLambda也存在同名的私有方法
• Lambda表達(dá)式的實(shí)現(xiàn)類型是VM生成的模板類，從結(jié)果上觀察，序列化前的實(shí)例和反序列化后得到的實(shí)例屬于不同的模板類，對(duì)于前一小節(jié)的例子某次運(yùn)行的結(jié)果中序列化前的模板類為club.throwable.lambda.SerializedLambdaApp$$Lambda$14/0x0000000800065840，反序列化后的模板類為club.throwable.lambda.SerializedLambdaApp$$Lambda$26/0x00000008000a4040

?

ObjectStreamClass是序列化和反序列化實(shí)現(xiàn)的類描述符，關(guān)于對(duì)象序列化和反序列化的類描述信息可以從這個(gè)類里面的成員屬性找到，例如這里提到的writeReplace和readResolve方法

?

圖形化的過(guò)程如下：

獲取SerializedLambda的方式

通過(guò)前面的分析，得知有兩種方式可以獲取Lambda表達(dá)式的SerializedLambda實(shí)例：

• 方式一：基于Lambda表達(dá)式實(shí)例和Lambda表達(dá)式的模板類反射調(diào)用writeReplace方法，得到的返回值就是SerializedLambda實(shí)例
• 方式二：基于序列化和反序列化的方式獲取SerializedLambda實(shí)例

基于這兩種方式可以分別編寫例子，例如反射方式如下：

//?反射方式
public?class?ReflectionSolution?{

????@FunctionalInterface
????public?interface?CustomerFunction<S,?T>?extends?Serializable?{

????????T?convert(S?source);
????}

????public?static?void?main(String[]?args)?throws?Exception?{
????????CustomerFunction?function?=?Long::parseLong;
????????SerializedLambda?serializedLambda?=?getSerializedLambda(function);
????????System.out.println(serializedLambda.getCapturingClass());
????}

????public?static?SerializedLambda?getSerializedLambda(Serializable?serializable)?throws?Exception?{
????????Method?writeReplaceMethod?=?serializable.getClass().getDeclaredMethod("writeReplace");
????????writeReplaceMethod.setAccessible(true);
????????return?(SerializedLambda)?writeReplaceMethod.invoke(serializable);
????}
}

序列化和反序列方式會(huì)稍微復(fù)雜，因?yàn)?code style="overflow-wrap: break-word;padding: 2px 4px;border-radius: 4px;margin-right: 2px;margin-left: 2px;background-color: rgba(27, 31, 35, 0.05);font-family: "Operator Mono", Consolas, Monaco, Menlo, monospace;word-break: break-all;">ObjectInputStream.readObject()方法會(huì)最終回調(diào)SerializedLambda.readResolve()方法，導(dǎo)致返回的結(jié)果是一個(gè)新模板類承載的Lambda表達(dá)式實(shí)例，所以這里需要想辦法中斷這個(gè)調(diào)用提前返回結(jié)果，方案是構(gòu)造一個(gè)和SerializedLambda相似但是不存在readResolve()方法的「影子類型」：

package?cn.vlts;
import?java.io.Serializable;

/**
?*?這里注意一定要和java.lang.invoke.SerializedLambda同名,可以不同包名,這是為了"欺騙"ObjectStreamClass中有個(gè)神奇的類名稱判斷classNamesEqual()方法
?*/
@SuppressWarnings("ALL")
public?class?SerializedLambda?implements?Serializable?{
????private?static?final?long?serialVersionUID?=?8025925345765570181L;
????private??Class?capturingClass;
????private??String?functionalInterfaceClass;
????private??String?functionalInterfaceMethodName;
????private??String?functionalInterfaceMethodSignature;
????private??String?implClass;
????private??String?implMethodName;
????private??String?implMethodSignature;
????private??int?implMethodKind;
????private??String?instantiatedMethodType;
????private??Object[]?capturedArgs;

????public?String?getCapturingClass()?{
????????return?capturingClass.getName().replace('.',?'/');
????}
????public?String?getFunctionalInterfaceClass()?{
????????return?functionalInterfaceClass;
????}
????public?String?getFunctionalInterfaceMethodName()?{
????????return?functionalInterfaceMethodName;
????}
????public?String?getFunctionalInterfaceMethodSignature()?{
????????return?functionalInterfaceMethodSignature;
????}
????public?String?getImplClass()?{
????????return?implClass;
????}
????public?String?getImplMethodName()?{
????????return?implMethodName;
????}
????public?String?getImplMethodSignature()?{
????????return?implMethodSignature;
????}
????public?int?getImplMethodKind()?{
????????return?implMethodKind;
????}
????public?final?String?getInstantiatedMethodType()?{
????????return?instantiatedMethodType;
????}
????public?int?getCapturedArgCount()?{
????????return?capturedArgs.length;
????}
????public?Object?getCapturedArg(int?i)?{
????????return?capturedArgs[i];
????}
}


public?class?SerializationSolution?{

????@FunctionalInterface
????public?interface?CustomerFunction<S,?T>?extends?Serializable?{

????????T?convert(S?source);
????}

????public?static?void?main(String[]?args)?throws?Exception?{
????????CustomerFunction?function?=?Long::parseLong;
????????cn.vlts.SerializedLambda?serializedLambda?=?getSerializedLambda(function);
????????System.out.println(serializedLambda.getCapturingClass());
????}

????private?static?cn.vlts.SerializedLambda?getSerializedLambda(Serializable?serializable)?throws?Exception?{
????????try?(ByteArrayOutputStream?baos?=?new?ByteArrayOutputStream();
?????????????ObjectOutputStream?oos?=?new?ObjectOutputStream(baos))?{
????????????oos.writeObject(serializable);
????????????oos.flush();
????????????try?(ObjectInputStream?ois?=?new?ObjectInputStream(new?ByteArrayInputStream(baos.toByteArray()))?{
????????????????@Override
????????????????protected?Class?resolveClass(ObjectStreamClass?desc)?throws?IOException,?ClassNotFoundException?{
????????????????????Class?klass?=?super.resolveClass(desc);
????????????????????return?klass?==?java.lang.invoke.SerializedLambda.class???cn.vlts.SerializedLambda.class?:?klass;
????????????????}
????????????})?{
????????????????return?(cn.vlts.SerializedLambda)?ois.readObject();
????????????}
????????}
????}
}

被遺忘的$deserializeLambda$方法

前文提到，Lambda表達(dá)式實(shí)例反序列化的時(shí)候會(huì)調(diào)用java.lang.invoke.SerializedLambda.readResolve()方法，神奇的是，此方法源碼如下：

private?Object?readResolve()?throws?ReflectiveOperationException?{
????try?{
????????Method?deserialize?=?AccessController.doPrivileged(new?PrivilegedExceptionAction<>()?{
????????????@Override
????????????public?Method?run()?throws?Exception?{
????????????????Method?m?=?capturingClass.getDeclaredMethod("$deserializeLambda$",?SerializedLambda.class);
????????????????m.setAccessible(true);
????????????????return?m;
????????????}
????????});

????????return?deserialize.invoke(null,?this);
????}
????catch?(PrivilegedActionException?e)?{
????????Exception?cause?=?e.getException();
????????if?(cause?instanceof?ReflectiveOperationException)
????????????throw?(ReflectiveOperationException)?cause;
????????else?if?(cause?instanceof?RuntimeException)
????????????throw?(RuntimeException)?cause;
????????else
????????????throw?new?RuntimeException("Exception?in?SerializedLambda.readResolve",?e);
????}
}

看起來(lái)就是"捕獲類"中存在一個(gè)這樣的靜態(tài)方法：

class?CapturingClass?{

????private?static?Object?$deserializeLambda$(SerializedLambda?serializedLambda){
????????return?[serializedLambda]?=>?Lambda表達(dá)式實(shí)例;
????}??
}

可以嘗試檢索"捕獲類"中的方法列表：

public?class?CapturingClassApp?{

????@FunctionalInterface
????public?interface?CustomRunnable?extends?Serializable?{

????????void?run();
????}

????public?static?void?main(String[]?args)?throws?Exception?{
????????invoke(()?->?{
????????});
????}

????private?static?void?invoke(CustomRunnable?customRunnable)?throws?Exception?{
????????Method?writeReplaceMethod?=?customRunnable.getClass().getDeclaredMethod("writeReplace");
????????writeReplaceMethod.setAccessible(true);
????????java.lang.invoke.SerializedLambda?serializedLambda?=?(java.lang.invoke.SerializedLambda)
????????????????writeReplaceMethod.invoke(customRunnable);
????????Class?capturingClass?=?Class.forName(serializedLambda.getCapturingClass().replace("/",?"."));
????????ReflectionUtils.doWithMethods(capturingClass,?method?->?{
????????????????????System.out.printf("方法名:%s,修飾符:%s,方法參數(shù)列表:%s,方法返回值類型:%s\n",?method.getName(),
????????????????????????????Modifier.toString(method.getModifiers()),
????????????????????????????Arrays.toString(method.getParameterTypes()),
????????????????????????????method.getReturnType().getName());
????????????????},
????????????????method?->?Objects.equals(method.getName(),?"$deserializeLambda$"));
????}
}

//?執(zhí)行結(jié)果
方法名:$deserializeLambda$,修飾符:private?static,方法參數(shù)列表:[class?java.lang.invoke.SerializedLambda],方法返回值類型:java.lang.Object

果真是存在一個(gè)和之前提到的java.lang.invoke.SerializedLambda注釋描述一致的"捕獲類"的SerializedLambda實(shí)例轉(zhuǎn)化為Lambda表達(dá)式實(shí)例的方法，因?yàn)樗阉鞫嗵幍胤蕉紱](méi)發(fā)現(xiàn)此方法的蹤跡，猜測(cè)$deserializeLambda$是方法由VM生成，并且只能通過(guò)反射的方法調(diào)用，算是一個(gè)隱藏得比較深的技巧。

小結(jié)

JDK中的Lambda表達(dá)式功能已經(jīng)發(fā)布很多年了，想不到這么多年后的今天才弄清楚其序列化和反序列化方式，雖然這不是一個(gè)復(fù)雜的問(wèn)題，但算是最近一段時(shí)間看到的比較有意思的一個(gè)知識(shí)點(diǎn)。

參考資料：

• JDK11源碼
• Mybatis-Plus相關(guān)源碼

（本文完 e-a-20211127 c-2-d）