# $Id: Changes,v 1.46 2003/09/12 01:17:45 petdance Exp $
Revision history for Perl extension HTML::Lint and Test::HTML::Lint.
1.24 September 11, 2003
[FIXES]
* Removed the check for "input tags can have height &
width", because they can't.
1.23 September 02, 2003
[ENHANCEMENTS]
* The types of errors to find may now be passed in the
HTML::Lint constructor.
* Can now export the error types from HTML::Lint::Error.
[DOCUMENTATION]
* Fixed a bug in Test::HTML::Lint docs. (Thanks to Leon
Brocard and about 90 other people)
1.22 JUNE 11, 2003
[ENHANCEMENTS]
* Added some docs explaining html_ok()'s optional lint object.
* Allowed the NAME attribute in the <IMG> tag.
[FIXES]
* Fixed the version check in t/99.pod.t
1.21 October 9, 2002
[ENHANCEMENTS]
* Added a test file for pod files
* html_ok() now diag()s the test name before spewing the errors.
* Made some docs a little more explicit. (Thanks to Ron Savage)
* Added the check for missing <HTML>, <HEAD>, <TITLE> and <BODY>.
* Added the check for nonrepeatable tags.
[FIXES]
* Ticket #1493: typo in HTML::Lint::HTML4
<ADDRESS> element in HTML4.pm was <ADDRESSS>. Thanks to Dominic
Mitchell for finding it.
1.20 August 22, 2002
[ENHANCEMENTS]
* Made it run under 5.5.3
1.13 August 5, 2002
[ENHANCEMENTS]
* Added text-use-entity for detecting control characters.
1.12 August 2, 2002
[FIXES]
* No longer squawks on html_ok( undef ). It now specifically fails.
1.11 July 25, 2002
[ENHANCEMENTS]
* Added the concept of HTML::Lint::Error type, which is one of
Structure, Helper or Fluff.
* weblint now takes --[no]structure, --[no]helper, --[no]fluff.
* Added the ability to overload html_ok()'s HTML::Lint object
with one of your own.
[FIXES]
* Fixed line number goof in Test::HTML::Lint
* Fixed line positions in weblint on files (URL reading was OK)
1.10 July 17, 2002
* Added Test::HTML::Lint
* Removed the "use 5.6.0" from everywhere
* Included Neil Bowers' original weblint script, in the
weblint-original/ directory. Note that it does NOT get
installed.
1.02 July 8, 2002
* No functional changes. It's all in the test suite.
* Added a whole mess of *.t files. They've been in CVS all along,
but I forgot to put them in the MANIFEST. Ooops.
1.01 July 3, 2002
* Tests use isa_ok() for more stringent checking of return objects
* Added <nobr>. Ooopsie.
1.00 June 5, 2002
* weblint has a --context option to show the line that the error
occurs on.
* Removed the original weblint log from this Changes file.
* Fixed: Unknown tags wouldn't get put into the stack, so you
would get two errors: One complaining that it didn't know the
tag, and another that the closing tag didn't make sense.
0.94 May 31, 2002
* Moved the %HTML::Lint::Error::errors hash to be a package
variable, and not initialize it in an INIT block.
0.93 May 28, 2002
* weblint can now read URLs as well as files
* Improved the docs in HTML::Lint::Error.
0.92 February 26, 2001
* Removed the INIT block in HTML::Error in preparation for
Apache::Lint, and so brian's weblint++ can use it.
* errors() method respects wantarray (brian d foy)
* file(), line(), column(), errcode() and errtext() all return
blank instead of undef.
0.91 January 8, 2002
* Fixed: Content description tags (<CITE>, <EM>, etc) were
not seen as being valid tags.
0.90 July 8, 2001
First version of the rewrite as a subclass of HTML::Parser
0.03 May 15, 2001
* Fixed t/*.t problems
0.02 May 14, 2001
* Packaged and bundled with new namespace
* First upload to CPAN
0.01 April 20, 2001
* original version; created by h2xs 1.21 with options
-n HTML::Lint -X -A
* adapted entirely from Neil Bowers' Weblint package