Hessian serializer

[danomatic]

Based on the custom-serializer branch. Adds an additional serializer, HessianSerializer, that should be more performant and avoid some of the issues we experienced with JavaSerializer.

[coveralls]

Coverage decreased (-11.3%) to 82.988% when pulling 2131854 on danomatic:hessian-serializer into 99ea324 on arhs:master.

[coveralls]

Coverage decreased (-11.3%) to 82.988% when pulling bcabd3a on danomatic:hessian-serializer into 99ea324 on arhs:master.

[coveralls]

Coverage decreased (-4.2%) to 90.041% when pulling bcabd3a on danomatic:hessian-serializer into 99ea324 on arhs:master.

[coveralls]

Coverage decreased (-3.4%) to 90.871% when pulling ad42513 on danomatic:hessian-serializer into 99ea324 on arhs:master.

[danomatic]

According to this, "The efficiency of Hessian 2.0 is about twice that of java.io serialization."

[danomatic]

My company has been using this in production for months without issue.

[danomatic]

That being said, this does add an extra Maven dependency. If that is problematic, it could be split to another project or we could just demonstrate how to use it in the docs.

[dnlprplt]

Ideally, we'd have to generate another jar with the serializer so that people could decide whether or not to include that dependency.

In the first place, I could live if we simply include the serializer in the main library.

[dnlprplt]

If Hessian2Output implements java.lang.AutoCloseable, could you please use a try-with-reource block?

It not, you should add another try block and close the stream in case an exception is raised.

[dnlprplt]

You should close the stream once it has been read. Please also consider the comment made on the serialize() part.

[coveralls]

Coverage decreased (-4.4%) to 89.837% when pulling d59bb4b on danomatic:hessian-serializer into 99ea324 on arhs:master.

[danomatic]

@danielprplt I've added code to make sure that the input and output streams are closed.

[coveralls]

Coverage decreased (-4.3%) to 89.96% when pulling 8d27dfb on danomatic:hessian-serializer into 99ea324 on arhs:master.

[cyrilschumacher]

@danomatic Could you resolve the conflicts present in your branch? In any case, thank you for your work.