A new --suppress option allows response code/URL combinations to not be reported as problems. Checkbot warnings are now handled as pseudo-HTTP status messages. There is more robust handling of (lack of) status messages, and documentation fixes.
This release tries to produce valid XHTML 1.1. URLs matching the --ignore option are now completely ignored instead of being checked and then not reported on, proxy support works again (for all links, not just external ones), and some documentation fixes have been applied.
URLs with query strings are now checked by default, the server results page contains shortcut links to each section, the warning for unqualified hostnames for news: URLs has been removed, signals like SIGINT are handled, and bugs and documentation fixes have been applied.
A new --filter option allows rewriting of URLs. Problematic links are now reported for each occurrence, not just the first one. Statistics should now be working. Simplified storage of problematic links should avoid some obscure bugs. Link handling has been rewritten, therefore internal and external links are now checked at the same pass. Small code cleanups have been made.
The Makefile and packaging were improved. A better default for --match is generated. Additional fallbacks to GET are used when HEAD fails on servers. Bugs in the printing of Web server feedback have been fixed.
