rspace-client-python

This project contains a client which helps calling RSpace APIs. There are some example Python scripts.

To begin with you'll need an account on an RSpace server and an API key which you can get from your profile page. You can sign up for a free RSpace account at https://community.researchspace.com

In these examples we'll be using the rspace_client package (code is in rspace_client folder) which provides an abstraction over lower-level libraries. It's compatible with both Python 2 and Python 3.

All the code listed here is in the project.

For full details of our API spec please see https://your.rspace.com/public/apiDocs

To install rspace-client and its dependencies, run

pip3 install rspace-client

You may need to install six and requests modules:

pip3 install six requests

To run the example scripts in the examples folder, cd to that folder, then run

python3 ExampleScript.py https://your.rspace.com MyAPIKey

replacing MyAPIKey with your key, and ExampleScript.py with the name of the script you want to run.

A basic query to list documents

First of all we'll get our URL and key from a command-line parameters.

parser = argparse.ArgumentParser()
parser.add_argument("server", help="RSpace server URL (for example, https://community.researchspace.com)", type=str)
parser.add_argument("apiKey", help="RSpace API key can be found on 'My Profile'", type=str)
args = parser.parse_args()

client = rspace_client.Client(args.server, args.apiKey)
documents = client.get_documents()

In the above example, the 'documents' variable is a dictionary that can easily be accessed for data:

print(document['name'], document['id'], document['lastModified'])

Iterating over pages of results

The JSON response also contains a _links field that uses HATEOAS conventions to provide links to related content. For document listings and searches, links to previous, next, first and last pages are provided when needed.

Using this approach we can iterate through pages of results, getting summary information for each document.

while client.link_exists(response, 'next'):
    print('Retrieving next page...')
    response = client.get_link_contents(response, 'next')

A complete example of this is examples/paging_through_results.py.

Searching

RSpace API provides two sorts of search - a basic search that searches all searchable fields, and an advanced search where more fine-grained queries can be made and combined with boolean operators.

A simple search can be run by calling get_documents with a query parameter:

  response = client.get_documents(query='query_text')

Here are some examples of advanced search constructs:

    // search by tag:
    search = json.dumps([terms:[[query:"ATag", queryType:"tag"]]])
    
    // by name
    search = json.dumps([terms:[[query:"AName", queryType:"name"]]])
    
    // for items created on a given date using IS0-8601 or yyyy-MM-dd format
    search = json.dumps([terms:[[query:"2016-07-23", queryType:"created"]]])
    
    // for items modified between 2  dates using IS0-8601 or yyyy-MM-dd format
    search = json.dumps([terms:[[query:"2016-07-23;2016-08-23 ", queryType:"lastModified"]]])
    
    // for items last modified on either of 2  dates:
    search = json.dumps([operator:"or",terms:[[query:"2015-07-06", queryType:"lastModified"],
                                    [query:"2015-07-07", queryType:"lastModified"] ])

    // search for documents created from a given form:
    search = json.dumps([terms:[[query:"Basic Document", queryType:"form"]]])
    
    // search for documents created from a given form and a specific tag:
    search = json.dumps([operator:"and", terms:[[query:"Basic Document", queryType:"form"], [query:"ATag", queryType:"tag"]]])

or by using AdvancedQueryBuilder

# Creation date (documents created between 2017-01-01 and 2017-12-01
advanced_query = rspace_client.AdvancedQueryBuilder().\
    add_term('2017-01-01;2017-12-01', rspace_client.AdvancedQueryBuilder.QueryType.CREATED).\
    get_advanced_query()

To submit these queries pass them as a parameter to get_get_documents_advanced_query:

    response = client.get_documents_advanced_query(advanced_query)
    for document in response['documents']:
        print(document['name'], document['id'], document['lastModified'])

Retrieving document content

Content can be retrieved from the endpoint /documents/{id} where {id} is a documentID.

Here is an example retrieving a document in CSV format taken from forms.py script:

advanced_query = rspace_client.AdvancedQueryBuilder(operator='and').\
    add_term(form_id, rspace_client.AdvancedQueryBuilder.QueryType.FORM).\
    get_advanced_query()

response = client.get_documents_advanced_query(advanced_query)

print('Found answers:')
for document in response['documents']:
    print('Answer name:', document['name'])
    document_response = client.get_document_csv(document['id'])
    print(document_response)

Getting attached files

Here's an example where we download file attachments associated with some documents. The code is in download_attachments.py.

try:
    response = client.get_document(doc_id=document_id)
    for field in response['fields']:
        for file in field['files']:
            download_metadata_link = client.get_link_contents(file, 'self')
            filename = '/tmp/' + download_metadata_link['name']
            print('Downloading to file', filename)
            client.download_link_to_file(client.get_link(download_metadata_link, 'enclosure'), filename)
except ValueError:
    print('Document with id %s not found' % str(document_id))

Creating / editing a new document

A document can be created by sending a POST request to /documents. Document name, form from which the document is created, tags and field values can be specified. The example code is in create_document.py.

# Creating a new Basic document in Api Inbox folder
new_document = client.create_document(name='Python API Example Basic Document', tags=['Python', 'API', 'example'],
                                      fields=[{'content': 'Some example text'}])

You can also supply the parentFolderId of the workspace folder you want the document created in:

# Creating a new Basic document in specified folder:
new_document = client.create_document(name='Python API Example Basic Document', tags=['Python', 'API', 'example'],
                                      fields=[{'content': 'Some example text'}], parent_folder_id=21);

It is possible to edit a document by sending a PUT request to /documents/{id}, where {id} is a documentID. Document name, tags and field values can be edited.

# Editing the document to link to the uploaded file
client.update_document(document['id'], fields=[{'content': 'Edited example text.'}])

Uploading a file to gallery

Any file that can be uploaded by using the UI can be uploaded by sending a POST request to /files. Also, it is possible to link to the file from any document as shown in create_document.py example.

# Uploading a file to the gallery
with open('resources/2017-05-10_1670091041_CNVts.csv', 'rb') as f:
    new_file = client.upload_file(f, caption='some caption')

Linking to files and other RSpace documents

There is a convenient syntax to link to either files or other RSpace documents.

To include links to files in your document content, you can use the syntax <fileId=12345> where '12345' is the ID of a file uploaded through the files/ endpoint.

# Editing a document to link to an uploaded file
client.update_document(new_document['id'], fields=[{
    'content': 'Some example text. Link to the uploaded file: <fileId={}>'.format(new_file['id'])
}])

To include links to RSpace documents, folders or notebooks in your document content, you can use the syntax <docId=12345> where '12345' is the ID of an RSpace document,folder or notebook. E.g.

# Editing a document to link to another RSpace document
client.update_document(new_document['id'], fields=[{
    'content': 'Some example text. Link to another document: <docId={}>'.format(anotherDocument['id'])
}])

Activity

Access to the information that is available from the RSpace audit trail. This provides logged information on 'who did what, when’.

For example, to get all activity for a particular document:

response = client.get_activity(global_id=document_id)

print('Activities for document {}:'.format(document_id))
for activity in response['activities']:
    print(activity)

To get all activity related to documents being created or modified last week:

date_from = date.today() - timedelta(days=7)
response = client.get_activity(date_from=date_from, domains=['RECORD'], actions=['CREATE', 'WRITE'])

print('All activity related to documents being created or modified from {} to now:'.format(date_from.isoformat()))
for activity in response['activities']:
    print(activity)

Creating a Folder / Notebook

A folder can be created by sending a POST request to /folders. All arguments are optional. Name, parent folder id and whether to create a notebook can be specified. For example, to create a folder named 'Testing Folder', create_folder method can be used:

# Creating a folder named 'Testing Folder'
new_folder = client.create_folder('Testing Folder')

Notebooks can be created by setting notebook=True. To create a new notebook inside the previously created folder:

# Creating a notebook named 'Testing Notebook' inside the previously created folder:
new_notebook = client.create_folder('Testing Notebook', parent_folder_id=new_folder['globalId'], notebook=True)

There are some restrictions on where you can create folders and notebooks, which are required to maintain consistent behaviour with the web application.

You can't create folders or notebooks inside notebooks
You can't create notebooks inside the Shared Folder; create them in a User folder first, then share. (Sharing is not yet supported in the API, but you can do this in the web application).

Getting Information About a Folder / Notebook

Folder or notebook information can be retrieved by sending a GET request to /folders/{folderId} where folder id is a numerical ID of a folder or a notebook. Python client accepts both numerical IDs and global IDs. Method get_folder can be used to get information about a folder:

# Get information about a folder
folder_info = client.get_folder('FL123')  # or client.get_folder(123)
print(folder_info['globalId'], folder_info['name'])

Forms

Published forms can be listed by sending a GET request to /forms. The results might be paginated if there are too many forms (see create_form.py example for a more in depth usage example).

# Listing all published forms
response = client.get_forms()
for form in response['forms']:
    print(form['globalId'], form['name'])

A new form can be created by sending a POST request to /forms. Name, tags (optionally) and fields can be specified. Currently, supported types of form fields are: 'String', 'Text', 'Number', 'Radio', 'Date'. More information about the available parameters can be found in API documentation or by looking at create_form.py source code.

# Creating a new form
fields = [
    {
      "name": "A String Field",
      "type": "String",
      "defaultValue": "An optional default value"
    }
]
client.create_form('Test Form', tags=['testing', 'example'], fields=fields)

Form information can be retrieved by sending a GET request to /forms/{formId} where formId is a numerical ID of a form. Python client accepts both numerical IDs and global IDs.

# Getting information about a form
response = client.get_form('FM3')  # or client.get_form(3)
print('Retrieved information about a form:', response['globalId'], response['name'])
print('Fields:')
for field in response['fields']:
    print(field['type'], field['name'])

A newly created form is not available to create documents from until it has been published. Sending a POST request to /forms/{formId}/publish publishes a form.

# Publishing form FM123
client.publish_form('FM123')  # or client.publish_form(123)

# Unpublish the form
client.unpublish_form('FM123')  # or client.unpublish_form(123)

It is possible to share a form with your groups. Once it is shared the accessControl.groupPermissionType property is READ.

# Sharing form FM123
client.share_form('FM123')

# Unsharing the form
client.unshare_form('FM123')

Export

From RSpace 1.47 (API version 1.3) you can programmatically export your work in HTML or XML format. This might be useful if you want to make scheduled backups, for example. If you're an admin or PI you can export a particular user's work if you have permission.

Because export can be quite time-consuming, this is an asynchronous operation. On initial export you will receive a link to a job that you can query for progress updates. When the export has completed there will be a link to access the exported file - which may be very large.

This Python API client provides an easy to use method that handles starting an export, polling the job's status and downloading the exported archive once it's ready. For example, to export current user's work in XML format:

export_archive_file_path = client.download_export('xml', 'user', file_path='/tmp')

There are start_export(self, format, scope, id=None) and get_job_status(self, job_id) functions to start the export and check its status as well.

danielvarela/rspace-client-python