Semi-Structured Data Manipulation

Data is generated by a wide range of systems, in an endless variety of shapes. Sometimes, the data you’ll need to work with won’t conform to the tabular structure of rows and columns that we tend to associate with relational databases. Semi-structured data refers to data that does not follow a strict tabular schema or data model. Such data may feature some or all the following properties:

• Composite data structures involving multiple values within a single entry, such as sequences of values and groupings of named values within a single entry
• Flexible schemas, both within composite values and in the shape of data across records
• Nested data, where composite values contain further composite values, enabling the modeling of hierarchical data within individual values

A particularly well-known form of semi-structured data is JavaScript Object Notation (JSON), which you will almost certainly find yourself having to wrangle...