Expand description
SerializedFileWriter: Low level Parquet writer API
Re-exportsยง
use crate::bloom_filter::Sbbf;๐use crate::file::metadata::page_index::PageIndex;๐use crate::file::metadata::thrift::PageHeader;๐use crate::file::page_index::column_index::ColumnIndexMetaData;๐use crate::file::page_index::offset_index::OffsetIndexMetaData;๐use crate::parquet_thrift::ThriftCompactOutputProtocol;๐use crate::parquet_thrift::WriteThrift;๐use bytes::Bytes;๐use std::fmt::Debug;๐use std::fmt::Debug;๐use std::io::BufWriter;๐use std::io::IoSlice;๐use std::io::Read;๐use std::io::Write;๐use std::sync::Arc;๐use crate::column::page_encryption::PageEncryptor;๐use crate::column::writer::ColumnCloseResult;๐use crate::column::writer::ColumnWriterImpl;๐use crate::column::writer::get_typed_column_writer_mut;๐use crate::column::page::CompressedPage;๐use crate::column::page::PageWriteSpec;๐use crate::column::page::PageWriter;๐use crate::column::writer::ColumnWriter;๐use crate::column::writer::get_column_writer;๐use crate::data_type::DataType;๐use crate::encryption::encrypt::FileEncryptionProperties;๐use crate::encryption::encrypt::FileEncryptor;๐use crate::encryption::encrypt::get_column_crypto_metadata;๐use crate::errors::ParquetError;๐use crate::errors::Result;๐use crate::file::PARQUET_MAGIC_ENCR_FOOTER;๐use crate::file::properties::BloomFilterPosition;๐use crate::file::properties::WriterPropertiesPtr;๐use crate::file::reader::ChunkReader;๐use crate::file::PARQUET_MAGIC;๐use crate::schema::types::ColumnDescPtr;๐use crate::schema::types::SchemaDescPtr;๐use crate::schema::types::SchemaDescriptor;๐use crate::schema::types::TypePtr;๐use crate::file::metadata::*;๐
Structsยง
- Page
Encryptor ๐Context - Context required to create a
PageEncryptorfor a column - Serialized
Column Writer - A wrapper around a
ColumnWriterthat invokes a callback onSelf::close - Serialized
File Writer - Parquet file writer API.
- Serialized
Page Writer - A serialized implementation for Parquet
PageWriter. Writes and serializes pages and metadata into output stream. - Serialized
RowGroup Writer - Parquet row group writer API.
- Tracked
Write - A wrapper around a
Writethat keeps track of the number of bytes that have been written. The givenWriteis wrapped with aBufWriterto optimize writing performance.
Functionsยง
- get_
file_ ๐magic - Get the magic bytes at the start and end of the file that identify this as a Parquet file.
- write_
bloom_ ๐filters - Serialize all the bloom filters of the given row group to the given buffer, and returns the updated row group metadata.
Type Aliasesยง
- OnClose
Column Chunk - Callback invoked on closing a column chunk
- OnClose
RowGroup - Callback invoked on closing a row group, arguments are: