Skip to content

WebServiceAPI

Jump to bottom Edit New page
Bert Frees edited this page Dec 15, 2017 · 4 revisions

Web Service API

This document outlines the web service API for the DAISY Pipeline.

Resources

RESTful web services describe a system in terms of its resources. Generally speaking, resources can be created, retrieved, modified, and deleted.

The main resources in this web service are scripts, jobs, and clients.

A script is a Pipeline XProc script. The script is identified with a unique URI. It is described in terms of its functionality and all of its inputs and options.

A job is the execution of a script. Jobs have unique IDs and are created using input and option values. Jobs produce downloadable results.

A client is an application using the Pipeline via its API. Examples of clients are web applications and command line utilities. A client is not a human user.

Accessing Resources

RESTful applications communicate via HTTP and use standard methods and status codes to create, get, and delete resources.

The following HTTP methods are used here:

  • GET
  • POST
  • PUT
  • DELETE

Every method is not implemented for every resource because in some cases, the method wouldn't be relevant. Any unimplemented methods will respond with HTTP 405 "Method Not Allowed".

The API

WS Status

  • HTTP Method: GET
  • URI: /alive
  • Response(s):

Scripts

Get all scripts

  • HTTP Method: GET
  • URI: /scripts
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains XML data. See a sample response document.
    • HTTP 401 Unauthorized: Client was not authorized to perform request.

Get a single script

  • HTTP Method: GET
  • URI: /scripts/$ID
    • Where $ID is the script's ID
  • Query parameters
  • Response(s):
    • HTTP 200 OK: Response body contains XML data. See a sample response document.
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Jobs

Create a job

  • HTTP Method: POST
  • URI: /jobs
  • Query parameters:
  • Request body:
  • Response(s):
    • HTTP 201 Created: Response body contains XML data about the new resource. See a sample document.
    • HTTP 400 Bad Request: The request was invalid
    • HTTP 401 Unauthorized: Client was not authorized to perform request.

Read more about creating jobs.

Get all jobs

  • HTTP Method: GET
  • URI: /jobs
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains XML data. See a sample response document.
    • HTTP 401 Unauthorized: Client was not authorized to perform request.

Get a single job

  • HTTP Method: GET
  • URI: /jobs/$ID
    • Where $ID is the job's unique ID
  • Query parameters:
    • msgSeq (optional): Returned job status messages will be a subset of messages where the sequence values are greater than that provided. The default is 0. This parameter allows incremental reporting on job status; i.e. "only return messages newer than X".
    • See Authentication
  • Response(s):
    • HTTP 200 OK: Response body contains XML data. See a sample response document.
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Delete a single job

  • HTTP Method: DELETE
  • URI: /jobs/$ID
    • Where $ID is the job's unique ID
  • Query parameters:
  • Response(s):
    • HTTP 204 No Content: Successfully processed the request, no content being returned
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Get the result for a job

  • HTTP Method: GET
  • URI: /jobs/$ID/result[/$TYPE/$NAME[/idx/$IDX]]
    • Where $ID is the job's unique ID
    • $TYPE is whether port or option
    • $NAME is the port or option name
    • $IDX is the index of a single output within the port/option.
    • An example of a result description in the job xml format.
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains Zip data
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Get the log file for a job

  • HTTP Method: GET
  • URI: /jobs/$ID/log
    • Where $ID is the job's unique ID
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains a plain text representation of the log file
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Admin

In addition to managing scripts and jobs, the web service also provides an administrative interface. Use of this API is restricted to use by authorized clients; i.e. those with ADMIN permissions.

Stop the web service

  • HTTP Method: GET
  • URI: /admin/halt/$KEY
    • Where $KEY is generated by the web service upon startup and stored locally as a text file.
  • Query parameters:
  • Response(s):
    • HTTP 204 No Content: Successfully processed the request, no content being returned
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 403 Forbidden: The service could not be halted. This error would be caused by an invalid key.

Get all clients

  • HTTP Method: GET
  • URI: /admin/clients
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains XML data. See a sample response document.
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Create a new client

  • HTTP Method: POST
  • URI: /admin/clients
  • Query parameters:
  • Request body:
  • Response(s):
    • HTTP 201 Created: Response body contains XML data about the new resource. See a sample document.
    • HTTP 400 Bad Request: The request was invalid
    • HTTP 401 Unauthorized: Client was not authorized to perform request.

Get a single client

  • HTTP Method: GET
  • URI: /admin/clients/$ID
    • Where $ID is the client's unique ID
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains XML data. See a sample response document.
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Delete a client

  • HTTP Method: DELETE
  • URI: /admin/clients/$ID
    • Where $ID is the client's unique ID
  • Query parameters:
  • Response(s):
    • HTTP 204 No Content: Successfully processed the request, no content being returned
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Modify a client

  • HTTP Method: PUT
  • URI: /admin/clients/$ID
    • Where $ID is the client's unique ID
  • Query parameters:
  • Response(s):
    • HTTP 200 OK: Response body contains XML data about the modified resource. See a sample response document.
    • HTTP 400 Bad Request: The request was invalid
    • HTTP 401 Unauthorized: Client was not authorized to perform request.
    • HTTP 404 Not Found: Resource not found

Schema

Schema for all the XML formats used in this API can be found here.

Authentication

When authentication is enabled, the Pipeline only accepts requests from authorized clients. In order to become authorized, a client must be added to the database of clients using the admin API described in [#Admin].

The client must add the following query parameters to each request:

  • authid: Identifies the client application
  • time: Timestamp (UTC)
  • nonce: A unique string generated by the client
  • sign: Hash of the URI string including all query params except this one

Note that it is possible to start the web service with authentication disabled. This is appropriate when the web service is running only locally or in other similarly controlled environments. In this case, none of the authentication parameters are required, and any that are sent are ignored.

Read more about authentication.

Local File System mode

The web service is by default run in what is referred to as local FS mode. The property org.daisy.pipeline.ws.localfs must be set to change it. In local mode, the web service assumes read/write access to the disk. Documents that would normally be sent as part of a zip archive or inline request can be referred to by their local file path instead. Output options are enabled and output is written to disk instead of being zipped up for download.

See a sample document.

Web Service internal errors

In the unlikely but not impossible case that the Pipeline WS suffered from an internal error, a descriptive xml is returned along with the HTTP 500 header.

See the xml sample document.

Add a custom footer
Clone this wiki locally