FSO Metadata Auto R Package

Introduction

This repository aims to simplify the access to the Swiss Federal Statistical Office metadata. Following the implementation in the interoperability platform and the SIS portal, the APIs are made available here in R. This public library is made available for the internal FSO staff, the federal administration and for external actors.

Installation

You can install the library with

install.packages("remotes")
remotes::install_gitlab("DSCC/fso-metadata")

then at the beginning of your R script, you will need to

library("fso.metadata")

Sometimes when you try to install a package, you get an error like this "ERROR: loading failed for 'i386' when using the install_gitlab function from remotes. Currently, if 32-bit and 64-bit versions of R are installed, it seems devtools tries to build for both of them. This yields a loading failed for 'i386' error. To force building the package for your currently running R version use the INSTALL_opts argument of install_github:

remotes::install_gitlab("DSCC/fso-metadata",  INSTALL_opts=c("--no-multiarch"))

Now you can install a function with

install.packages64("fso-metada")

For more information, visit Confluence.

In case of ERROR: loading failed for 'i386'

Functionnalities

Based on the metadata that you want, you will call certain functions and parameters.

Codelists

1. Export a codelist based on an identifier

codelist <- get_codelist(identifier, environment, language, export_format, version_format, annotations)

Parameters:
    - identifier ("character"): the codelist's identifier
    - environment ("character", default="PRD" for production)
        Available are 'PRD', 'ABN', 'TEST', 'QA' and 'DEV'.
    - language ("character", default="all" for all languages, no filtering)
        Available are 'all', 'fr', 'de', 'it', 'en'.
    - export_format ("character", default="SDMX-ML"): the export's format. 
        Available are CSV, XLSX, SDMX-ML or SDMX-JSON.
    - version_format ("numeric", default=2.1): the export format's version 
        (2.0 or 2.1 when format is SDMX-ML).
    - annotations (bool, default=FALSE): flag to include annotations
Returns:
    - codelist (data.frame) based on the export format
        - a data.frame if export_format was CSV or XLSX
        - a json if export_format was SDMX-ML or SDMX-JSON.

Nomenclatures

1. Export one level of a nomenclature

one_level_df <- get_nomenclature_one_level(identifier, environment, level_number, filters, language, annotations)

Parameters:
    - identifier ("character"): nomenclature's identifier
    - environment ("character", default="PRD" for production)
        Available are 'PRD', 'ABN', 'TEST', 'QA' and 'DEV'.
    - level_number ("numeric"): level to export
    - filter (list): additionnal filters in form of named list
    - language ("character", default='fr'): response data's language 
        Available are 'fr', 'de', 'it', 'en'.
    - annotations (bool, default=FALSE): flag to include annotations
Returns:
    - response (data.frame): dataframe with 3 columns 
        (Code, Parent and Name in the selected language)

2. Export multiple levels of a nomenclature (from level_from to level_to)

multiple_levels_df = get_nomenclature_multiple_levels(identifier, environment, level_from, level_to, filters, language, annotations)

Parameters:
    - identifier ("character"): nomenclature's identifier
    - environment ("character", default="PRD" for production)
        Available are 'PRD', 'ABN', 'TEST', 'QA' and 'DEV'.
    - level_from ("numeric"): the 1st level to include
    - level_to ("numeric"): the last level to include
    - filter (list): additionnal filters in form of named list
    - language ("character", default='fr'): response data's language 
        Available are 'fr', 'de', 'it', 'en'.
    - annotations (bool, default=FALSE): flag to include annotations
Returns:
    - multiple_levels_df (data.frame): dataframe columns from `level_from` to `level_to` codes

As the APIs continue to be implemented, further functionnalities will be added.

Upcoming function: Data Structures

1. Get the data structure (not available yet)

data_structure <- get_data_structure(identifier, language)

Parameters:
    - identifier ("character"): the nomenclature's identifier
    - language ("character", default='fr'): the language of the response data. 
        Available are 'fr', 'de', 'it', 'en'.
Returns:
    - data_structure: data structure

Background

All the APIs made available in this library are also documented in Swagger UI should you want to do more experiments through a UI.

• Here for APIs of the interoperability platform (public).
• Here for dcat APIs (internal to configuration).

Example

Examples for each API are provided in the R Markdown.

Practical demo.

A documentation page is also available here.