polars_io/parquet/read/
mod.rs

Help
1//! Functionality for reading Apache Parquet files.
2//!
3//! # Examples
4//!
5//! ```
6//! use polars_core::prelude::*;
7//! use polars_io::prelude::*;
8//! use std::fs::File;
9//!
10//! fn example() -> PolarsResult<DataFrame> {
11//!     let r = File::open("example.parquet").unwrap();
12//!     let reader = ParquetReader::new(r);
13//!     reader.finish()
14//! }
15//! ```
16
17#[cfg(feature = "cloud")]
18mod async_impl;
19mod mmap;
20mod options;
21mod predicates;
22mod read_impl;
23mod reader;
24mod utils;
25
26const ROW_COUNT_OVERFLOW_ERR: PolarsError = PolarsError::ComputeError(ErrString::new_static(
27    "\
28Parquet file produces more than pow(2, 32) rows; \
29consider compiling with polars-bigidx feature (polars-u64-idx package on python), \
30or set 'streaming'",
31));
32
33pub use options::{ParallelStrategy, ParquetOptions};
34use polars_error::{ErrString, PolarsError};
35pub use polars_parquet::arrow::read::infer_schema;
36pub use polars_parquet::read::FileMetadata;
37pub use read_impl::{create_sorting_map, try_set_sorted_flag};
38#[cfg(feature = "cloud")]
39pub use reader::ParquetAsyncReader;
40pub use reader::{BatchedParquetReader, ParquetReader};
41pub use utils::materialize_empty_df;
42
43pub mod _internal {
44    pub use super::mmap::to_deserializer;
45    pub use super::predicates::{collect_statistics_with_live_columns, read_this_row_group};
46    pub use super::read_impl::{PrefilterMaskSetting, calc_prefilter_cost};
47    pub use super::utils::ensure_matching_dtypes_if_found;
48}
polars_io/parquet/read/mod.rs

polars_io/parquet/read/
mod.rs
