SBECK / Math-SigFigs-1.20 / Changes 
Written by:
   Sullivan Beck (sbeck@cpan.org)
Any suggestions, bug reports, or donations :-) should be sent to me.

1.00  1996-12-05
  - Initial creation

1.01  1997-01-28
  - Used croak and changed die's to confess.
  - "101" is now returned as "101." .
  - Fixed where 9.99 wasn't being correctly returned with 1 sigfig.
    Kyle Krom

1.02  2000-01-10
  - Fixed where 1249.01 wasn't correctly rounded to 1200.
    Janna Wemekamp

1.03  2003-09-11
  - Fixed a bug where I left off the sign.  Steve Reaser
  - Fixed a bug in subSF where numbers ending in zero were truncated.
    Andrew Grall

1.04  2005-06-30
  - Complete rewrite of addSF.
     * stopped using sprintf (which does not return the same results on
       all platforms.
     * replaced IsReal with Simplify.

1.05  2007-02-22
  - Fixed a bug where subSF didn't work with some values.

1.06  2008-05-05
  - Updated test.pl.
  - Added Build.PL and several other things to meet requirements on
    http://cpants.perl.org/

1.07  2008-05-08
  - Additional changes to meet requirements on http://cpants.perl.org/

1.08  2008-05-09
  - Final changes to meet requirements on http://cpants.perl.org/
  - Bug fix so it won't fail with "Too early to specify a build action"

1.09  2009-09-24
  - FormatSigFigs will preserve a '+' sign.  Denby Angus
  - Better handling a value of 0.  Denby Angus

1.10  2015-07-10
  - Switched test suite to Test::Inter
  - Changed HISTORY file to new Changes specification.
  - Put repository on GitHub.  Gabor Szabo
  - Improved POD documentation formatting.
  - Improved test suite (and made a number of simplifications to the module)
    based on Devel::Cover.  Test suite now has 100% coverage.

1.20  2016-10-03
  - Added some examples to the docs that illustrate trailing zero handling.
    Christopher King
  - Change addSF/subSF to return undef if any non-numeric argument is passed in.
  - Corrected how 0 (and 0.00, etc.) significant figures are counted and used
  - Now supports exponentials
  - Fixed a bug where subtracting numbers very close to each other gave
    incorrect results.  Richard Spinney
  - Return numbers in their simplest form (without trailing decimal points
    when they are not required)

1.21