Table of Contents
- Terms of Use
- GraphiQL implementation
- Queryable Objects
- Simple Examples
- Item Data
- Pagination
- More Example Queries
- Writing Data
- Ruby Example--Retrieve First 5 Items
- Ruby Example--Using Pagination To Retrieve All Items
- Python Example
- Feedback
Use the eScholarship API to query for Items (publications), Authors, and Units (journals, series, departments / research units) within eScholarship.org, returned as JSON. The API uses the GraphQL query language.
Terms of Use
Use of the eScholarship Public (Read) API must abide by the eScholarship API Terms of Use. Please do not abuse the API with thousands of requests per hour for hours at a time. Users who make excessive API requests that compromise service quality may be throttled or cut off.
GraphiQL implementation
https://escholarship.org/graphql
This web based IDE provides a tabbed interface for editing and testing GraphQL queries against the eScholarship API. The Documentation Explorer allows you to drill down into each object, its fields, and arguments.
Queryable Objects
The IDE is also where to find the set of queryable objects. Click on the "Docs" link in the upper right hand corner to search for objects and their components and attributes.
Simple Examples
Get all eScholarship items (default: first 100 listed, ordered by most recently added):
{ items{ nodes { title permalink } } }
Get an eScholarship item using a DOI:
{ item(id: "10.3945/jn.114.202333", scheme: DOI) { permalink title journal volume issue issn } }
Note that the primary id for an Item is an ARK ID. The above example is a query using a alternative identifier scheme: e.g specifying item(id: "ark:/13030/qt01r1171s") points to the same Item.
Item Data
The Item query provides the richest amount of data of all the objects, including:
permalink - the page where the item and corresponding metadata resides on eScholarship: i.e. https://escholarship.org/uc/item/65k8r3bd
contentLink - the download link for PDF/content file (if applicable)
Explore the GraphiQL implementation to see all fields available.
Pagination
Arrays of objects (Authors, Contributors, Items, and Units) are delivered using the nodes field. These object arrays can also be accompanied in the selection set with the following utilitarian fields describing the set:
total - Approximate count of all objects that can be returned
more - Opaque cursor string for next page, if total exceeds limit (set to 100 by default)
The results of the following query include the cursor string more that is used to retrieve the next page of 10 results:
{ unit(id: "ucsdecon_rw") { name items (first: 10) { total more nodes { permalink title } } } }
To retrieve the next page of results, query again. This time, embed the string provided by the more field into the more argument for items. Do not specify any other arguments with this one; the string already encodes the prior set of arguments (in this example, first: 10 should be removed in next query). Each query using a more argument provides a new cursor string for the following page. If it has reached the end, the returned more value will be null.
{ unit(id: "ucsdecon_rw") { id name items(more: "eyJmaXJzdCI6MTAsIm9yZGVyIjoiQURERURfREVTQyIsImxhc3RJRCI6InF0MWZ6MTc4NDciLCJsYXN0RGF0ZSI6IjIwMTYtMDMtMTEifQ") { total more nodes { permalink title } } } }
More Example Queries
Get the title, published date and authors for all the Items in a Unit:
{ unit(id: "lmri_pr") { id name items { nodes { title permalink published authors { nodes { name } } } } } }
Get the first 50 Items with a certain “tag” (i.e. keyword or subject) within a given Unit:
{ unit(id: "lbnl") { id name items(first: 50, tags: "subject:Environmental sciences") { total more nodes { title permalink subjects } } } }
Get all the items associated with an author (exact for now, fuzzy in the future):
{ author(email: "[email protected]") { id name items { total nodes { id title } } } }
Get all the items of a certain type:
{ items(tags: "type:CHAPTER") { total nodes { bookTitle title permalink published } } }
Get all the items published between certain dates:
{ unit(id: "ucd") { name items(after: "2017-01-01", before: "2017-12-31", order: PUBLISHED_DESC) { total more nodes { permalink title published } } } }
Determine when an item first appeared on eScholarship
{ item(id: "ark:/13030/qt3dw6f08j") { id title journal volume issue issn added } }
Get an item using a UCPMS ID:
{ item(id: "709820", scheme: OA_PUB_ID) { id title journal volume issue issn added } }
Get the date of expiry of an embargo of a thesis or dissertation:
{ item(id: "ark:/13030/qt2k3957sr") { id title type added embargoExpires } }
Writing Data
The eScholarship API also allows writing data using QraphQL’s mutations, including the ability to create, replace, or withdraw items. See the mutation root type in the GraphiQL implementation for more details.
Ruby example : retrieve first five items
require 'httparty' query = "{ items(first:5) { nodes { title permalink } } }" result = HTTParty.post("https://escholarship.org/graphql", :headers => { 'Content-Type' => 'application/json' }, :body => { "query" => query }.to_json) result.code == 200 or raise "HTTP error #{result.code}: #{result.body}" result['errors'] and raise "GraphQL error: #{result['errors']}" puts JSON.parse(result.body)['data']
Ruby example : use pagination to retrieve all items
require 'httparty' more = "null" loop do query = "{unit(id: \"vertebrate_pest_conference\"){ id name items (more:#{more}) {total more nodes { permalink } } }}" result = HTTParty.post("https://escholarship.org/graphql", :headers => { 'Content-Type' => 'application/json' }, :body => { "query" => query }.to_json) result.code == 200 or raise "HTTP error #{result.code}: #{result.body}" result['errors'] and raise "GraphQL error: #{result['errors']}" #puts JSON.parse(result.body)['data'] data = JSON.parse(result.body).dig('data', 'unit','items') more = data.dig('more') #print ARKS data['nodes'].each {|item| puts "#{item['permalink']}" } break if ! more more = "\"" + more + "\"" end
Python example
import requests
import json
query = "{ items(first:5) { nodes { title permalink } } }"
response = requests.post('https://escholarship.org/graphql',
headers = { 'Content-Type': 'application/json' },
data = json.dumps({ 'query': query }))
if response.status_code != 200:
raise Exception("HTTP error %d: %s" % (response.status_code, str(response.content)))
if 'errors' in response.json():
raise Exception("GraphQL error: " + str(response.json()['errors']))
print(response.json()['data'])
Feedback
We'd love to hear how you're making use of the eScholarship API - please share your story with us!