Spark Notebook Versions Save

This minor patches v0.4.1 that broke scala 2.11 due to usage of sbt.

• fix click on logo goes to /
• fix launching spark-notebook from bin folder
• fix :dp and :cp for spark 1.3 and scala 2.10
• fix rendering of list of string ouput
• output stream toggle (shortcut t)

• Adds the auto plotting for sequence data returned in a block (using reflection)
• The auto plotting components are also available as such (Chart) and provide a simple way to present data whatever the type is. Chek also the ``
• inputs can be hidden using the keystroke i (like o for the output) and according menu entries have been added
• Support for spark 1.3.0 → becomes the default version
• Wizard for the "cluster" definition, which has to be renamed "configuration template"
• repos, deps and import are now a list of string in the metadata
• some fixes here and there.

Aknowledgments:

• @axlpado for the plots and reflection
• @virtualirfan for the ideas, time,... and the yarn support
• @minyk for the yarn support
• @marcinjurek for the mesos support
• @lucaventurini for the yarn support
• @huitseeker for the kind words :-D
• @pkerpedjiev for the docker support (on mac for instance)
• @xtordoir for everything you did
• @mandubian for the PR
• all 245 stargazers so far!

• notebook can now declare metadata for deps, repos, and specially cluster
• spark 1.2.1 support has been added
• scala 2.11 support added
• notebooks can be arranged in folders
• a logging panel has been added to display background tasks
• duplicate, rename and delete have been fixed

WIP and KI

• creation of a folder misses some UI to name it
• move notebooks is not possible yet
• cluster profiles (notebook metadata) aren't reusable easily

This version cleans further the subprocesses for the REPL, that is enables remote akka in order to deploy everything unrelated to the web app to it.

The classloading and serialization problems are now gone when running the spark notebook in play run.

Check out the S3 repo or docker repo for available releases:

• S3
• docker

This new version integrates new distributions, for different hadoop versions:

• 1.0.4 (default)
• 2.0.0-cdh4.2.0 (helpful when using ec2 scripts with hadoop version 2 enabled)

So the version concatenates (sep by _)

• the notebook version
• the spark version
• the hadoop version

New features are:

• display datatables for sequences of type having a Codec defined for.
• :sh scope to execute simple shell commands

Releases are available:

• on S3: https://s3.eu-central-1.amazonaws.com/spark-notebook/
• on Docker: andypetrella/spark-notebook

Fixes #52

Fixes the reload (CTRL+R, F5, ...) of notebooks containing space in the name.

The zip file can be found here.

Or the docker image can be used as well, check this.

This version is the very first one allowing binary distribution, thanks to play framework!

Check the README.md for information about its features.

Use

wget https://s3.eu-central-1.amazonaws.com/spark-notebook/spark-notebook-0.1.zip
unzip spark-notebook-0.1.zip
cd spark-notebook-0.1
./bin.spark-notebook

The default configuration will start the server on http://localhost:9000