Allow both read and write of expressions (${key:default}) in the metadata

Port the IronJacamar tracer, plus HTML report generator

Validation (#478) should be part of deployer chain.
Add aa RA throwing an appropriate exception when created if validation rules are not passing.

TODOs are

• Separate configurations for the Ant, Maven and Gradle artifacts
• Standalone validator scripts
• Migrating the test suite to testsuite/

Add support for XMLStreamWriter to the parsers, except RaParser

Hello

I search the source code of ironjacamar-deployers-fungal jar, do you know where it is ?
(particularly org.jboss.jca.deployers.fungal.RADeployer)

Thanks in advance

Need support on embedded and also on standalone

Add the specification javax.resource package as a module to the project

Implement a new inflow architecture based on deployment repository

Port logging codes and messages from IJ 1.x
Review common, core and deployers module adding all logs needed for debugging.
i18n is required for log level < DEBUG
Take care also of trace message using our Tracer API #487

Create XSDs for IronJacamar 2.0, based on the latest 1.x ones

Both for outbound and inbound

Port form 1.2 and update user guide in new documentation environment (#528)
(PDF, HTML, web)

Documentation too

Foreground and background validation of connections, with fast fail escape

Porting from 1.2 paying attention to:

1. Cleanups in regards to security integration
2. "Encrypted" connection properties

Port form 1.2 and update user guide in new documentation environment (#528)
(PDF, HTML, web)

We need test cases for DWM using embedded multiple instances.

Evaluate also tests for crashing instances

Port the metadata from IronJacamar 1.2, remove enum usage, and hide XML node information

The new pool architecture should allow different strategies to be used on a per deployment basis

In org.jboss.jca.adapters.jdbc.WrappedPreparedStatement the different delegated methods all use lock/unlock and checkState but some of them use the ps field directly, others use getUnderlyingPreraedStatement (which does lock+check itself). Is this intentional or should all use ps. directly?

It looks like at the beginning of the flle ps. was used (setBoolean, setByte, ..., setArray, getParameterMetaDate), and since getPoolable/setPoolable used the underyling one the following methds used it by accident:

https://github.com/ironjacamar/ironjacamar/blob/1.2/adapters/src/main/java/org/jboss/jca/adapters/jdbc/WrappedPreparedStatement.java

If it is intentional, I would suspect it should be used in the old methods as well (and we can skip the checkState() call then.

Porting all things from 1.2 branch.
Create a specific module for rarinfo and add it to distribution
update also user documentation if needed

Enable Travis CI integration on GitHub for master, so every pull request will be automatically tested . Of course, this adds to the existing Jenkins jobs that are still run daily against current master on cloudbees.

Add a janitor concept to the pool.

Functionality:

• dumpQueuedThreads
• getConnectionListeners
• killConnectionListener

https://issues.jboss.org/browse/JBJCA-1326

Include them in distribution's documentation.
Consider also to include IJ specific public API javadoc

My whatever-ds.xml has:

<?xml version="1.0" encoding="UTF-8"?>

<datasources xmlns="http://www.ironjacamar.org/doc/schema"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.ironjacamar.org/doc/schema http://www.ironjacamar.org/doc/schema/datasources_1_3.xsd">

   <datasource jndi-name="whatever-ds" pool-name="ExampleDS">
        <connection-url>
            jdbc:h2:target/h2-test;MODE=MySQL
        </connection-url>
        <driver>h2</driver>
        <transaction-isolation>TRANSACTION_READ_COMMITTED</transaction-isolation>
        <pool>
            <min-pool-size>2</min-pool-size>
            <max-pool-size>5</max-pool-size>
            <prefill>true</prefill>
        </pool>

        <security>
            <user-name>sa</user-name>
            <password>sa</password>
        </security>

        <statement>
            <prepared-statement-cache-size>32</prepared-statement-cache-size>
            <share-prepared-statements>true</share-prepared-statements>
        </statement>
    </datasource>
</dataSources>

So, <share-prepared-statements>true</share-prepared-statements> generates an error, even though I think this is valid. I only took a quick glance, but I couldn't figure out why the XSD didn't allow this.

New ConnectionManager implementations, especially for LocalTransaction and XATransaction scenarios

Add a CachedConnectionManager that support merging of ConnectionListener upon transaction start

Port form 1.2 and update user guide in new documentation environment (#528)
(PDF, HTML, web)

Include JUnit4 testing extension