HTTP Fetcher is a small, robust, flexible library for downloading files via HTTP using the GET method. It's easy to use, but it also allows you to customize and manipulate your file requests through altering the User Agent, Referer, timeout, etc. The error reporting functions give you a simple, clean interface through which to obtain information about a problem. HTTP Fetcher is designed to let you retrieve files via HTTP quickly and easily without having to reinvent the wheel. From this base you can build interactive download tools (like fetch), site mirroring tools, Web spiders, or anything with a need to do HTTP GETs.
|Operating Systems||POSIX Linux|
Release Notes: This version adds the ability to (transparently) follow redirects, as well as a function to allow callers to control the number of redirects to follow (if any). It also includes some minor portability improvements and some cleanups.
Release Notes: HTTP control structures now use RFC compliant line breaks. Support has been added for connecting to arbitrary ports. Server response values with leading 0's are now handled correctly. A timeout bug has been fixed and a memory leak plugged.
Release Notes: A buffer overflow and a memory leak have been fixed. There are portability improvements and build process checks for necessary functions. Included are a spec file for building RPM packages and an http-fetcher.m4 macro file for inclusion with autoconf. It checks for hstrerror and provides a workaround if not present.
Release Notes: A new "Host: " field in HTTP requests to accommodate some servers that require it (even with HTTP 1.0 requests); servers would previously return a 404 without the Host specified.