statistikzh/statbot

statbot

PROJECT STATBOT.SWISS

LATEST: THE PROJECT IS BEING CONTINUED ON RENKU UNDER THE FOLLOWING LINK: https://renkulab.io/projects/christian.ruiz/statbot-swiss

In this folder are the main data warehouse scripts that run all the different R and python scripts in order to generate the newest data for statbot. It is supposed to be built as some form of cronjob to be executed once a day. All the statistical offices can contribute here with scripts. Or they can contribute by adding additional data-values-files and we will find a way to grab those files and insert the values.

For those who would like to participate, we prepared 'under this link a documentation' that can be followed step by step.

Important for this project is a DB-schema, in which all the data are collected. Right now the following schema is used:

There are several elements to it. The fact-table is the central element, where the actual data of the datasets are saved. Thus, every dataset is saved in this format and only differs according to certain dimensions (as of V3).

Next to the fact-table, there is a class-table that contains all the datasets (called classes) and their meta-data. It contains a table, where all the spatial units are saved. 'For a detailed explanation on the spatial units table click here'. And it contains a dimensions-table where all the dimensions such as gender, age and origin and their corresponding values are saved.

Important Temporary-Notice: The pxR-package to read px-files within R has a bug. Use the script here in statbot_read_px.R to read large px-files until the authors fix that package (PR is there pending for a while). It is included in this R-environment and is thus loaded automatically if you run the R-project.

We added some additional documentation to some functions, with which 'metadata can be extracted from px-cubes'.

Repo-structure

• R/ and python/ folders: include here all the R- and python scripts of your office (see also the possibility to generate a template with R)

• temp: files are downloaded here temporarily

• data/hash.csv: contains all the MD5sum hashes of the files - so we can check whether files are up-to-date

• data/: contains all the main DB-structural csv-files about classes, spatial units, dimensions.

• data/values: here the actual dataset data is stored (the so-called fact tables)

• scripts/init_scripts/: contains some of the initialization scripts. However, some have been moved to R/ as they are run as functions

• log/folder will be created if you run the main.R script and contains a log of the main-script and all of the elements.

• constants/ contains a constants.json file with constants that can be shared in R and python environments

• db_schema/ contains a txt-file with the current db_schema that can be visualized e.g. with https://dbdiagram.io/d

• statbot_env/ is the virtual environment for python (you still need to pip install the requirements)

• readme_folder/ contains some additional readmes that are linked in this file already

• man/ R-environment specific manuals that are automatically built

• renv/ R-specific R-environment

• presentations: contain slides about project STATBOT (contact us for more)

• old: contains all the files and folders from STATBOT V.0.x.x