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.

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

Fix skiprows issue with ORC Reader #7359

Merged
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
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
5 changes: 3 additions & 2 deletions cpp/src/io/orc/stripe_data.cu
Original file line number Diff line number Diff line change
Expand Up @@ -1211,8 +1211,9 @@ __global__ void __launch_bounds__(block_size)
uint32_t skippedrows = min(static_cast<uint32_t>(first_row - row_in), nrows);
uint32_t skip_count = 0;
for (uint32_t i = t * 32; i < skippedrows; i += 32 * 32) {
uint32_t bits = s->vals.u32[i >> 5];
if (i + 32 > skippedrows) { bits &= (1 << (skippedrows - i)) - 1; }
vuule marked this conversation as resolved.
Show resolved Hide resolved
uint32_t bits = (i + 32 <= skippedrows)
? s->vals.u32[i >> 5]
: (rle8_read_bool32(s->vals.u32, i) & ((1 << (skippedrows - i)) - 1));
skip_count += __popc(bits);
}
skip_count = warp_reduce(temp_storage[t / 32]).Sum(skip_count);
Expand Down
26 changes: 26 additions & 0 deletions python/cudf/cudf/tests/test_orc.py
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@
import pandas as pd
import pyarrow as pa
import pyarrow.orc
import pyorc
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

CI fails because of this import

ModuleNotFoundError: No module named 'pyorc'

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Created a PR for it rapidsai/integration#215

import pytest

import cudf
Expand Down Expand Up @@ -318,6 +319,31 @@ def test_orc_read_rows(datadir, skiprows, num_rows):
np.testing.assert_allclose(pdf, gdf)


def test_orc_read_skiprows(tmpdir):
fname = tmpdir.join("TestOrcFile.skiprows.orc")
rgsl888prabhu marked this conversation as resolved.
Show resolved Hide resolved
df = pd.DataFrame(
{"a": [1, 0, 1, 0, None, 1, 1, 1, 0, None, 0, 0, 1, 1, 1, 1]},
dtype=pd.BooleanDtype(),
)
output = open(fname, "wb")
writer = pyorc.Writer(output, pyorc.Struct(a=pyorc.Boolean()))
tuples = list(
map(
lambda x: (None,) if x[0] is pd.NA else x,
list(df.itertuples(index=False, name=None)),
)
)
writer.writerows(tuples)
writer.close()

skiprows = 10

expected = cudf.read_orc(fname)[skiprows::].reset_index(drop=True)
got = cudf.read_orc(fname, skiprows=skiprows)

assert_eq(expected, got)


def test_orc_reader_uncompressed_block(datadir):
path = datadir / "uncompressed_snappy.orc"
try:
Expand Down