This requires Kotlin 1.2 versions of core and serialization (core-1.2 and serialization-1.2 respectively), which are just "shell" modules and which compile the existing source code with Kotlin 1.2. The 1.2 plugin does not work with the current version of Gradle and so the 1.2 compiler has to be called directly.
Now with two versions of Kotlin in the code base, each module needs to have its version manually specified to ensure a clean separation. Otherwise, the default Kotlin version can override 1.2 when needed.
Some of the code was tidied-up or improved to enable it to be cross-compiled. For post-1.2 APIs being used, they have been copied into core-1.2 with the same method signatures. OpenTelemetryComponent was moved to node-api, along with the dependency, to avoid also having a 1.2 version for the opentelemetry module.
Major changes due to JDK 17:
1. JDK17 JCE Provider now has built-in support for eddsas, corda uses
the bouncycastle (i2p) implementation. This PR removes the conflicting
algorithms from the built-in JCE provider.
2. JavaScript scripting has been removed from the JDK, the corda log4j config was using
scripting to conditionally output additional diagnostic info if the MDC
was populated. This PR has removed the scripting.
3. The artifactory plug-ins used are now deprecated, this PR has removed them
and uses the same code as Corda 5 for publishing to artifactory.
4. Javadoc generation has been modified to use the latest dokka plug-ins.
5. Gradle 7.6 has implemented an incredibly annoying change where transitive
dependencies are not put on the compile classpath, so that they have to be
explicitly added as dependencies to projects.
6. Mockito has been updated, which sadly meant that quite a few source files
have to changes to use the new (org.mockito.kotlin) package name. This makes
this PR appear much larger than it is.
7. A number of tests have been marked as ignored to get a green, broadly they fall
into 3 classes.
The first is related to crypto keypair tests, it appears some logic
in the JDK prefers to use the SunJCE implementation and we prefer to use
bouncycastle. I believe this issue can be fixed with better test setup.
The second group is related to our use of a method called "uncheckedCast(..)",
the purpose of this method was to get rid of the annoying unchecked cast compiler
warning that would otherwise exist. It looks like the Kotlin 1.9 compiler type
inference differs and at runtime sometimes the type it infers is "Void" which causes
an exception at runtime. The simplest solution is to use an explicit cast instead of
unchecked cast, Corda 5 have removed unchecked cast from their codebase.
The third class are a number of ActiveMQ tests which appear to have a memory leak somewhere.
* [NOTICK] Add a custom detekt rule for tests with no timeout, and fix remaining missing timeouts
* [NOTICK] Add a test for custom detekt rules and tidying
* add timeout annotation to new test
Co-authored-by: Stefano Franz <roastario@gmail.com>
Adds support for understanding both Maps and Arrays
Irritatingly, whilst arrays are mostly serialized as lists, we cannot simply use a restricted List reader to deserialize them because there are subtle differences in the way we need to work out if its actually a restricted type or not. Rather than add a bunch of random logic into the factory method I've isolated it in the class hierarchy. So a little bit more code makes the implementations a lot neater. We also need to deal with the fact arras of unboxed primitives exist, which whilst Java really gets excited about, we don't need to care about. An int, is an int, is an int!.
Map support required we add a slightly better Value dumper, essentially the "key" component of the KV pair needs to be more flexible than a simple string when we're dumping out param:value pairs.
Testing
Added a lot more unit tests to both the ordered type notation code to ensure we build up the schema dependency struct in the correct order. Quite important as we rely on that in the composite factory to be strictly ordered to ensure we're not constructing a reader for a type we don't yet understand... and there were some small bugs in the version that predates this PR
Also added a lot higher level tests to ensure actual reading out of the blob works
* migrate PersistentIdentityService to use key.toShortString()
update definition of PublicKeyToExternalId mapping to allow fast lookup by externalId/publicKey
* fix misspelled table name
* add test of migration script
* add design document for proposal to move IdentityService to using the correct PK.toStringShort() method for hashing a publickey
* add enterprise testing considerations to design
* address review comments
* fix compilation errors
* modify PublicKeyToOwningIdentityCache to use toStringShort() as it's lookup key
* address syzmon's code review comments
The Blob Writer is a small kotlin app that allows arbitrary things to be serialized and then those bytes written to a file, quite useful for working on non JVM parsing of such things.
Along a similar vein, add a schema dumper alongside the blob-inspector to highlight the contents of the header
As having them all live in a single PropertyReader header / source
file was getting far too big. Additionally, sue some of our aliased
types to shorted lines (sVec, sList, uPtr etc)
Now things are working for the most part, move the readers into their
own internal namespace. Also start creating an actual separation of the
interface and the implementation with some Interface classes at the
non internal level.
Currently, it's more of a blob inspector than a full-blown
implementation, but this code does have the ability to understand a Corda
serialized blob and build a set of deserialisers to handle a blob. It
does this in the same way the native java implementation works by
interrogating the types (as per the definition in the envelope) and
building recursive type readers that are able to pull the relevant
information out of the byte stream.
Lots of future work to undertake, but as a starting point its a little
bit useful with the scope it could be made much more useful with some
work.
* Start pooling classpath scanning
Quickly patch synchronisation of attachment class loader cache. Needs a revisit but more complicated due to DJVM.
Annotate away for DJVM
Take ClassGraph utils into their own file so we can exclude for DJVM
Clean up a little
* Daemonize the threads
* Seems to be some concurrency problems with use of ClassGraph. Using a mutex for now to work around.
Move Raft and BFT-Smart notaries back into node to preserve backwards compatibility.
* Allow overriding full node config when using internal mock network parameters.
* Make BFT-Smart notary start up in prod mode as well
* Move raft & bftsmart notaries to net.corda.notary.experimental package
* Make sure Raft notary handles reference state edge cases correctly.
* Make sure BFT-Smart notary handles reference state edge cases correctly.
* Include notary schemas in node internal schemas
* Undo Raft notary table schema changes to maintain compatibility.
* Split Workflow and contracts of Finance App into separate Cordapps, part 1 - content which is different between OS and ENT is still in contract Cordapp.
* Move CashSelection implementations to workflow module.
* Move CashSelection implmentations to workflow module.
* Move finance module to finance-flows, top level finance module is empty.
* Move finance module to finance-flows, top level finance module is empty.
* Updated build comment.
* Revert publication of combined (contracts and flows) corda-finance.jar (to maintain backwards compatibility with 3rd party cordapps dependent on finance)
* Added backwards compatibility clarification comment.
* Re-instate new cordapp metadata.
* Global rename of `finance-flows` to `finance-workflows` to follow adopted naming conventions.
* Addressed final review comments.
* Rename application to "Corda Finance Demo"
* Generation of original corda-finance jar from new sub-modules.
* Fixed and tested demobench with new split finance contract and workflow jars.
* Renamed finance sub-modules to contracts and workflows.
* Remove Michele!!!
* Minor fix to filtering logic.
* Align CorDapp configuration filename with workflows jar.
* Fix breaks caused by finance module naming changes.
* Final alignment between OS/ENT of finance contract code.
* Implementation of Contract and Workflow attribute identifiers.
* Fixes following rebase from master.
* Fix broken JUnit test.
* Fix broken JUnit test.
* Fix broken JUnit test.
* Added missing constants.
* Further clean-up.
* Updated documentation.
* Added changelog entry.
* Updated all samples (using new Gradle Plugin 4.0.37 functionality)
* Temporarily resolve gradle plugins from latest published snapshot.
* Temporarily resolve gradle plugins from latest published snapshot.
* Updates following feedback from PR review.
* Move constants into CordappInfo companion object.
* Contract and Workflow attribute `version` to `versionId` (as version is a reserved gradle variable)
* Clarified warning message on incorrect version identifier.
* Align version identifier processing logic with gradle cordapp plugin.
* Updated comment.
* Minor fixes following rebase from master.
* Fixed broken unit test.
* Improved exception reporting.
* Update to use 4.0.37 of Gradle Plugins.
* Added support for combined Contract and Workflow CorDapp info.
* Updated following discussions with Shams + cleanup.
* Updated following Shams PR review.
* Minor API improvements.
* Added missing cordapp info causing deployNodes to fail.
FinalityHandler is insecure in that it is open to receive any transaction from any party.
Any CorDapp targeting platform version 4 or above is required use the new c'tors which take in FlowSession objects to the counterpart flow. This flow must subcall ReceiveFinalityFlow to receive and record the finalised transaction.
Old CorDapps (with target platform version < 4) will continue to work as previously. However if there are no old CorDapps loaded then the node will disable FinalityHandler.
* Introduce public subset of config to tweak config via mock net work without exposing internal node config.
* Removal of functions exposing (internal) NodeConfiguration from the public test API
* Code review fixes
* Blank lines removed
* Documented mock network API change in upgrade notes.
* Updated documentation and API doc.
* More documentation/API doc
* Migrated all non-BFT notary implementations to use async commits.
* Mock network: await for async operation completion. When calling runNetwork()
it keeps "pumping" messages between participants until no more messages are generated.
The problem comes in when a flow suspends on an async operation: the mock network
thinks the flow finished the work for the current step, and since no more messages
are generated, completes the runNetwork() function. The message that the flow
generates once it resumes after async operation completion never gets processed.
This change makes runNetwork() wait until all flow async operations finish,
and only then check whether no more messages can be transferred.
