ENH melt uses column name if available

[hayd]

Currently melt allows you to manually specify the val_name and var_name, this changes the default behaviour to grab the column name if one is available.

In [1]: df = pd.DataFrame({'a': {0: 1.0676829999999999, 1: -1.3214049999999999, 2: -0.80733299999999997}, 'b': {0: -1.110463, 1: 0.36891499999999999, 2: 0.082979999999999998}, 'c': {0: 0.20866999999999999, 1: -1.055342, 2: -0.87336100000000005}})

In [2]: df.columns.name = 'foo'

In [3]: df
Out[3]:
foo         a         b         c
0    1.067683 -1.110463  0.208670
1   -1.321405  0.368915 -1.055342
2   -0.807333  0.082980 -0.873361

In [4]: pd.melt(df)  # before this would have columns ['variable', 'value']
Out[4]:
  foo     value
0   a  1.067683
1   a -1.321405
2   a -0.807333
3   b -1.110463
4   b  0.368915
5   b  0.082980
6   c  0.208670
7   c -1.055342
8   c -0.873361

Also, cleaned up all the tests (and added some more) in test_reshape.

[jreback]

isn't this already in with #3649, or this is just a cleanup of that?

[jreback]

oh i c, you name is more 'automatic'...nice!

[jreback]

@hayd there is a release notes/whatsnew associated with 3649, why don't you add on to it with this change?

[hayd]

I was just pondering whether I needed to bother :) will do

[hayd]

As an aside, wonder if melt should work with multiindexes...

[cpcloud]

u could do frame.columns.name or 'variable' too :)

[hayd]

...but just in case they had named their column False(y) :)

[hayd]

merge? (Have another WIP for MultiIndexes)

[jreback]

look good