Struct polars::prelude::ParquetReader

pub struct ParquetReader<R>
where
    R: Read + Seek,
{ /* private fields */ }

Available on crate feature polars-io only.

Read Apache parquet format into a DataFrame.

Implementations

impl<R> ParquetReader<R>

where R: MmapBytesReader,

pub fn set_low_memory(self, low_memory: bool) -> ParquetReader<R>

Try to reduce memory pressure at the expense of performance. If setting this does not reduce memory enough, turn off parallelization.

pub fn read_parallel(self, parallel: ParallelStrategy) -> ParquetReader<R>

Read the parquet file in parallel (default). The single threaded reader consumes less memory.

pub fn with_n_rows(self, num_rows: Option<usize>) -> ParquetReader<R>

Stop parsing when n rows are parsed. By settings this parameter the csv will be parsed sequentially.

pub fn with_columns(self, columns: Option<Vec<String>>) -> ParquetReader<R>

Columns to select/ project

pub fn with_projection(self, projection: Option<Vec<usize>>) -> ParquetReader<R>

Set the reader’s column projection. This counts from 0, meaning that vec![0, 4] would select the 1st and 5th column.

pub fn with_row_index(self, row_index: Option<RowIndex>) -> ParquetReader<R>

Add a row index column.

pub fn with_schema(self, schema: Option<Arc<ArrowSchema>>) -> ParquetReader<R>

Set the Schema if already known. This must be exactly the same as the schema in the file itself.

pub fn schema(&mut self) -> Result<Arc<ArrowSchema>, PolarsError>

Schema of the file.

pub fn use_statistics(self, toggle: bool) -> ParquetReader<R>

Use statistics in the parquet to determine if pages can be skipped from reading.

pub fn num_rows(&mut self) -> Result<usize, PolarsError>

Number of rows in the parquet file.

pub fn with_hive_partition_columns( self, columns: Option<Vec<Series>> ) -> ParquetReader<R>

pub fn get_metadata(&mut self) -> Result<&Arc<FileMetaData>, PolarsError>

pub fn with_predicate( self, predicate: Option<Arc<dyn PhysicalIoExpr>> ) -> ParquetReader<R>

impl<R> ParquetReader<R>

where R: MmapBytesReader + 'static,

pub fn batched( self, chunk_size: usize ) -> Result<BatchedParquetReader, PolarsError>

Trait Implementations

impl<R> SerReader<R> for ParquetReader<R>

where R: MmapBytesReader,

fn new(reader: R) -> ParquetReader<R>

Create a new ParquetReader from an existing Reader.

fn set_rechunk(self, rechunk: bool) -> ParquetReader<R>

Make sure that all columns are contiguous in memory by aggregating the chunks into a single array.

fn finish(self) -> Result<DataFrame, PolarsError>

Take the SerReader and return a parsed DataFrame.