Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix binary operations Index by Series. #2046

Merged
merged 2 commits into from
Feb 9, 2021
Merged

Fix binary operations Index by Series. #2046

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7beecb4
Fix binary operations Index by Series.
ueshin Feb 8, 2021
3687b30
Fix.
ueshin Feb 8, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 21 additions & 10 deletions databricks/koalas/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,15 +96,19 @@ def align_diff_index_ops(func, this_index_ops: "IndexOpsMixin", *args) -> "Index
cols = [arg for arg in args if isinstance(arg, IndexOpsMixin)]

if isinstance(this_index_ops, Series) and all(isinstance(col, Series) for col in cols):
combined = combine_frames(this_index_ops.to_frame(), *cols, how="full")
combined = combine_frames(
this_index_ops.to_frame(),
*[cast(Series, col).rename(i) for i, col in enumerate(cols)],
how="full"
)

return column_op(func)(
combined["this"]._kser_for(combined["this"]._internal.column_labels[0]),
*[
combined["that"]._kser_for(label)
for label in combined["that"]._internal.column_labels
]
)
).rename(this_index_ops.name)
else:
# This could cause as many counts, reset_index calls, joins for combining
# as the number of `Index`s in `args`. So far it's fine since we can assume the ops
Expand Down Expand Up @@ -137,10 +141,10 @@ def align_diff_index_ops(func, this_index_ops: "IndexOpsMixin", *args) -> "Index
elif isinstance(this_index_ops, Series):
this = this_index_ops.reset_index()
that = [
cast(Series, col.to_series() if isinstance(col, Index) else col).reset_index(
drop=True
)
for col in cols
cast(Series, col.to_series() if isinstance(col, Index) else col)
.rename(i)
.reset_index(drop=True)
for i, col in enumerate(cols)
]

combined = combine_frames(this, *that, how="full").sort_index()
Expand All @@ -155,13 +159,16 @@ def align_diff_index_ops(func, this_index_ops: "IndexOpsMixin", *args) -> "Index
combined["that"]._kser_for(label)
for label in combined["that"]._internal.column_labels
]
)
).rename(this_index_ops.name)
else:
this = cast(Index, this_index_ops).to_frame().reset_index(drop=True)

that_series = next(col for col in cols if isinstance(col, Series))
that_frame = that_series._kdf[
[col.to_series() if isinstance(col, Index) else col for col in cols]
[
cast(Series, col.to_series() if isinstance(col, Index) else col).rename(i)
for i, col in enumerate(cols)
]
]

combined = combine_frames(this, that_frame.reset_index()).sort_index()
Expand All @@ -176,8 +183,12 @@ def align_diff_index_ops(func, this_index_ops: "IndexOpsMixin", *args) -> "Index
other.index.names = that_series._internal.index_names

return column_op(func)(
self_index, *[other._kser_for(label) for label in other._internal.column_labels]
)
self_index,
*[
other._kser_for(label)
for label, col in zip(other._internal.column_labels, cols)
]
).rename(that_series.name)


def booleanize_null(left_scol, scol, f) -> Column:
Expand Down
42 changes: 38 additions & 4 deletions databricks/koalas/tests/test_ops_on_diff_frames.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1365,15 +1365,41 @@ def test_series_repeat(self):

def test_series_ops(self):
pser1 = pd.Series([1, 2, 3, 4, 5, 6, 7], name="x", index=[11, 12, 13, 14, 15, 16, 17])
pser2 = pd.Series([1, 2, 3, 4, 5, 6, 7], index=[11, 12, 13, 14, 15, 16, 17])
pidx1 = pd.Index([10, 11, 12, 13, 14, 15, 16])
pser2 = pd.Series([1, 2, 3, 4, 5, 6, 7], name="x", index=[11, 12, 13, 14, 15, 16, 17])
pidx1 = pd.Index([10, 11, 12, 13, 14, 15, 16], name="x")
kser1 = ks.from_pandas(pser1)
kser2 = ks.from_pandas(pser2)
kidx1 = ks.from_pandas(pidx1)

self.assert_eq((kser1 + 1 + 10 * kser2).sort_index(), (pser1 + 1 + 10 * pser2).sort_index())
self.assert_eq(
(kser1 + 1 + 10 * kser2.rename()).sort_index(),
(pser1 + 1 + 10 * pser2.rename()).sort_index(),
)
self.assert_eq(
(kser1.rename() + 1 + 10 * kser2).sort_index(),
(pser1.rename() + 1 + 10 * pser2).sort_index(),
)
self.assert_eq(
(kser1.rename() + 1 + 10 * kser2.rename()).sort_index(),
(pser1.rename() + 1 + 10 * pser2.rename()).sort_index(),
)

self.assert_eq(kser1 + 1 + 10 * kidx1, pser1 + 1 + 10 * pidx1)
self.assert_eq(kser1.rename() + 1 + 10 * kidx1, pser1.rename() + 1 + 10 * pidx1)
self.assert_eq(kser1 + 1 + 10 * kidx1.rename(None), pser1 + 1 + 10 * pidx1.rename(None))
self.assert_eq(
kser1.rename() + 1 + 10 * kidx1.rename(None),
pser1.rename() + 1 + 10 * pidx1.rename(None),
)

self.assert_eq(kidx1 + 1 + 10 * kser1, pidx1 + 1 + 10 * pser1)
self.assert_eq(kidx1 + 1 + 10 * kser1.rename(), pidx1 + 1 + 10 * pser1.rename())
self.assert_eq(kidx1.rename(None) + 1 + 10 * kser1, pidx1.rename(None) + 1 + 10 * pser1)
self.assert_eq(
kidx1.rename(None) + 1 + 10 * kser1.rename(),
pidx1.rename(None) + 1 + 10 * pser1.rename(),
)

pidx2 = pd.Index([11, 12, 13])
kidx2 = ks.from_pandas(pidx2)
Expand All @@ -1389,12 +1415,20 @@ def test_series_ops(self):
kidx2 + kser1

def test_index_ops(self):
pidx1 = pd.Index([1, 2, 3, 4, 5])
pidx2 = pd.Index([6, 7, 8, 9, 10])
pidx1 = pd.Index([1, 2, 3, 4, 5], name="x")
pidx2 = pd.Index([6, 7, 8, 9, 10], name="x")
kidx1 = ks.from_pandas(pidx1)
kidx2 = ks.from_pandas(pidx2)

self.assert_eq(kidx1 * 10 + kidx2, pidx1 * 10 + pidx2)
self.assert_eq(kidx1.rename(None) * 10 + kidx2, pidx1.rename(None) * 10 + pidx2)

if LooseVersion(pd.__version__) >= LooseVersion("1.0"):
self.assert_eq(kidx1 * 10 + kidx2.rename(None), pidx1 * 10 + pidx2.rename(None))
else:
self.assert_eq(
kidx1 * 10 + kidx2.rename(None), (pidx1 * 10 + pidx2.rename(None)).rename(None)
)

pidx3 = pd.Index([11, 12, 13])
kidx3 = ks.from_pandas(pidx3)
Expand Down