Aggregation script should write to existing file

[peterdesmet]

The aggregation script currently drops file with aggregated data and creates a new one. This is problematic:

1. The created file is for some reason not recognized by Google Drive (and thus CartoDB) as a CSV. It has a different icon.
2. The CartoDB sync is using the file ID to synchronize. If the script deletes the file to replace it with a new one, the sync will still use the old file (in the trash)

Proposal:

The script should use a file ID of an existing file, empty it and write the new data to it. As a fail safe, it could back up the old file.

[bartaelterman]

I don't think this is still relevant:

• We'll be using R scripts now, not Google Scripts. This means the scripts will run locally and file ids can differ on different machines and different operating systems.
• We wont sync this file with CartoDB anymore. We'll use an aggregated file for that, so this issue will apply to that file.

[peterdesmet]

Agreed. All issues have to be reviewed, and new issues and milestones set.