Takes a elasticsearch mapping file, flattens it, filters out exact fields and returns a unique list of keys that can be consumed by get_deep()

Simple drugname normalization: strip punctuation and whitespace and lowercase.

Simple ndc normalization: strip letters, whitespace, and trim 10 digit ndcs.

A helper function for extracting a specific subset of keys from a dictionary.

Doing cleanup work here so that the json to be loaded in to ES is date friendly and the naming conventions line up with the API standard

Moving this code from the `spl_harmonization` file, since it is the only part of that file that is needed now that we have converted to a

Converts input dates for known formats to a standard format that is Elasticsearch friendly. Returns the input_date if it is not a known f

Hash function used to create unique IDs for the reports

Identifying the NDC type is based on string length WITHOUT '-', we need to strip all non-numeric characters before we look at the length

Helper function to make the export quarters code more readable. This function does two things: exports all data that is NOT in between two

Isolating the thorny naming logic for the display name in a helper function. The naming is solely present for url cleanliness. Possib

Takes several rows for the same report_number and merges them into a single report object, which is the final JSON representation, ba

Just like self.assertTrue(a >= b), but with a nicer default message.

The default behavior for splitting files is to have one split per file.