Datasync python script

[eldang]

Resolves https://github.com/eldang/ballard-4culture/issues/1

[eldang]

@coregis I think this does everything we need it to, though I won't be surprised if we end up needing changes to json formatting details as we go. What I've ended up making is:

• data/places.geojson is the only true geojson file, and the only one that should be loaded as a Mapbox layer. Each feature has a people field which is a list of cross-references to people IDs.
• data/people.json is a non-spatial data file, which compiles everything else about a person including a places field which is a list of cross-references to place IDs.
• The mp3, transcripts, images, and other_media folders contain local copies of everything that is both referenced from a person record and found in the appropriate Google Drive folder. The fields with the same name in each person record are just lists of filenames in those folders.
• All code and auxiliary files are in datasync, and it makes temp subdirectories which it cleans up on completion.

I think if you make a map that uses places.geojson as a data source, and has mocked-up popups and controls, I can take back over from there and do the in-browser parsing of people.json to fill popups and populate search menus. I think I've structured this in such a way as to make that work easy, but we'll see how it goes in reality....

[coregis]

@eldang WOW you've made some good progress! I am crunching to prep for a 1:00 pm dog-and-pony and will review this in detail once preparations are complete or the presentation is over, whichever comes first