Stream: python
Topic: intro to the Python FHIRClient
Eric Haas (Jun 12 2019 at 18:20):
@ Bob Milius and I created some Jupyter notebooks that are helpful to get started.
https://github.com/bmilius-nmdp/pytestfhir
https://github.com/bmilius-nmdp/pytestfhir/blob/master/makePatient.ipynb
https://github.com/bmilius-nmdp/pytestfhir/blob/master/makeSpecimen.ipynb
https://github.com/bmilius-nmdp/pytestfhir/blob/master/makeSequences.ipynb
https://github.com/Healthedata1/smart_pythonclient_example if you are truly new to python this Notebook links to several helpful YouTube videos on how to set up a local Python environment.
Ward Weistra (Jun 12 2019 at 22:19):
Thanks so much @Eric Haas (and @ Bob Milius), especially the useful help functions I hadn't found yet.
Would you also know if there is any functionality to work with custom profiles and validating data before sending it to the server?
Eric Haas (Jun 13 2019 at 00:34):
For custom profiles you have to create the object class using the fhirparser and add it to your library. The validation is not as thorough as for example doing a $validate operation on ref implementation that has registered the profile (E.g. terminology)
Eric Haas (Jun 15 2019 at 00:18):
I was mistaken... I tried creating python classes out of profiles on the Smart fhir parser and unfortunately it is not yet capable of generating classes for profiles. I have however successfully created logical models as python classes.
Eric Haas (Jun 15 2019 at 00:21):
@Harold Solbrig This ( creating python classes out of profiles like this one ) is something we should as to our TODO list.
Last updated: Apr 12 2022 at 19:14 UTC
