https://github.com/deacondesperado/nektar

A command line client for Apache Hive metastore implemented in Rust
https://github.com/deacondesperado/nektar

Last synced: 2 months ago
JSON representation

A command line client for Apache Hive metastore implemented in Rust

• Host: GitHub
• URL: https://github.com/deacondesperado/nektar
• Owner: DeaconDesperado
• License: mit
• Created: 2023-09-14T17:40:41.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-10-07T00:48:58.000Z (3 months ago)
• Last Synced: 2024-10-12T10:13:46.089Z (3 months ago)
• Language: Rust
• Size: 342 KB
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 6
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Nektar

A fast and lightweight command line interface for the [Apache Hive](https://hive.apache.org/) metastore.

## Motivation

Hive's metastore is ubiquitous in many big data deployments, whether the deployment in question makes use

of the Hive execution engine or not.  It is a common component that provides table virtualization for other 

distributed processing engines such as [Flink](https://nightlies.apache.org/flink/flink-docs-master/docs/connectors/table/hive/hive_catalog/), [Spark](https://spark.apache.org/docs/latest/sql-data-sources-hive-tables.html) and [Trino](https://trino.io/docs/current/connector/hive.html).  

More modern analytic table formats such as [Iceberg](https://iceberg.apache.org/docs/latest/hive/) and [Hudi](https://hudi.apache.org/docs/syncing_metastore/) leverage Hive's metastore to some extent for the serving and storage layer of their metadata.  Cloud data catalogs such as GCP's [DataProc Metastore](https://cloud.google.com/dataproc-metastore/docs/hive-metastore) and [AWS's Glue Data Catalog](https://docs.aws.amazon.com/emr/latest/ReleaseGuide/emr-hive-metastore-glue.html) also implement the metastore interface.

Typically, operations against the metastore are done via these engines' catalog procedures, or in SQL DDL 

using the Hive Cli itself.  There are limited options for calling the metastore's endpoints themselves directly.  This

project provides a lightweight command line interface to the metastore's Thrift endpoints directly, to allow lower-level

debugging of metastore payloads without JVM startup overhead.

# Usage

Install via `cargo install` or download a binary directly from releases.

```

A fast, lightweight CLI for Hive Metastore

Usage: nektar [OPTIONS]  

Commands:

  get-table                     Get a single table by database and table name

  get-catalog                   Get a single catalog by name

  get-catalogs                  Get a list of all catalogs in metastore

  get-partitions                Get the partitions of a table

  get-partition-names-by-parts  Get the partitions of a table by partition value

  get-databases                 Get all databases in the metastore

  create-catalog                Create a catalog

  create-table                  Create a table from a table definition file

  list-tables                   Get tables in database with optional glob on name

  drop-table                    Drop a single table by database and table name

  help                          Print this message or the help of the given subcommand(s)

Arguments:

    Thrift metastore endpoint, eg: host.com:9083

Options:

      --format   [default: json] [possible values: json, yaml]

  -h, --help             Print help

  -V, --version          Print version

```