This crate contains the official Native Rust implementation of Apache Parquet, which is part of the Apache Arrow project.
Example usage of reading data:
use std::fs::File;
use std::path::Path;
use parquet::file::reader::{FileReader, SerializedFileReader};
let file = File::open(&Path::new("/path/to/file")).unwrap();
let reader = SerializedFileReader::new(file).unwrap();
let mut iter = reader.get_row_iter(None).unwrap();
while let Some(record) = iter.next() {
println!("{}", record);
}
See crate documentation on available API.
This crate is tested with the latest stable version of Rust. We do not currrently test against other, older versions of the Rust compiler.
If you are upgrading from version 3.0 or previous of this crate, you
likely need to change your code to use [ConvertedType
] rather than
[LogicalType
] to preserve existing behaviour in your code.
Version 2.4.0 of the Parquet format introduced a LogicalType
to replace the existing ConvertedType
.
This crate used parquet::basic::LogicalType
to map to the ConvertedType
, but this has been renamed to parquet::basic::ConvertedType
from version 4.0 of this crate.
The ConvertedType
is deprecated in the format, but is still written
to preserve backward compatibility.
It is preferred that LogicalType
is used, as it supports nanosecond
precision timestamps without using the deprecated Int96
Parquet type.
- Parquet-format 2.6.0
To update Parquet format to a newer version, check if parquet-format
version is available. Then simply update version of parquet-format
crate in Cargo.toml.
- All encodings supported
- All compression codecs supported
- Read support
- Primitive column value readers
- Row record reader
- Arrow record reader
- Statistics support
- Write support
- Primitive column value writers
- Row record writer
- Arrow record writer
- Predicate pushdown
- Parquet format 2.6.0 support
Licensed under the Apache License, Version 2.0: http://www.apache.org/licenses/LICENSE-2.0.