Metadata API

Overview

One of the core features of the library is to allow retrieving metadata that can be used to power statistical processes.

Note

Discover the range of statistical processes enabled by the SDMX metadata through the metadata API in the following tutorials:

• Generate the Filesystem Layout.

• Create the Physical Data Model.

• Validate your data.

• Map Your Data.

These metadata are typically stored in an SDMX Registry, such as the FMR. However, retrieval is not limited to specific registries; any service compliant with the SDMX-REST 2.0.0 specification, allowing metadata retrieval in SDMX-JSON 2.0.0, can be used.

To retrieve metadata, a pysdmx client is necessary. Two types of clients are available: one for synchronous (blocking) retrieval and another for asynchronous (non-blocking) retrieval. At a minimum, the registry endpoint must be provided during client instantiation to specify where the metadata will be fetched from.

Once the client instance is obtained, various methods can be utilized to retrieve the desired metadata.

Examples:

Let’s illustrate the process with an example. Assume we wish to retrieve information from the SDMX Global Registry for data validation of the EDUCAT_CLASS_A dataflow maintained by the UNESCO Institute for Statistics (UIS). The following code demonstrates this:

>>> from pysdmx.fmr import RegistryClient
>>> gr = RegistryClient("https://registry.sdmx.org/sdmx/v2/")
>>> schema = gr.get_schema("dataflow", "UIS", "EDUCAT_CLASS_A", "1.0")

API Reference

Two clients are available for metadata retrieval, both exposing the same API. The only distinction is that one client operates synchronously (RegistryClient), while the other operates asynchronously (AsyncRegistryClient).

• Synchronous client
• Asynchronous client