[FEA] update() method is missing

[vivanov10]

Is your feature request related to a problem? Please describe.

update() method is missing

Describe the solution you'd like

Implement the update() method as in pandas

Describe alternatives you've considered

cuDF implementation of update() I am using:

def update(df, other, join="left", overwrite=True, errors="ignore"):
    # TODO: Support other joins
    if join != "left":  # pragma: no cover
        raise NotImplementedError("Only left join is supported")
    if errors not in ["ignore", "raise"]:
        raise ValueError("The parameter errors must be either 'ignore' or 'raise'")

    other = other.reindex(df.index, axis=0)
    other = other.reindex(df.columns, axis=1)

    for col in df.columns:
        this = df[col]
        that = other[col]

        if errors == "raise":
            mask_this = that.notna()
            mask_that = this.notna()
            if any(mask_this & mask_that):
                raise ValueError("Data overlaps.")

        if overwrite:
            mask = that.isna()
        else:
            mask = this.notna()

        # don't overwrite columns unnecessarily
        if mask.all():
            continue

        df.loc[mask, col] = this[mask]
        df.loc[~mask, col] = that[~mask]

cuDF example:

import cudf as pd

data = pd.DataFrame(
   data=([[1, 10, 100, 1000, 10000], [2, 20, 200, 2000, 20000]]), columns=['a', 'b', 'c', 'd', 'e']
)

data1 = pd.DataFrame(
   data=([[3, 33], [2, 22]]), columns=['a', 'b']
)
update(data, data1)
data

Additional context

This is how it works in pandas:

import pandas as pd

data = pd.DataFrame(
   data=([[1, 10, 100, 1000, 10000], [2, 20, 200, 2000, 20000]]), columns=['a', 'b', 'c', 'd', 'e']
)

data1 = pd.DataFrame(
   data=([[3, 33], [2, 22]]), columns=['a', 'b']
)
data.update(data1)
data

While it raises an exception in cuDF:

import cudf as pd

data = pd.DataFrame(
   data=([[1, 10, 100, 1000, 10000], [2, 20, 200, 2000, 20000]]), columns=['a', 'b', 'c', 'd', 'e']
)

data1 = pd.DataFrame(
   data=([[3, 33], [2, 22]]), columns=['a', 'b']
)
data.update(data1) # no attribute

Traceback:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-23-deb292f322d6> in <module>
      9    data=([[3, 33], [2, 22]]), columns=['a', 'b']
     10 )
---> 11 data.update(data1) # no attribute

~/miniconda3/envs/rapids14/lib/python3.7/site-packages/cudf/core/dataframe.py in __getattr__(self, key)
    430                 return self[key]
    431 
--> 432         raise AttributeError("'DataFrame' object has no attribute %r" % key)
    433 
    434     @annotate("DATAFRAME_GETITEM", color="blue", domain="cudf_python")

AttributeError: 'DataFrame' object has no attribute 'update'

[kkraus14]

Pushing to 0.18