We were relying on the Java typename conversion to work for our AMQP
envelope type selection, and only special casing arrays of primitives.
However with nested arrays this breaks as the intermediate serialises
for the nested arrays still have no idea as to what the underlaying type
is as the assumption will be being made that the type is a boxed
primitive.
Solution is to compute the typename properly, walking down the nested
array chain
* Cache deserialized rather than serialized WireTransaction. Prevent repeated deserialization when adding signatures to the SignedTransaction.
* Added a test to check that stx copying and signature collection still works properly after (de)serialization
Update to use LedgerTransaction api
Push query output logic onto BaseTransaction and update usages where possible
Migrate a few more uses
Address some PR comments
Address some PR comments
Fixup after rebase
* Replace buggy classes in rt.jar with patched versions.
* Apply Java bugfixes to MacOSX.
* Initial support for patching Windows JRE.
* Resign DemoBench DMG after the bugfixe source has been deleted.
* Apply Java bugfixes for DemoBench on Windows.
* Registering anonymous identities now takes in AnonymisedIdentity
* AnonymousParty.toString() now uses toStringShort() to match other toString() functions
* Add verifyAnonymousIdentity() function to verify without storing an identity
* Replace pathForAnonymous() with anonymousFromKey() which matches actual use-cases better
* Add unit test for fetching the anonymous identity from a key
* Update verifyAnonymousIdentity() function signature to match registerAnonymousIdentity()
* Rename AnonymisedIdentity to AnonymousPartyAndPath
* Remove certificate from AnonymousPartyAndPath as it's not actually used.
* Rename registerAnonymousIdentity() to verifyAndRegisterAnonymousIdentity()
* * Store composite key in keystore from file for notaries's identity.
* Some refactoring.
* * Addressed PR issues
* * Remove unintended format changes
* * Fixed failing test due to getting keys from wrong keystore
Remove `createTwoNodes()` from mock network as its behaviour is inconsistent with creating a set of nodes. `createSomeNodes()` is generally a better fit and creates a network map and notary, and registers all nodes. Where that's not the intention, it's acceptable to manually create each node.
* POMs generated by publishing are now correct. The publish extension now requires an explicit call to configure the publishing instead of waiting until after evaluation. This prevents evaluation order issues with the artifact renaming code that causes the POM to have the original, incorrect, artifact names.
* Fixed new test compile issues caused by removal of some dependencies in test utils that caused webserver code to be automatically included in any project also compiling test utils.
* Implemented Kryo custom serializers for Field and KProperty types.
* Adjusted KPropertySerializer to use kotlin member properties upon read() due to failing RPC tests.
Added additional Kotlin and Java tests (CordaRPCClient, StandaaloneCordaRPCClient)
Annotated schemas to be CordaSerializable (required when referencing as KProperty in custom queries).
Cleanup some outstanding compiler warnings.
* Added client RPC Java integration and smoke tests to build.
* Clean up compiler warnings in Java tests.
* Fixed incorrect assertion expectation.
* Backed out Field and KProperty custom serializers.
* Backed out Field and KProperty custom serializers.
* Store VaultQueryCustom custom column references as name and class (from Java Field and Kotlin KProperty1 types respectively).
Custom serialization of Field and KProperty type no longer required.
* Removed blank lines as per RP review comments.
