Skip building R vignettes if Spark is not built

[shivaram]

What changes were proposed in this pull request?

When we build the docs separately we don't have the JAR files from the Spark build in
the same tree. As the SparkR vignettes need to launch a SparkContext to be built, we skip building them if JAR files don't exist

How was this patch tested?

To test this we can run the following:

build/mvn -DskipTests -Psparkr clean
./R/create-docs.sh

You should see a line Skipping R vignettes as Spark JARs not found at the end

[shivaram]

The only downside of this is that we won't be uploading our vignette to the website. I think this is fine -- Any thoughts @junyangq @felixcheung

cc @rxin

[SparkQA]

Test build #65778 has finished for PR 15200 at commit eb7e8b8.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[rxin]

I'm going to merge this in master/branch-2.0 because I need this to cut an rc.

[felixcheung]

we could hand write the result instead of running test/Spark to generate the vignettes?
It seems that could be problematic if output are getting out of sync - and similar problem if we build doc without jar and then just skip the vignettes?

maybe we should add vignettes to profile -Psparkr in Maven?

[shivaram]

Yeah - so I'm thinking we should just auto-generate this and check in the html file in git. Its not that big. When somebody updates the vignette we need to remind them to regenerate it though as a part of the PR ?

[felixcheung]

if it's part of the -Psparkr profile of the build it will be regenerated by default. If it's changed and not in .gitignore it should be flagged for commit..