groupby().apply is too opinionated

[ghost]

apply looks at the returned values from the user-supplied transformer function
and does convenient things based on the type returned.

However, there's no opt-out if you just want to transform groups
into something and get it back as-is:

In [16]: df=mkdf(10,2,data_gen_f=lambda x,y: randint(1,10))
    ...: df
    ...: 
    ...: 
    ...: 
Out[16]: 
C0       C_l0_g0  C_l0_g1
R0                       
R_l0_g0        9        1
R_l0_g1        3        7
R_l0_g2        8        1
R_l0_g3        4        3
R_l0_g4        5        3
R_l0_g5        7        2
R_l0_g6        4        1
R_l0_g7        5        4
R_l0_g8        9        7
R_l0_g9        4        8

In [17]: def f1(g):
    ...:     return g.sort('C_l0_g0')
    ...: # group on the suffix of the running index 
    ...: g=df.groupby(lambda key: int(key.split("g")[-1]) >= 5)
    ...: r=g.apply(f1)
    ...: 

# we want to get a bunch of sorted dataframes, but get concat-ed against our will
In [18]: r
Out[18]: 
C0             C_l0_g0  C_l0_g1
      R0                       
False R_l0_g1        3        7
      R_l0_g3        4        3
      R_l0_g4        5        3
      R_l0_g2        8        1
      R_l0_g0        9        1
True  R_l0_g6        4        1
      R_l0_g9        4        8
      R_l0_g7        5        4
      R_l0_g5        7        2
      R_l0_g8        9        7

More failed guessing of what the user wants: #5908.

[jreback]

could add a raw keyword here to avoid wrapping the output

[jreback]

not sure what this is actually about. Lots of ways to do this anyhow.