Summary
Use concurrency/multithreading to significantly speed up downloads of a dataset's files from the cloud. Download times were measured to be almost 7 times faster on a test dataset of 30 datafiles, although this will vary depending on the number of files downloaded.
<!--- SKIP AUTOGENERATED NOTES --->
Contents
Enhancements
- Make `Dataset.download_all_files` multithreaded
Testing
- Mock Pub/Sub in subscription creation test
Other
- Improve documentation of `AppFrom` class
<!--- END AUTOGENERATED NOTES --->