|
| 1 | +// Licensed to the Apache Software Foundation (ASF) under one |
| 2 | +// or more contributor license agreements. See the NOTICE file |
| 3 | +// distributed with this work for additional information |
| 4 | +// regarding copyright ownership. The ASF licenses this file |
| 5 | +// to you under the Apache License, Version 2.0 (the |
| 6 | +// "License"); you may not use this file except in compliance |
| 7 | +// with the License. You may obtain a copy of the License at |
| 8 | +// |
| 9 | +// http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | +// |
| 11 | +// Unless required by applicable law or agreed to in writing, |
| 12 | +// software distributed under the License is distributed on an |
| 13 | +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
| 14 | +// KIND, either express or implied. See the License for the |
| 15 | +// specific language governing permissions and limitations |
| 16 | +// under the License. |
| 17 | + |
| 18 | +use super::utils::*; |
| 19 | +use async_trait::async_trait; |
| 20 | +use hive_metastore::{TThriftHiveMetastoreSyncClient, ThriftHiveMetastoreSyncClient}; |
| 21 | +use iceberg::table::Table; |
| 22 | +use iceberg::{Catalog, Namespace, NamespaceIdent, Result, TableCommit, TableCreation, TableIdent}; |
| 23 | +use std::collections::HashMap; |
| 24 | +use std::fmt::{Debug, Formatter}; |
| 25 | +use std::sync::{Arc, Mutex}; |
| 26 | +use thrift::protocol::{TBinaryInputProtocol, TBinaryOutputProtocol}; |
| 27 | +use thrift::transport::{ |
| 28 | + ReadHalf, TBufferedReadTransport, TBufferedWriteTransport, TIoChannel, WriteHalf, |
| 29 | +}; |
| 30 | +use typed_builder::TypedBuilder; |
| 31 | + |
| 32 | +/// Hive metastore Catalog configuration. |
| 33 | +#[derive(Debug, TypedBuilder)] |
| 34 | +pub struct HmsCatalogConfig { |
| 35 | + address: String, |
| 36 | +} |
| 37 | + |
| 38 | +/// TODO: We only support binary protocol for now. |
| 39 | +type HmsClientType = ThriftHiveMetastoreSyncClient< |
| 40 | + TBinaryInputProtocol<TBufferedReadTransport<ReadHalf<thrift::transport::TTcpChannel>>>, |
| 41 | + TBinaryOutputProtocol<TBufferedWriteTransport<WriteHalf<thrift::transport::TTcpChannel>>>, |
| 42 | +>; |
| 43 | + |
| 44 | +/// # TODO |
| 45 | +/// |
| 46 | +/// we are using the same connection everytime, we should support connection |
| 47 | +/// pool in the future. |
| 48 | +struct HmsClient(Arc<Mutex<HmsClientType>>); |
| 49 | + |
| 50 | +impl HmsClient { |
| 51 | + fn call<T>(&self, f: impl FnOnce(&mut HmsClientType) -> thrift::Result<T>) -> Result<T> { |
| 52 | + let mut client = self.0.lock().unwrap(); |
| 53 | + f(&mut client).map_err(from_thrift_error) |
| 54 | + } |
| 55 | +} |
| 56 | + |
| 57 | +/// Hive metastore Catalog. |
| 58 | +pub struct HmsCatalog { |
| 59 | + config: HmsCatalogConfig, |
| 60 | + client: HmsClient, |
| 61 | +} |
| 62 | + |
| 63 | +impl Debug for HmsCatalog { |
| 64 | + fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result { |
| 65 | + f.debug_struct("HmsCatalog") |
| 66 | + .field("config", &self.config) |
| 67 | + .finish_non_exhaustive() |
| 68 | + } |
| 69 | +} |
| 70 | + |
| 71 | +impl HmsCatalog { |
| 72 | + /// Create a new hms catalog. |
| 73 | + pub fn new(config: HmsCatalogConfig) -> Result<Self> { |
| 74 | + let mut channel = thrift::transport::TTcpChannel::new(); |
| 75 | + channel |
| 76 | + .open(config.address.as_str()) |
| 77 | + .map_err(from_thrift_error)?; |
| 78 | + let (i_chan, o_chan) = channel.split().map_err(from_thrift_error)?; |
| 79 | + let i_chan = TBufferedReadTransport::new(i_chan); |
| 80 | + let o_chan = TBufferedWriteTransport::new(o_chan); |
| 81 | + let i_proto = TBinaryInputProtocol::new(i_chan, true); |
| 82 | + let o_proto = TBinaryOutputProtocol::new(o_chan, true); |
| 83 | + let client = ThriftHiveMetastoreSyncClient::new(i_proto, o_proto); |
| 84 | + Ok(Self { |
| 85 | + config, |
| 86 | + client: HmsClient(Arc::new(Mutex::new(client))), |
| 87 | + }) |
| 88 | + } |
| 89 | +} |
| 90 | + |
| 91 | +/// Refer to <https://github.com/apache/iceberg/blob/main/hive-metastore/src/main/java/org/apache/iceberg/hive/HiveCatalog.java> for implementation details. |
| 92 | +#[async_trait] |
| 93 | +impl Catalog for HmsCatalog { |
| 94 | + /// HMS doesn't support nested namespaces. |
| 95 | + /// |
| 96 | + /// We will return empty list if parent is some. |
| 97 | + /// |
| 98 | + /// Align with java implementation: <https://github.com/apache/iceberg/blob/9bd62f79f8cd973c39d14e89163cb1c707470ed2/hive-metastore/src/main/java/org/apache/iceberg/hive/HiveCatalog.java#L305C26-L330> |
| 99 | + async fn list_namespaces( |
| 100 | + &self, |
| 101 | + parent: Option<&NamespaceIdent>, |
| 102 | + ) -> Result<Vec<NamespaceIdent>> { |
| 103 | + let dbs = if parent.is_some() { |
| 104 | + return Ok(vec![]); |
| 105 | + } else { |
| 106 | + self.client.call(|client| client.get_all_databases())? |
| 107 | + }; |
| 108 | + |
| 109 | + Ok(dbs.into_iter().map(NamespaceIdent::new).collect()) |
| 110 | + } |
| 111 | + |
| 112 | + async fn create_namespace( |
| 113 | + &self, |
| 114 | + _namespace: &NamespaceIdent, |
| 115 | + _properties: HashMap<String, String>, |
| 116 | + ) -> Result<Namespace> { |
| 117 | + todo!() |
| 118 | + } |
| 119 | + |
| 120 | + async fn get_namespace(&self, _namespace: &NamespaceIdent) -> Result<Namespace> { |
| 121 | + todo!() |
| 122 | + } |
| 123 | + |
| 124 | + async fn namespace_exists(&self, _namespace: &NamespaceIdent) -> Result<bool> { |
| 125 | + todo!() |
| 126 | + } |
| 127 | + |
| 128 | + async fn update_namespace( |
| 129 | + &self, |
| 130 | + _namespace: &NamespaceIdent, |
| 131 | + _properties: HashMap<String, String>, |
| 132 | + ) -> Result<()> { |
| 133 | + todo!() |
| 134 | + } |
| 135 | + |
| 136 | + async fn drop_namespace(&self, _namespace: &NamespaceIdent) -> Result<()> { |
| 137 | + todo!() |
| 138 | + } |
| 139 | + |
| 140 | + async fn list_tables(&self, _namespace: &NamespaceIdent) -> Result<Vec<TableIdent>> { |
| 141 | + todo!() |
| 142 | + } |
| 143 | + |
| 144 | + async fn create_table( |
| 145 | + &self, |
| 146 | + _namespace: &NamespaceIdent, |
| 147 | + _creation: TableCreation, |
| 148 | + ) -> Result<Table> { |
| 149 | + todo!() |
| 150 | + } |
| 151 | + |
| 152 | + async fn load_table(&self, _table: &TableIdent) -> Result<Table> { |
| 153 | + todo!() |
| 154 | + } |
| 155 | + |
| 156 | + async fn drop_table(&self, _table: &TableIdent) -> Result<()> { |
| 157 | + todo!() |
| 158 | + } |
| 159 | + |
| 160 | + async fn stat_table(&self, _table: &TableIdent) -> Result<bool> { |
| 161 | + todo!() |
| 162 | + } |
| 163 | + |
| 164 | + async fn rename_table(&self, _src: &TableIdent, _dest: &TableIdent) -> Result<()> { |
| 165 | + todo!() |
| 166 | + } |
| 167 | + |
| 168 | + async fn update_table(&self, _commit: TableCommit) -> Result<Table> { |
| 169 | + todo!() |
| 170 | + } |
| 171 | +} |
0 commit comments