Parsons

**BigQuery**
A new connector that allows you to query BigQuery databases. Future releases will include the ability to copy data into BigQuery.
- `BigQuery.query()`

**Twilio**
A new connector allows you to access account usage and messages that you have sent from your account.
- `Twilio.get_messages()`
- `Twilio.get_account()`
- `Twilio.get_accounts()`
- `Twilio.get_account_usage()`

**VAN**
- `VAN.upload_scores()` - New method that significantly simplifies the loading and approving of scores to VAN.
- Various bug fixes to improve the stability of the connector. Under the hood, all of the connectors have been refactored to make the code more rational and the responses more standardized.

**Table**
- `Table.first` is a new attribute that returns the first value in a Parsons table. This is useful for database queries that only return a single value and keeps you from having to index the table directly.


tbl = rs.query("select count(*) from schema.table")
tbl.first
>>> 10



my_count = rs.query("select count(*) from schema.table").first
>>> 10


- `Table.from_csv()` Improved the error handling with this to raise an exception if you are trying to load in any empty CSV file.

**Windows Usage**
There was a bug in the handling of temporary files which prohibited Windows users from running many Parsons methods. This has been resolved.

**Potentially Breaking Changes**

- *ActionKit Class*: Previously, to instantiate the class, you passed in the subdomain of your AK instance. Now, you have to pass in the entire domain. Previously: `myorg`, Currently `myorg.actionkit.com`. This was done to support users who have fully custom domains for their AK instance. (Ref `ActionKit()`).

- *Repo Renamed*: The GitHub repo has been renamed from `parsons_public` to `parsons`.

- Removed `pandas` dependency. This is now an optional dependency. It is required for the `Table.to_dataframe()` method.

**Other Updates**

- Removed various documentation build dependencies that are irrelevant to usage of the package. This and the removal of `pandas` significantly shrinks the size of package on install.

- Fixed various `Table.to_csv()` compression bugs when archiving with zip.

- Added `SFTP.get_file_size()` method to access the size of file without having to first download it.

- Internal improvements to VAN class.

- Added new `Hustle` class with access to all of their API endpoints.
- Added new `CrowdTangle` class with access to many of their API endpoints.
- Added `MobilizeAmerica.get_events_organizations()`
- Added `MobilizeAmerica.get_people()`
- Added `zip` compression to `Table.to_csv()` and `Table.to_s3_csv()`
- Added `Table.to_csv_zip()` method to add multiple csv to the same archive.
- Added `specifycolumns` argument to `Redshift.copy()` that if appending to an existing table or truncating, will map to existing columns. This allows for copying of tables that might have column names out of order to include fewer columns than are in the target table.
- Added the `CivisClient.client` attribute which allows you to access the Civis API client directly.
- Updated VAN documentation.
- Grammar and formatting updates to the documentation.

Initial public release.