Skip to content

POC (do not merge): Add row-group and column select #222

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

Draft
martindurant wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

POC (do not merge): Add row-group and column select #222

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
22 changes: 18 additions & 4 deletions arro3-io/src/parquet.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,15 +1,16 @@
use std::any::Any;
use std::collections::HashMap;
use std::str::FromStr;
use std::sync::Arc;

use arrow_array::{RecordBatchIterator, RecordBatchReader};
use parquet::arrow::arrow_reader::ParquetRecordBatchReaderBuilder;
use parquet::arrow::arrow_writer::ArrowWriterOptions;
use parquet::arrow::ArrowWriter;
use parquet::arrow::{ArrowWriter, ProjectionMask};
use parquet::basic::{Compression, Encoding};
use parquet::file::properties::{WriterProperties, WriterVersion};
use parquet::format::KeyValue;
use parquet::schema::types::ColumnPath;
use parquet::schema::types::{ColumnPath, SchemaDescriptor};
use pyo3::exceptions::{PyTypeError, PyValueError};
use pyo3::prelude::*;
use pyo3_arrow::error::PyArrowResult;
Expand All @@ -19,8 +20,21 @@ use pyo3_arrow::PyRecordBatchReader;
use crate::utils::{FileReader, FileWriter};

#[pyfunction]
pub fn read_parquet(py: Python, file: FileReader) -> PyArrowResult<PyObject> {
let builder = ParquetRecordBatchReaderBuilder::try_new(file).unwrap();
pub fn read_parquet(
py: Python,
file: FileReader,
rgs: Option<Vec<usize>>,
Copy link
Owner

@kylebarron kylebarron Oct 9, 2024

Choose a reason for hiding this comment

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

I'd call this row_groups for symmetry with pyarrow https://arrow.apache.org/docs/python/generated/pyarrow.parquet.ParquetFile.html#pyarrow.parquet.ParquetFile.read_row_groups

Copy link
Contributor Author

@martindurant martindurant Oct 10, 2024

Choose a reason for hiding this comment

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

Certainly! It really ought to be on the reader (.select() or such) anyway, so that you only read the parquet metadata once. Or maybe have a separate "fetch details" function. This was only minimalist POC to show that you can pick data portions like this.

columns: Option<Vec<usize>>,
) -> PyArrowResult<PyObject> {
let mut builder = ParquetRecordBatchReaderBuilder::try_new(file).unwrap();
if let Some(nn) = rgs {
builder = builder.with_row_groups(nn);
}

if let Some(cols) = columns {
let projection = ProjectionMask::leaves(builder.parquet_schema(), cols);
builder = builder.with_projection(projection);
}

let metadata = builder.schema().metadata().clone();
let reader = builder.build().unwrap();
Expand Down