指向新合同版本的国家仍然试图废除旧合同？

Question

我有一个场景，我已经将我的义务从V1升级到V2，并成功指向正确的V2合约。接下来，我试图对这些升级后的V2国家做一个< code>SettleObligation。当在< code > CollectSignatureFlow 中形成并发送事务时，在我的< code>isGreaterThan方法中发现了一个< code > Java . lang . nosuchmethoderror ，该方法在契约中的< code>verifySettle命令中调用。

这个特定的函数最初存在于一个文件BaseHelper.kt中的包“com.example.base”上，一路上，在V2中，我们将函数迁移到另一个文件MathHelper.kt，这一移动不会导致任何未解析的引用，因为包是相同的。

想了解：

1. 合约是否通过文件名而不是包名来引用函数位置？并且编译合约 v1.jar后它是不可更改的？
2. 为什么 V2 状态仍试图对 V1 合约进行反序列化？这是通过以某种方式走链来完成的吗？

下面的堆栈跟踪

[WARN ] 2018-11-13T00:05:12,777Z [Node thread-1] flow.[cd538d42-1715-4ed3-bde6-38eca94ef79f].run - Flow ended due to receiving exception {}
net.corda.core.contracts.TransactionVerificationException$ContractRejection: Contract verification failed: com.example.base.BaseHelperKt.isGreaterThan(Ljava/math/BigDecimal;Ljava/math/BigDecimal;)Z, 
contract: com.example.contracts.ObligationContractV1, transaction: 8B8780A16D330A93A361F747B77C227442BD310C9DAAA561376DED69F580C794
        at net.corda.node.services.statemachine.FlowStateMachineImpl.erroredEnd(FlowStateMachineImpl.kt:497) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowStateMachineImpl.confirmNoError(FlowStateMachineImpl.kt:481) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowStateMachineImpl.waitForMessage(FlowStateMachineImpl.kt:444) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowStateMachineImpl.receiveInternal(FlowStateMachineImpl.kt:376) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowStateMachineImpl.receive(FlowStateMachineImpl.kt:229) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowSessionImpl.receive(FlowSessionImpl.kt:44) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowSessionImpl.receive(FlowSessionImpl.kt:48) ~[corda-node-3.2.1847-corda.jar:?]
        at net.corda.core.flows.CollectSignatureFlow.call(CollectSignaturesFlow.kt:290) ~[corda-core-3.2.1847-corda.jar:?]
        at net.corda.core.flows.CollectSignatureFlow.call(CollectSignaturesFlow.kt:135) ~[corda-core-3.2.1847-corda.jar:?]
        at net.corda.core.flows.FlowLogic.subFlow(FlowLogic.kt:290) ~[corda-core-3.2.1847-corda.jar:?]
        at net.corda.core.flows.CollectSignaturesFlow.call(CollectSignaturesFlow.kt:114) ~[corda-core-3.2.1847-corda.jar:?]
        at net.corda.core.flows.CollectSignaturesFlow.call(CollectSignaturesFlow.kt:64) ~[corda-core-3.2.1847-corda.jar:?]
        at net.corda.core.flows.FlowLogic.subFlow(FlowLogic.kt:290) ~[corda-core-3.2.1847-corda.jar:?]
        at com.example.flows.flows.SettleObligation$Initiator.collectSignature(SettleObligation.kt:178) ~[obligation-1.0.jar:?]
        at com.example.flows.flows.SettleObligation$Initiator.call(SettleObligation.kt:87) ~[obligation-1.0.jar:?]
        at com.example.flows.flows.SettleObligation$Initiator.call(SettleObligation.kt:51) ~[obligation-1.0.jar:?]
        at net.corda.node.services.statemachine.FlowStateMachineImpl.run(FlowStateMachineImpl.kt:96) [corda-node-3.2.1847-corda.jar:?]
        at net.corda.node.services.statemachine.FlowStateMachineImpl.run(FlowStateMachineImpl.kt:44) [corda-node-3.2.1847-corda.jar:?]
        at co.paralleluniverse.fibers.Fiber.run1(Fiber.java:1092) [quasar-core-0.7.9-jdk8.jar:0.7.9]
        at co.paralleluniverse.fibers.Fiber.exec(Fiber.java:788) [quasar-core-0.7.9-jdk8.jar:0.7.9]
        at co.paralleluniverse.fibers.RunnableFiberTask.doExec(RunnableFiberTask.java:100) [quasar-core-0.7.9-jdk8.jar:0.7.9]
        at co.paralleluniverse.fibers.RunnableFiberTask.run(RunnableFiberTask.java:91) [quasar-core-0.7.9-jdk8.jar:0.7.9]
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [?:1.8.0_181]
        at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:1.8.0_181]
        at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180) [?:1.8.0_181]
        at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293) [?:1.8.0_181]
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) [?:1.8.0_181]
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) [?:1.8.0_181]
        at net.corda.node.utilities.AffinityExecutor$ServiceAffinityExecutor$1$thread$1.run(AffinityExecutor.kt:62) [corda-node-3.2.1847-corda.jar:?]
Caused by: net.corda.core.CordaRuntimeException: java.lang.NoSuchMethodError: com.example.base.BaseHelperKt.isGreaterThan(Ljava/math/BigDecimal;Ljava/math/BigDecimal;)Z
[INFO ] 2018-11-13T00:05:12,807Z [RxIoScheduler-2] network.PersistentNetworkMapCache.addNode - Previous node was identical to incoming one - doing nothing {}
[INFO ] 2018-11-13T00:05:12,807Z [RxIoScheduler-2] network.PersistentNetworkMapCache.addNode - Done adding node with info: NodeInfo

Answer 1

在柯特林中，每个文件都被编译为一个 JVM 级类。名为福巴.kt 的文件将成为名为福巴的类。

因此，当您将一个顶级函数从一个文件移动到另一个文件时，该文件的名称将被嵌入到编译后的代码中，这样的更改不是二进制兼容的，即使它是源代码兼容的。

不幸的是，这是所有软件开发隐藏的复杂性之一——源代码编译器解析符号范围的方式并不总是与运行时链接器相同。它在每种语言和运行时都会发生，尽管JetBrains的人有办法在很长时间内修复它，如果他们选择这样做的话。抱歉。您必须将函数移回来，或者提供一个别名(查看< code>@JvmName注释)。

Answer 2

resolveTransactionFlow.kt（沿着链走时）将按顺序循环以获取Tx，同时验证它。在这种情况下，需要所有旧的合同代码（作为cordapp）都存在，否则您将无法解析链的来源。

val result = topologicalSort(newTxns)
    result.forEach {
        // For each transaction, verify it and insert it into the database. As we are iterating over them in a
        // depth-first order, we should not encounter any verification failures due to missing data. If we fail
        // half way through, it's no big deal, although it might result in us attempting to re-download data
        // redundantly next time we attempt verification.
        it.verify(serviceHub)
        serviceHub.recordTransactions(StatesToRecord.NONE, listOf(it))
    }