20140709 Wed Jul 9 16:28:37 PDT 2014 New Features * The "git" scheme is supported. (Schwern) * svn, ssh and svn+ssh schemes are supported. [rt.cpan.org 57490] (Schwern) * Added a --schemeless option to urifind. (Schwern) Bug Fixes * http:// is no longer matched [rt.cpan.org 63283] (Schwern) Backwards Incompatibilities * Previously, URIs stringified to their canonical version. Now they stringify as written. This results in less loss of information. For example. "Blah HTTP:://FOO.COM" previously would stringify as "http://foo.com/" and now it will stringify as "HTTP://FOO.COM". To restore the old behavior you can call $uri->canonical. (Schwern) Distribution Changes * No longer using URI::URL. (Schwern) * Now requires URI 1.60 for Unicode support. (Schwern) 20140702 Wed Jul 2 13:41:47 PDT 2014 New Features * IDNA (aka Unicode) domains are now supported. [github 3] (GwenDragon) * The list of TLDs for schemeless matching has been updated. [github 3] (GwenDragon) Bug Fixes * Handle balanced [], {} and quotes in addition to (). [rt.cpan.org 85053] (Schwern) * Don't mangle IPv6 URLs. [rt.cpan.org 85053] (Schwern) * Schemeless is more accurate about two letter TLDs. [github 3] (GwenDragon) Distribution Changes * Switched the issue tracker to Github. (Schwern) 20111103 Thu Nov 3 12:14:21 PDT 2011 Bug Fixes * URI::URL::strict will no longer leak out of find() if the callback or filter fails. [rt.cpan.org 71153] (Carl Chambers) 20111020 Thu Oct 20 17:31:56 PDT 2011 Bug Fixes * Things which look like URIs, but aren't, are now properly escaped like other text. [rt.cpan.org 71658] New Features * Balanced parens in URIs are no longer stripped. Example: "http://example.com/foo(bar)" (Merten Falk) 20100505 Wed May 5 18:48:44 PDT 2010 Test Fixes * Fixed t/urifind/find.t on Windows 20100504.1039 Tue May 4 10:39:23 PDT 2010 Doc Fixes * Forgot to mention that we ship with urifind now. 20100504 Tue May 4 10:29:52 PDT 2010 New Features * Added a urifind program. (Darren Chamberlain) Bug Fixes * The final semi-colon was being strippped form URLs found in HTML that ended with HTML entities. (Michael Peters) Example: http://google.com/search?q=<html> * URLs with leading dots, pluses and minuses are now found. [rt.cpan.org 57032] Example: stuff...http://example.com 20100211 Thu Feb 11 04:02:26 PST 2010 Bug Fixes * Finding URIs inside brackets was pretty badly broken by the last release. (Michael Peters) 20090319 Thu Mar 19 12:17:53 PDT 2009 Bug Fixes * Schemeless now ignores the case of the TLD. New Features * Updated the list of accepted domains for finding schemeless URIs from the latest ICANN list. Docs * Add LICENSE section * Remove wildly out of date CAVEATS * Added an example of how to get a list of all URIs. * Updated INSTALL section to reflect new dependencies and Module::Build installation process * Regenerated the README file 20090316 Mon Mar 16 16:18:10 PDT 2009 New Features * Added optional replacement function to find(). Now you can not only replace URLs found, but also the rest of the text around them in one fell swoop. (Mike Schilli) [rt.cpan.org 20486] * Whitespace inside <...> is now ignored as per the suggestion of RFC 3986 appendix C. [rt.cpan.org 20483] Other * Michael G Schwern is now primary maintainer again. Thanks for all your work, Roderick! * Repository moved to http://github.com/schwern/uri-find * Now requires Test::More * Verisoning scheme changed to ISO date integers * Minimum Perl version is now 5.6.0. 0.16 Fri Jul 22 06:00:24 EDT 2005 - Oops, make the URI::Find::Schemeless->top_level_domain_re case insensitive, as it should be and the docs claimed it was. Thanks to Todd Eigenschink. 0.15 Tue Mar 22 07:23:17 EST 2005 - Have all functions croak if invoked with the wrong number of arguments. Add URI::Find->badinvo. https://rt.cpan.org/NoAuth/Bug.html?id=1845 - Mention DARREN's urifind script in the man page. - Oops, URI::URL::strict was turned on and left on. Put it back the way you found it. Thanks to Chris Nandor. https://rt.cpan.org/NoAuth/Bug.html?id=11906 - Schemeless.pm: - Find '<x>intag.com</x>'. - Get $tldRe from a new class method, ->top_level_domain_re. - Update top level domain list. 0.14 Sat Oct 9 08:20:04 EDT 2004 - Add copyright notice. - Add ] to main $cruftSet, } to schemeless $cruftSet, for [http://square.com] and {brace.com}. - quotemeta() $cruftSet. 0.13 Mon Jul 1 10:37:54 EDT 2002 - Don't find any schemeless URIs with a plain URI::Find. Previously it'd find ones which started with "ftp." and "www.", but it was more prone to false positives than URI::Find::Schemeless. - Have schemeless_to_schemed use http:// except in the specific case in which it uses ftp://. Remove URI::Find::Schemeless's version. 0.12 Wed Mar 20 14:39:21 EST 2002 - Improve the "wrap each URI found in an HTML anchor" example. - Release a new version so CPAN sees the maintainer change. 0.11 Thu Jul 26 14:43:49 EDT 2001 - Michael passed the module to Roderick for maintenance. - Improve test suite. - Tweak URI::Find::Schemeless not to find Foo.p[ml]. 0.10 Mon Jul 10 20:14:08 EDT 2000 - Rearchitected the internals to allow simple subclassing - Added URI::Find::Schemeless (thanks Roderick) 0.04 Sat Feb 26 09:05:11 GMT 2000 - Added # to the uric set of characters so HTML anchors are caught. 0.03 Tue Feb 1 16:15:22 EST 2000 - Added some heuristic discussion to the docs. - Added some heuristics to avoid picking up perl module names - Improved schemeless URI heuristic to avoid picking up usenet board names. - Handling the <URL:...> case better as suggested in RFC 2396 Apdx E - Added ; to the cruft heuristic 0.02 Tue Feb 1 13:11:56 EST 2000 - Added heuristic to handle 'URL:http://www.foo.com' - Added heuristic to handle trailing quotes. 0.01 Mon Jan 31 19:12:23 EST 2000 - First working version released to CPAN.