HttpStream

Overview

Currently we are working on providing to the user a way to create streams that interact with any valid pathname, including that of any schema. This implementation of the stream class will give support to Http pathnames, which will in turn allow the user to interact with http URIs.

It should be noted that our implementation of stream support of Http will be both concrete and of our own description. First, Http streams can only originate from IO, character streams; any attempt to create anything else, such as input, should generate an error.

The open, read, and write functions are also unique for HttpStreams. The process for dealing with Http streams occur in the following order:

• open
• write - [optional] this acts as a request-sender, as in it can direct to another html page or directory from the base given http pathname, e.g. (write-char "/docs/index/support.html")
• read - this acts as a request-returner, in that it returns html code of the given last write command. So from (read-line s) we would get the first line (up to a line break) of the html page.

From here the user can keep on opening more streams and continuing these steps. It should be noted that these functions must occur in this order for http streams, anything else should throw and error exception.

Example Usage

(open-stream-p s)

(file-write-data s)

(read-line s)

(close s)

Implementation

Methods of HttpStream

Discussions

Links to Blog issues

Current Status of HttpStream

Status: Implemented and working

Release Level:

Open bug count:

Test Suites

Links to JUnit results

-- JerryBoetje - 12 Jul 2003