kproductivity / onsapir Goto Github PK

Note: 'A dataset (or slice) is represented in SDMX by two responses: Structure and Data. The ONS API, for convenience, allows you to receive the two responses in a combined message, but as this is not valid in SDMX an artificial ONSDataPackage entity is used to encapsulate them. If you are using SDMX-based tools (e.g. from Eurostat) you should always use the separate outputs (/dsd for the structure and /set for the data).
The Structure part of an SDMX message can contain Key Families, Concepts and Codelists, these three together comprise a DSD (Data Structure Definition). Parts of the DSD can be returned separately, or in a single response thing using dataset/[datasetid]/dsd.'

Need to consider that if dataset is too big, the extraction will be in the form of a zip file.

Controlling for the API [error] response codes:

• 500 - internal server error (unexpected)
• 407 - proxy authentication required
• 406 - not acceptable (usually an incorrect accept header)
• 404 - not found (could be misspelling or dataset id not found on database)
• 400 – Bad Request (usually a syntax error)
• 202 – Accepted, not complete (download being generated)
• 200 - OK (only returned for successful requests, never used for any type of error condition as this can cause ambiguity for the client application)

According to the API documentation, 'the root node of the API can be queried and the response includes the URLs for a list of contexts, a list of concepts, a list of collections and a list of classifications'.

A function that interactively queried the API to build the url can be programmed.