CHANGELOG

## Version 0.5.0

* Retries to download files and perform other api interactions 3 times
* Properly checks exceptions instead of using general cases
* Immidiately stops download and/or copy when receiving a end signal
* Handles keyboard interupts properly
* Reduces memory usage during download by performing os.fsync
* Manually copy the file in chunks in order to stop network copies which may stall during interupt
* Reorganize files
* Added oauth generator
* Fix issues that would cause a custom web app implementation to improperly redirect
* In web app, use the python generated oauth when present
* In web app, show warning when not using python oauth token
* Added --noconsole param to prevent console logging
* Added versioning

## Version 0.5.1
* bugfix

## Version 0.5.2
* Save cpu by first trying download without flush
* Fix bug where a keyboard interrupt would not stop execution when using a single thread
* The access token now needs to be generated and stored using getfiles.py --oauth and getfiles.py --settoken
* Got rid of the local flag
* Added the --testauth flag to ensure a connection can be made

## Version 0.5.3
* Fix bug causing program to stall and recursion to fail
* Added -p, --progress flag to occasionally log progress
* Let the logging module print the thread number
* Better logging messages

## Version 0.5.4
* Major performance improvements, specifically with traversing filestructures
* Better exception handling
* When the api rate limit is reached, the program will always retry at an increased timeout
* Multi-threaded folder traversing
* Folder traversing is no longer done via recursion. Instead, there are multi-threaded "queuers" that will scan a directory, queue file downloads, and folder traversals.
* Better interrupt handling
* Better decoding of filenames and convert names with disallowed characters to proper os format
* Removed bug where some files would get copied twice
* File copies are now retried
* Temporary downloads are now done directly in the temp folder so there is no file structure left over
* If a file returns the same incorrect size 3 times it is kept
* Some other stuff I can't remember lol

## Version 0.6.0
* Converted the FileLister into a python web application
* Better handling of foreign characters
* Added upload to google drive
* Better thread management
* Better handling of shutdown
* Resume interrupted downloads
* Better handling of api limit errors when downloading
* Tons more, its late, I'm going to sleep LOL

## Version 0.6.1
* Using chardet to hopefully detect the proper encoding for file names
* Fixed a bug in the gdrive wrapper
* Set upload and download retry to 10
* During upload google drive will refresh the auth token if needed
* Use base64 path for sha1 hash to prevent unicode errors

## Version 0.6.2
* Using codecs.open to write raw unicode chars to files
* Fixed a small bug in copy.py
* Changed the name of getfiles.py so people will stop trying to use it :)
* Added capability to do a --dryrun
* Added capability to set --folderthreads max
* When testing auth take into account api limit
* Don't launch browser when using --nolaunch with oauth

## Version 0.6.3
* Log the full path in successfiles
* Waits for uploads/moves to complete if temp disk is full
* Bugfixes

## Version 0.7.0
* Better analytics printed every 2 minutes when --progress is present
* Upload local files
* Better handling of exit
* Prevent bug that would improperly cause threads to shutdown
* Better handling of google drive api limit errors

## Version 0.7.1
* Fixed issues with `--dryrun`
* Proper log when running local
* Put custom api creds in bitcasa.ini fixes #56 
* Fix encoding issues. Fixes #62 Fixes #55 
* Fix some api rate limit issues where it wouldn't sleep properly
* Fix a bug where the improper variable was used