iceberg-rust - GitLite

KO | EN

gitlite — search

#hacktoberfest #react #ai #go #golang #android #linux #docker #nodejs #windows #awesome #vue #awesome-list #pytorch

iceberg-rust

(★ 245)

Unofficial rust implementation of Apache Iceberg with integration for Datafusion

파일 탐색기

최종 버전 다운로드 (.zip)

- - rust.yml
- dependabot.yml
- - Cargo.toml
- - Cargo.toml
- - .gitignore
  - Cargo.toml
  - README.md
- - Cargo.toml
- - Cargo.toml
- - context.rs
  - lib.rs
  - schema.rs
- Cargo.toml
- - dataframe.rs
  - insert_csv.rs
  - insert_table.rs
  - refresh_materialized_view.rs
- - error.rs
  - lib.rs
  - planner.rs
  - pruning_statistics.rs
  - statistics.rs
- - empty_insert.rs
  - equality_delete.rs
  - insert_csv.rs
  - insert_sequence_number.rs
  - integration_df_create_table.rs
  - integration_df_ctas.rs
  - integration_df_dml.rs
  - integration_df_filters.rs
  - integration_df_namespace_sql.rs
  - integration_df_query.rs
  - integration_df_subqueries.rs
  - integration_df_types.rs
  - integration_df_unsupported.rs
  - integration_df_views.rs
  - integration_pyiceberg.rs
  - integration_spark.rs
  - integration_trino.rs
  - roundtrip_types.rs
  - sort_order_scan.rs
  - table_statistics.rs
  - upstream_sql_engine_placeholders.rs
- Cargo.toml
- README.md
- - config.rs
  - error.rs
  - lib.rs
  - spec.rs
  - sql.rs
- - overwrite_test.rs
  - sort_order_test.rs
- .gitignore
- Cargo.toml
- README.md
- - error.rs
  - lib.rs
  - util.rs
- Cargo.toml
.gitignore
.history
Cargo.lock
Cargo.toml
CLAUDE.md
LICENSE
Makefile
README.md

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/JanKaul/iceberg-rust

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd iceberg-rust

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. Rust

보통 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Rust (rustup) rustup으로 설치하면 cargo도 함께 설치됩니다.

cargo build --release

Rust 프로젝트를 컴파일합니다.

cargo run

빌드 후 바로 실행까지 진행합니다.

✅ cargo build가 에러 없이 끝나면 성공입니다. target/ 폴더에 실행 파일이 생성됩니다.

3. Make

보통

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Make Linux/macOS는 보통 기본 설치되어 있습니다. Windows는 별도 설치(예: MSYS2, WSL)가 필요합니다.

make

생성된 빌드 설정을 바탕으로 실제 컴파일을 진행해 실행 파일을 만듭니다.

✅ 에러 없이 끝나면 성공입니다. 생성된 실행 파일을 직접 실행해보세요.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Rust implementation of [Apache Iceberg](https://iceberg.apache.org)

Apache Iceberg is Open Table Format that brings ACID quarantees to large analytic datasets. 
This repository contains a Rust implementation of Apache Iceberg that focuses on the interoperability with the Arrow ecosystem.
It provides an Iceberg integration for the [Datafusion](https://arrow.apache.org/datafusion/) query engine.

[![Crates.io][crates-badge]][crates-url]
[![Apache V2.0 licensed][apache-badge]][apache-url]
[![Build Status][actions-badge]][actions-url]

[crates-badge]: https://img.shields.io/crates/v/iceberg-rust
[crates-url]: https://crates.io/crates/iceberg-rust
[apache-badge]: https://img.shields.io/badge/License-Apache_2.0-blue.svg
[apache-url]: https://github.com/JanKaul/iceberg-rust/blob/main/LICENSE
[actions-badge]: https://github.com/JanKaul/iceberg-rust/actions/workflows/rust.yml/badge.svg?branch=main
[actions-url]: https://github.com/JanKaul/iceberg-rust/actions/workflows/rust.yml

## Features

### Iceberg tables

| Feature | Status |
| --- | --- |
| Read | :white_check_mark: |
| Read partitioned | :white_check_mark: |
| Insert | :white_check_mark: |
| Insert partitioned | :white_check_mark: |
| Equality deletes | :white_check_mark: |
| Positional deletes | |

### Table Maintenance

| Feature | Status |
| --- | --- |
| Expire snapshots | :white_check_mark: |
| Orphan file cleanup | :white_check_mark: |

### Iceberg Views

| Feature | Status |
| --- | --- |
| Read | :white_check_mark: |

### Iceberg Materialized Views

| Feature | Status |
| --- | --- |
| Read | :white_check_mark: |
| Full refresh | :white_check_mark: |
| Incremental refresh | :white_check_mark: |

### Catalogs

- REST
- S3Tables
- Filesystem
- Glue
- RDBMS (Postgres, MySQL)

### File formats

- parquet

### Integrations

- [Datafusion](https://arrow.apache.org/datafusion/)

## Example

Check out the [datafusion examples](datafusion_iceberg/examples).

### Basic Table Operations

```rust
use datafusion::{arrow::array::Int64Array, prelude::SessionContext};
use datafusion_iceberg::DataFusionTable;
use iceberg_rust::{
    catalog::Catalog,
    spec::{
        partition::{PartitionField, PartitionSpec, Transform},
        schema::Schema,
        types::{PrimitiveType, StructField, StructType, Type},
    },
    table::Table,
};
use iceberg_sql_catalog::SqlCatalog;
use object_store::memory::InMemory;
use object_store::ObjectStore;

use std::sync::Arc;

#[tokio::main]
pub(crate) async fn main() {
    let object_store: Arc<dyn ObjectStore> = Arc::new(InMemory::new());

let catalog: Arc<dyn Catalog> = Arc::new(
        SqlCatalog::new("sqlite://", "test", object_store.clone())
            .await
            .unwrap(),
    );

let schema = Schema::builder()
        .with_fields(
            StructType::builder()
                .with_struct_field(StructField {
                    id: 1,
                    name: "id".to_string(),
                    required: true,
                    field_type: Type::Primitive(PrimitiveType::Long),
                    doc: None,
                })
                .with_struct_field(StructField {
                    id: 2,
                    name: "customer_id".to_string(),
                    required: true,
                    field_type: Type::Primitive(PrimitiveType::Long),
                    doc: None,
                })
                .with_struct_field(StructField {
                    id: 3,
                    name: "product_id".to_string(),
                    required: true,
                    field_type: Type::Primitive(PrimitiveType::Long),
                    doc: None,
                })
                .with_struct_field(StructField {
                    id: 4,
                    name: "date".to_string(),
                    required: true,
                    field_type: Type::Primitive(PrimitiveType::Date),
                    doc: None,
                })
                .with_struct_field(StructField {
                    id: 5,
                    name: "amount".to_string(),
                    required: true,
                    field_type: Type::Primitive(PrimitiveType::Int),
                    doc: None,
                })
                .build()
                .unwrap(),
        )
        .build()
        .unwrap();

let partition_spec = PartitionSpec::builder()
        .with_partition_field(PartitionField::new(4, 1000, "day", Transform::Day))
        .build()
        .expect("Failed to create partition spec");

let table = Table::builder()
        .with_name("orders")
        .with_location("/test/orders")
        .with_schema(schema)
        .with_partition_spec(partition_spec)
        .build(&["test".to_owned()], catalog)
        .await
        .expect("Failed to create table");

let table = Arc::new(DataFusionTable::from(table));

let ctx = SessionContext::new();

ctx.register_table("orders", table).unwrap();

ctx.sql(
        "INSERT INTO orders (id, customer_id, product_id, date, amount) VALUES 
        (1, 1, 1, '2020-01-01', 1),
        (2, 2, 1, '2020-01-01', 1),
        (3, 3, 1, '2020-01-01', 3),
        (4, 1, 2, '2020-02-02', 1),
        (5, 1, 1, '2020-02-02', 2),
        (6, 3, 3, '2020-02-02', 3);",
    )
    .await
    .expect("Failed to create query plan for insert")
    .collect()
    .await
    .expect("Failed to insert values into table");

let batches = ctx
        .sql("select product_id, sum(amount) from orders group by product_id;")
        .await
        .expect("Failed to create plan for select")
        .collect()
        .await
        .expect("Failed to execute select query");

for batch in batches {
        if batch.num_rows() != 0 {
            let (product_ids, amounts) = (
                batch
                    .column(0)
                    .as_any()
                    .downcast_ref::<Int64Array>()
                    .unwrap(),
                batch
                    .column(1)
                    .as_any()
                    .downcast_ref::<Int64Array>()
                    .unwrap(),
            );
            for (product_id, amount) in product_ids.iter().zip(amounts) {
                if product_id.unwrap() == 1 {
                    assert_eq!(amount.unwrap(), 7)
                } else if product_id.unwrap() == 2 {
                    assert_eq!(amount.unwrap(), 1)
                } else if product_id.unwrap() == 3 {
                    assert_eq!(amount.unwrap(), 3)
                } else {
                    panic!("Unexpected product id")
                }
            }
        }
    }

ctx.sql(
        "INSERT INTO orders (id, customer_id, product_id, date, amount) VALUES 
        (7, 1, 3, '2020-01-03', 1),
        (8, 2, 1, '2020-01-03', 2),
        (9, 2, 2, '2020-01-03', 1);",
    )
    .await
    .expect("Failed to create query plan for insert")
    .collect()
    .await
    .expect("Failed to insert values into table");

let batches = ctx
        .sql("select product_id, sum(amount) from orders group by product_id;")
        .await
        .expect("Failed to create plan for select")
        .collect()
        .await
        .expect("Failed to execute select query");

for batch in batches {
        if batch.num_rows() != 0 {
            let (product_ids, amounts) = (
                batch
                    .column(0)
                    .as_any()
                    .downcast_ref::<Int64Array>()
                    .unwrap(),
                batch
                    .column(1)
                    .as_any()
                    .downcast_ref::<Int64Array>()
                    .unwrap(),
            );
            for (product_id, amount) in product_ids.iter().zip(amounts) {
                if product_id.unwrap() == 1 {
                    assert_eq!(amount.unwrap(), 9)
                } else if product_id.unwrap() == 2 {
                    assert_eq!(amount.unwrap(), 2)
                } else if product_id.unwrap() == 3 {
                    assert_eq!(amount.unwrap(), 4)
                } else {
                    panic!("Unexpected product id")
                }
            }
        }
    }
}
```