v3.0.1

Changed a few things:

• Added tests folder and initial pass at unit and integration test cases.
• Refactored the way data sources were being loaded to make it easier to test and support. Initially we scanned the data sources folder and added the found classes. Now we are having the data sources self-register.
• Changed the backfill feature from maintaining a list of ticker mappings in this project to support a backfill_ticker option.
• Removed ability to merge cached data with new data. Original intent was to allow the ability to cumulatively build up a longer history if the data source were changing limits. However, this was causing issues with events like stock splits so I decided to only cache the data returned from the data source.

Full Changelog: v2.0.1...v3.0.1

v2.0.1

Fixed a few issues:

• Moved data source imports to inside the calling methods to support lazy loading. Now that they are lazy-loaded, the data sources are not required to be installed.
• Added init.py to the /backfill folder so that it would get added to the release

Full Changelog: v2.0.0...v2.0.1

v2.0.0

Added ability to backfill ticker data with oldest ticker from the fund share classes.

Some notes on the backfill capabilities:

• This is a breaking change as the 'ticker' column was added to the returned dataframe which will cause issues with any previously cached csv data files.
  • To not lose existing cached data, write a script to read in the existing csv file, add a ticker column and set the value to the ticker, save the results back into the same file.
  • If you don't need to preserve your cached data, the easiest option is to just delete the current cache files and let the cache rebuild.
• The backfill crosswalk file only maps Vanguard mutual funds.

Full Changelog: 1.1.0...v2.0.0

v1.1.0

First official release.

• Added ability to register custom data sources.

Full Changelog: https://github.com/mattsmith321/fin-ds/commits/1.1.0