v0.2.0

Much has changed! Some parameters names have been deprecated and others have been added, below there is a summary of all the pull requests.
The most important changes are:

• Some fuctions have been added to avoid using query_to_dataframe.
• Bundle caching
• If multiple resources are queried, a dictionary with multiple DataFrames is returned.
• Read operations are now supported

What's Changed

• Update to 3.8 because of numpy security update by @giuliabaldini in #64
• Convert basic bundle collection to generator by @giuliabaldini in #60
• Add history support by @giuliabaldini in #74
• Add handling for dataframe functions using the _id parameter by @giuliabaldini in #72
• Fix the FHIR server URL bug for local servers by introducing a new variable that contains the domain by @giuliabaldini in #78
• Allow to specify multiple query arguments in df contraints by @giuliabaldini in #76
• Specify Package Version Inside Package by @giuliabaldini in #80
• Allow to turn off checks for dicom download by @giuliabaldini in #85
• Add resource name to the TQDM description by @giuliabaldini in #88
• Merge trade rows for dataframe and trade rows for dataframe with ref by @giuliabaldini in #89
• Allow addition of any column to the trade rows for dataframe result by @giuliabaldini in #90
• Convert logging.warn to warnings when appropriate by @giuliabaldini in #105
• Convert received resources from read operations to bundles for consistency by @giuliabaldini in #106
• Smarter bundle caching by @giuliabaldini in #107
• Make current beta compatible with 0.1.0 by @giuliabaldini in #112
• Fix efficiency problem with merge on by returning dataframes @giuliabaldini in #113
• Make fhirpathpy import greedy by @giuliabaldini in #116
• Support for custom certificates needed by @giuliabaldini in #119
• Improve documentation by @giuliabaldini in #122

Full Changelog: v0.1.0...v0.2.0