Do not generate incorrect SQL in presence of -and=>[]

[fabzzap]

An empty array ref as value for -and can cause incorrect SQL. Example:

$ perl -Mv5.12 -MSQL::Abstract -e 'my ($sql,@bind)=SQL::Abstract->new->where({-and=>[],-or=>[a=>1]});say $sql'
Use of uninitialized value in join or string at /usr/share/perl5/SQL/Abstract.pm line 1482.
 WHERE ( (  AND a = ? ) )

This is a proposed fix to that bug

[codecov-io]

Codecov Report

Merging #15 into master will increase coverage by 0.12%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #15      +/-   ##
==========================================
+ Coverage   83.06%   83.18%   +0.12%     
==========================================
  Files           3        3              
  Lines         927      928       +1     
  Branches      189      189              
==========================================
+ Hits          770      772       +2     
+ Misses        107      106       -1     
  Partials       50       50

Impacted Files	Coverage Δ
lib/SQL/Abstract.pm	82.47% <100%> (+0.18%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4805666...3e759b1. Read the comment docs.

[coveralls]

Pull Request Test Coverage Report for Build 28

• 4 of 4 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-0.2%) to 88.414%

---

Totals
Change from base Build 26:	-0.2%
Covered Lines:	786
Relevant Lines:	889

---

💛 - Coveralls

[mohawk2]

Surely there should be tests? And rather than just silently dropping things, this should cause an exception?

[fabzzap]

I just pushed a test for the patch. I don't know what "silently drop[s] things" here. -and=>[] is supposed to have no effect, but the main code instead generates a warning and incorrect SQL. With the patch, it behaves correctly: it does nothing.