Skip to main content

8. HTTP Servers and APIs

03/03/23

MoodlePDF

HTTP Servers​

  • HTTP servers normally run on well-defined ports
  • http (80), https(443)
  • So all TCP connection requests to a host will (initially) arrive at the same process
  • So TCP port numbers (layer 4) are NOT normally useful for distinguishing between services using HTTP on a single host
  • Each server will require typically explicit configuration

Routing Requests​

  • Server can handle each request differently depending on its:
  • URL domain (host) name
  • URL path
  • Identified file type/extension
  • Client IP address
  • HTTP Headers

Handling Requests​

  • Static file serving - the web server uses the URL path to identify a file which is copied back as the response body
  • CGI - the server executes a program (identified by the URL) in a separate process to handle the request
  • Dynamically loadable module - the server loads a runtime engine to execute a file in a scripting language
  • Reverse proxy - the original server becomes an HTTP client to pass the request onto another web server
  • Custom web server written using a language-specific web server framework - the web server includes the code to handle specific requests

HTTP APIs​

  • HTTP is often used as a standard transport mechanism for program-to-program interaction
  • Normally defined in terms of permitted combinations of
    • URL -> services and operation or resource (REST)
    • HTTP method
    • Authentication requirements
    • Request data
    • Response data
    • Error responses

Defining HTTP APIs​

  • Usually language-independent
  • Can be defined; manual documentation, automatically from the server implementation, machine-readable interface definition/specification

Types of HTTP API​

  • Form-/query-style
  • REST (REpresentational State Transfer)
  • RPC (Remote Procedure Call)
  • Can be a mixture of both

HTML Form- or Query-Style Operations​

  • URL defines the server and operation (path)
  • HTTP method is GET (normal navigation or data requests) or POST (operations with side effects)
  • Arguments are a list of names and text variables
    • In the URL query (GET)
    • In the request body (form POST), encoded either
  • Response body is machine readable results
  • Errors may be signalled by HTTP status and/or response body

REST - Representational State Transfer​

  • URL identifies a specific resource (database record)
  • HTTP method identifies operation to perform
  • For GET the response body (result) is an external data representation of the identified resource
  • For PUT/POST the only argument - the request body is also a new representation of the identified resource
  • Errors are signalled by HTTP status whenever possible

Remote Procedure Call (RPC)​

  • URL typically identifies a specific service
  • HTTP method is normally POST
  • Body of request encodes a list of arguments, and a specific operation name is not specified by URL
  • Response body encodes status and results of operation
  • Errors normally signalled within the response body