add download_files_example.R to hackathon folder

[rabramoff]

#32
This commit adds a file with an example using the Walz, Josefine (2016) dataset that Cristy suggested in #27. It only downloads one file for now.

[ktoddbrown]

Thanks Rose! This is a great example and let's keep an eye on how best it can be incorporated into the main functions. I'm going to go ahead and accept the pull request but do you take a look at the reports markdowns and see where this could fit in there?

[rabramoff]

As long as ISCN3 doesn't take up too much space, then I think your original plan is best <- add data to ISCN3 so that people only have to load one data frame in the report markdown.
If ISCN3 gets really massive it might be useful for people to download data, make a frame with what they want, then delete the loaded data. Or even better (and more unrealistic as a volunteer-driven project :P) would be to save only metadata in this package - and have the user specify what kind of data they want (14C profiles, heterotrophic respiration, whatever), and then use scripts to download only the files that contain these data. For this the process_Data function would become a fetch data function with extra arguments to collect something specific.

[ktoddbrown]

ISCN3 is getting to the point where some versions of R might have issues loading it into memory (at one point in the ISCN3 load function I think we reach 500MB, though we end at 180MB). I think we are on the cusp of needing an actual database back end instead of a set of dataframes. But can probably limp along for a little while longer.

Actually saving the variable names for each dataset and having a way for users to only load data with specific variables isn't a bad idea. Using the key structure that shouldn't be too hard to accomplish.

I've add it to the feature wish list! here: 49a33e6