parquet/file/
mod.rs

1// SPDX-License-Identifier: Apache-2.0
2// SPDX-FileCopyrightText: See the NOTICE file distributed with this work for additional information regarding copyright ownership
3// SPDX-FileContributor: Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements
4
5//! APIs for reading parquet data.
6//!
7//! Provides access to file and row group readers and writers, record API, metadata, etc.
8//!
9//! # See Also:
10//! * [`SerializedFileReader`] and [`SerializedFileWriter`] for reading / writing parquet
11//! * [`metadata`]: for working with metadata such as schema
12//! * [`statistics`]: for working with statistics in metadata
13//!
14//! [`SerializedFileReader`]: serialized_reader::SerializedFileReader
15//! [`SerializedFileWriter`]: writer::SerializedFileWriter
16//!
17//! # Example of writing a new file
18//!
19//! ```rust,no_run
20//! use std::{fs, path::Path, sync::Arc};
21//!
22//! use parquet::{
23//!     file::{
24//!         properties::WriterProperties,
25//!         writer::SerializedFileWriter,
26//!     },
27//!     schema::parser::parse_message_type,
28//! };
29//!
30//! let path = Path::new("/path/to/sample.parquet");
31//!
32//! let message_type = "
33//!   message schema {
34//!     REQUIRED INT32 b;
35//!   }
36//! ";
37//! let schema = Arc::new(parse_message_type(message_type).unwrap());
38//! let file = fs::File::create(&path).unwrap();
39//! let mut writer = SerializedFileWriter::new(file, schema, Default::default()).unwrap();
40//! let mut row_group_writer = writer.next_row_group().unwrap();
41//! while let Some(mut col_writer) = row_group_writer.next_column().unwrap() {
42//!     // ... write values to a column writer
43//!     col_writer.close().unwrap()
44//! }
45//! row_group_writer.close().unwrap();
46//! writer.close().unwrap();
47//!
48//! let bytes = fs::read(&path).unwrap();
49//! assert_eq!(&bytes[0..4], &[b'P', b'A', b'R', b'1']);
50//! ```
51//! # Example of reading an existing file
52//!
53//! ```rust,no_run
54//! use parquet::file::reader::{FileReader, SerializedFileReader};
55//! use std::{fs::File, path::Path};
56//!
57//! let path = Path::new("/path/to/sample.parquet");
58//! if let Ok(file) = File::open(&path) {
59//!     let reader = SerializedFileReader::new(file).unwrap();
60//!
61//!     let parquet_metadata = reader.metadata();
62//!     assert_eq!(parquet_metadata.num_row_groups(), 1);
63//!
64//!     let row_group_reader = reader.get_row_group(0).unwrap();
65//!     assert_eq!(row_group_reader.num_columns(), 1);
66//! }
67//! ```
68//! # Example of reading multiple files
69//!
70//! ```rust,no_run
71//! use parquet::file::reader::SerializedFileReader;
72//! use std::convert::TryFrom;
73//!
74//! let paths = vec![
75//!     "/path/to/sample.parquet/part-1.snappy.parquet",
76//!     "/path/to/sample.parquet/part-2.snappy.parquet"
77//! ];
78//! // Create a reader for each file and flat map rows
79//! let rows = paths.iter()
80//!     .map(|p| SerializedFileReader::try_from(*p).unwrap())
81//!     .flat_map(|r| r.into_iter());
82//!
83//! for row in rows {
84//!     println!("{}", row.unwrap());
85//! }
86//! ```
87#[cfg(feature = "encryption")]
88pub mod column_crypto_metadata;
89pub mod metadata;
90pub mod page_index;
91pub mod properties;
92pub mod reader;
93pub mod serialized_reader;
94pub mod statistics;
95pub mod writer;
96
97/// The length of the parquet footer in bytes
98pub const FOOTER_SIZE: usize = 8;
99const PARQUET_MAGIC: [u8; 4] = *b"PAR1";
100const PARQUET_MAGIC_ENCR_FOOTER: [u8; 4] = *b"PARE";
parquet/file/mod.rs

parquet/file/
mod.rs
