build.xml

Issue: MIRTH-1793

Instead of "Delete After Read", there is now an "After Processing Action". You can select "None", "Move", or "Delete", depending on what you want to do after a message has been successfully processed through the channel. The Move-to Directory / File Name are still there, and apply to this action.

There are now two different error actions. The first, "Error Reading Action", determines what should happen when the file failed to be read in and processed by the channel.

The other one is "Error in Response Action", which determines what to do when the message was successfully processed, but the status of the selected response is ERROR. In this case, you can choose "After Processing Action", "Move", or "Delete". If "After Processing Action" is chosen (the default), ERROR responses will just cause the After Processing Action to be performed, and not do anything different.

In addition to the error move-to directory, there is also an error move-to file name. Both of these fields apply only to the Error Reading and Error in Response actions.

Added two SerializationTypes that each data type must define (RAW or XML). Currently, the raw datatype uses RAW, and all other data types use XML. The SerializationType determines what datatype the transformed content is stored as, as well as what type of object msg and tmp will be in the transformer scripts.

DataType now stores the SerializationType so the transformer executors can utilize them.

Issue: MIRTH-2275

DataType now contains a BatchAdaptor.

Plugin classes have been updated to accommodate this change.

DataType (and related) have been moved from model to server since they are not used by the client.

AutoResponder (and related) have been moved from model to server since they are not used by the client.

Issue: MIRTH-2357

Highlights:

- The JMS connectors now have the ability to automatically reconnect to the JMS broker if the connection goes down.

- JMS message acknowledgments are sent only after the message has been persisted by Mirth Connect, for guaranteed delivery.

Issue: MIRTH-1945

Inbound properties consist of SerializationProperties, BatchProperties, and ResponseGenerationProperties.

Outbound properties consist of DeserializationProperties, ResponseValidationProperties, and Template SerializationProperties.

The data types dialog has been completely revamped.

The "Properties" button is gone.

Both inbound and outbound properties are now visible once a connector is selected.

The property editor panel (tree and description) has been written from scratch.

The old editor (bean version) and beaninfo classes have been removed. The property descriptor information now comes from the respective Property Group object.

X12 data type has been combined with EDI. They are now referenced as "EDI / X12".

Updated DataType client and server plugins.

Moved getDefaultProperties from DataTypeDelegate to DataTypeClientPlugin since the server doesn't use it.

Serializers now take in SerializerProperties (which contains Serialization, Deserialization, and Batch properties).

MIRTH-2274

Issue: MIRTH-2273

Issue: MIRTH-2325

Issue: MIRTH-2325

The "Basic TCP" transmission mode is technically a plugin, but it's hard-coded and can't be removed by the Extension Manager.

"MLLP" was refactored as a separate plugin.

StreamHandler classes were moved from tcp.stream to model.transmission because they can potentially be used by more than just TCP connectors (e.g. RS-232).

Issue: MIRTH-2250

Changed the access rules in all projects that include Server so that they can't access anything that isn't in mirth-client-core.jar or any connector-shared.jar

Cleaned up the classpath orders and build connector orders.

Issue: MIRTH-2250

Renamed VmListenerProperties to VmReceiverProperties

Renamed VmSender to VmDispatcher

Renamed VmSenderProperties to VmDispatcher

Updated build references. Please rebuild and recreate database.

Issue: MIRTH-2250

Separated coupling between JdbcDao and ChannelController so that JdbcDao no longer calls ChannelController to lookup the localChannelId for a channelId, which in turn called JdbcDao again. The map of channelIds to localChannelIds is now stored in JdbcDao.

Donkey's channel storage is now initialized on channel deploy.

Modified channel deploy to check if a ResponseSelector has been defined and if not, create one using the source connector's inbound data type.

Fixed some potential problems with error handling in donkey Channel/DestinationChain.

Changed XmlQuerySource.getQuery() and CachedPreparedStatementSource.getPreparedStatement() so that they return null if there is no query for the given queryId, rather than throwing an exception. This was done since some queries are only defined for specific databases.

Cleaned up donkey's unit tests and ensured that database connections, statements, and result sets are only opened when needed and are closed as soon as possible in finally blocks.

Removed the com.mirth.connect.server.tasks package and the classes within it, which are unused.

Added / cleaned up indexes on donkey tables.

Added debug log output to JdbcDao.

Made the custom meta-data columns case-insensitive by automatically converting them to uppercase.

Issue: MIRTH-2250

Issue: MIRTH-2263

Issue: MIRTH-2250

Issue: MIRTH-2050

Issue: MIRTH-2050

Issue: MIRTH-2004

Issue: MIRTH-2004

Issue: MIRTH-1890

Issue: MIRTH-1909