Tutorial 7: Using REST API

GLAMpipe has a full API, which means that everything that can be done via user interface, can be done via API also. Unfortenately the REST API is not currently well documented.

Task 1

  • get list of all projects in GLAMpipe.
Background information

The api endpoint is /api/v1/projects. If you want to get only the names of the projects, then you can use /api/v1/projects/titles

curl http://demo.glampipe.org/api/v1/projects/titles

Task 2

  • Create a project called "Hello World".
Background information

The api endpoint is /api/v1/projects. You need to make a post request with "title" parameter.

curl -X POST -d "title=Hello World" http://localhost:3000/api/v1/projects

Note that this creates a project without any collections. In order to add collection you must create it as follows:

curl -X POST -d "{'params': {'title':'my collection'}}" http://localhost:3000/api/v1/projects/5b217d24951720000816fe85/nodes/collection_basic?type=collection

Task 3

  • Create a script that fetches the Helsinki City Orchestra concert data, makes it browseable by composer and year, and shows 10 the most popular composers at the end.

There is a python script that: https://github.com/GLAMpipe/python-api-examples

Basically the code is like this:

# create online csv read node
data = {'params': params_csv, 'collection': collection_id}
csv_read = addNode('source_web_csv', data, project_id);

#create extract node
data = {'params': params_extract_year, 'collection': collection_id}
extract_year = addNode('process_field_extract', data, project_id);

#create facet view node
data = {'params': {}, 'collection': collection_id}
facet = addNode('view_facet', data, project_id);

# run csv read node
print("Reading CSV...")
runNode(csv_read, settings_csv)

# run csv read node
print("Extracting years... this takes a while...")
runNode(extract_year, settings_extract_year)

# run facet node
print("Creating facet view...")
runNode(facet, settings_facet)

You can exceute the code like this:

python csv-read-project.py