Statcubeserver serves statistical data over a REST-API. The API is a fairly thin REST-wrapper for pydatacube functionality.
Statcubeserver requires pydatacube (usually latest git, developed in sync), CherryPy 3 and Python 2 (most likely 2.7).
There's no separate installer, just run scripts in the project root. Currently
supports loading PC-Axis datasets of the Helsinki Region Infoshare service
using the load_hri.py script, by running eg:
mkdir hri_data_files
python2 load_hri.py hri_data_files > hri_data_index.json
which downloads and stores the PC-Axis files to hri_data_files-directory and
writes metadata/index of them to hri_data_index.json.
The px_json_server.py can be used to serve this data:
python2 px_json_server.py hri_data_index.json
And if everything goes fine, the API can be accessed (by default) at http://localhost:8080/resources/ and a (quite spartan) web-UI at http://localhost:8080/browser/ .
WARNING! The API is still subject to change. Please inform the developers if you are using the server, so we can ensure backwards compatibility.
NOTE! The links here point to a server instance serving Helsinki Region's statistics. To work with a local version, change the host, port and path accordingly (eg. http://dev.hel.fi/stats/ -> http://localhost:8080/).
Click the links to see example results.
Browse datasets with a browser: /browser/
Preview a dataset and see available methods: /browser/?resource=<resource uri>
Get available datasets: /resources/
Get dataset metadata and methods: /resources/<resource id>/
Get a filtered dataset: /resources/<resource_id>/filter<&col1=cat1,cat2&col2=cat3 ...>
Get dataset's data (can be filtered) as "entries": /resources/<dataset path>/entries
Get data as "table": /resources/<dataset path>/table<?start=firstrow&end=lastrow>
Get data as "columns": /resources/<dataset path>/columns
Get grouped data ("pivot") as columns: /resources/<dataset path>/group_for_columns?as_values=<col1,col2...>
Get data as JSON-stat: /resources/<dataset path>/jsonstat