REST APIs Lecture BigData Analytics Julian M. Kunkel julian.kunkel@googlemail.com University of Hamburg / German Climate Computing Center (DKRZ) 2016-12-09 Disclaimer: Big Data software is constantly updated, code samples may be outdated. REST APIs Accessing RESTful Services Outline 1 REST APIs 2 Accessing RESTful Services Julian M. Kunkel Lecture BigData Analytics, 2016 2 / 11 REST APIs Accessing RESTful Services REST [11] Representational state transfer (REST): System API utilizing HTTP / TCP RESTful: Term indicates the system is conforming to REST constraints Advantages (due to HTTP) Simplicity of the interfaces Portability Cachable Tracable: Communication can be inspected Semantics of HTTP request verbs [13] GET: retrieve a representation of a resource (no updates) PUT: store the enclosed data under the given URI POST: transfer an entity/data as subordinate of the web resource DELETE: remove the given URI PUT and DELETE are idempotent and GET (w/o concurrent updates) Julian M. Kunkel Lecture BigData Analytics, 2016 3 / 11 REST APIs Accessing RESTful Services HTTP 1.1 [13] The Hypertext Transfer Protocol (HTTP) is a stateless protocol Request via TCP ⇒ Response (status and content) Request/Response are encoded in ASCII Include a header with standardized key/value pairs [14] Non-standard key/value pairs can be added Usually prefixed with X for eXtension One data section (at the end) according to the media type Separation between header and data via one newline Example HTTP Request 1 2 3 4 5 GET /dir/file HTTP/1.1 Host: www.test.de:50070 User-Agent: mozilla Cache-Control: no-cache Accept: */* Julian M. Kunkel Lecture BigData Analytics, 2016 4 / 11 REST APIs Accessing RESTful Services HTTP 1.1 Media types [15] Based on Multipurpose Internet Mail Extensions (MIME) types Media type is composed of type, subtype and optional parameters e.g., image/png e.g., text/html; charset=UTF-8 Media types should be registered by the IANA1 Example

Lire le document entier