Revision history for Perl extension PadWalker.

0.01  Thu Nov  9 12:58:10 2000
	- original version; created by h2xs 1.19

Revision history between 0.01 and 0.03 has been lost in the mists
of time. Sorry about that.

0.03 was the first public release.

0.04  Thu Jul 19 13:50:19 BST 2001
	- Applied patch from Richard Soderberg to let it compile
	  under ithreads

0.05  Thu Jan 10 21:12:10 GMT 2002
	- Experimental peek_sub routine

0.06  Wed Mar  6 22:16:13 GMT 2002
	- Proper test script
	- Two bug fixes

0.07  Thu Mar 14 19:56:29 GMT 2002
	- Clean up POD documentation
	- Behave properly with debugger
	- Work under Perl 5.005

0.08  Mon Mar 18 17:54:16 GMT 2002
	- _upcontext() XSUB, for Richard Clamp
	- work properly with recursion
	- see past eval

0.09  Wed May 21 17:24:32 BST 2003
        - compile on threaded builds
        - skip test 9 if we're on Perl 5.8. (Otherwise it fails.)

0.10  Wed Jul 30 18:40:03 BST 2003
	- accommodate the new versions of perl (5.8.1)
	- acknowledge that test 9 fails because nested eval handling
          is simply broken. (Fix in a future release.)

0.11  Wed Aug 10 20:06:41 BST 2005 
	- fix various nasty bugs, specifically:
	 o peek_my(1) now works correctly in a sub called from the top level;
	 o deal better with sub calls across source files
	- don't return 'our' variables. (It is peek_my(), after all!
	  And the values were never properly returned.)

0.12  Tue Aug 16 15:40:04 BST 2005
	- make it work (up to a point) with Perl 5.6.

0.13  Mon Oct  3 11:54:23 BST 2005
	- don't build a debugging build unless explicitly told to!

0.14  Thu Oct  6 17:19:06 BST 2005
	- Fix the bugs reported by Dave Mitchell:
	 o if one variable masks another, make sure we return the
	   appropriate one;
	 o for a variable whose value has been lost, return undef
	   rather than the name of the variable;
	 o Don't die if PadWalker is called from a closure whose
           containing scope has disappeared.

0.99  Fri Oct  7 17:23:09 BST 2005
	- Make peek_sub return the values, if possible,
 	  even when it's not in the call chain;
	- Allow an our variable to mask a my variable,
	  and vice versa;
	- Add peek_our and closed_over routines.

0.99_91  Thu Oct 13 17:35:11 BST 2005
	- Make peek_my work correctly when called from a do "file";
	- Add var_name routine;
	- Add an :all EXPORT_TAG;
	- Drop support for perl < 5.8;
	- Eliminate compiler warnings (at least on gcc).

0.99_92  Sat Oct 15 00:17:24 BST 2005
	- Add license info to META.yml;
	- Use tabs consistently in the Changes file...
	- Don't use snprintf (apparently Bill Gates doesn't approve);
	- ignore 'our' variables in closed_over();
	- Eliminate more compiler warnings, this time on Win32.

0.99_93  Fri Oct 28 13:18:20 BST 2005
	- Change the sekrit undocumented second return value
	  of closed_over() to something slightly different.

1.0  Wed Nov  2 12:25:49 GMT 2005
	- Bump version number to 1.0

1.1  Sun Oct 22 16:13:40 BST 2006
	- Accommodate change 27312 "Store the stash for our in magic slot"
	  (See http://www.mail-archive.com/perl5-changes@perl.org/msg14073.html
	  or http://public.activestate.com/cgi-bin/perlbrowse/27312)
	
	  See also http://public.activestate.com/cgi-bin/perlbrowse/27306

1.2  Thu Nov 16 22:33:27 GMT 2006
	- Change prerequisites to accurately reflect versions of Perl that
	  PadWalker actually works with (i.e. 5.8.2 or later).
	- Fix memory leak: thanks to Rocco Caputo

1.3  Tue Jan  2 23:10:35 GMT 2007
	- Accommodate changes 29629-29630 "Move the low/high cop sequences
	  from NVX/IVX to a two U32 structure".

1.4  Fri Jan  5 09:12:11 GMT 2007
	- Accommodate change 29679 "Rename OURSTASH to SvOURSTASH and
	  OURSTASH_set to SvOURSTASH_set". (Dear Nick, please stop
	  breaking PadWalker. kthxbye.)

1.5  Fri Jan  5 16:22:27 GMT 2007
	- Fix egregrious bug in 1.4 :-(

1.6  Mon Jan 14 10:48:09 GMT 2008
	- Make _upcontext work in 64-bit architectures.
	  (http://rt.cpan.org/Ticket/Display.html?id=32287)
	  Thanks to Niko Tyni.