catalog/system_schema/information_schema/
table_constraints.rs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
// Copyright 2023 Greptime Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

use std::sync::{Arc, Weak};

use arrow_schema::SchemaRef as ArrowSchemaRef;
use common_catalog::consts::INFORMATION_SCHEMA_TABLE_CONSTRAINTS_TABLE_ID;
use common_error::ext::BoxedError;
use common_recordbatch::adapter::RecordBatchStreamAdapter;
use common_recordbatch::{RecordBatch, SendableRecordBatchStream};
use datafusion::execution::TaskContext;
use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
use datafusion::physical_plan::streaming::PartitionStream as DfPartitionStream;
use datafusion::physical_plan::SendableRecordBatchStream as DfSendableRecordBatchStream;
use datatypes::prelude::{ConcreteDataType, MutableVector};
use datatypes::scalars::ScalarVectorBuilder;
use datatypes::schema::{ColumnSchema, Schema, SchemaRef};
use datatypes::value::Value;
use datatypes::vectors::{ConstantVector, StringVector, StringVectorBuilder, VectorRef};
use futures::TryStreamExt;
use snafu::{OptionExt, ResultExt};
use store_api::storage::{ScanRequest, TableId};

use super::{InformationTable, TABLE_CONSTRAINTS};
use crate::error::{
    CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
};
use crate::information_schema::key_column_usage::{
    PRI_CONSTRAINT_NAME, TIME_INDEX_CONSTRAINT_NAME,
};
use crate::information_schema::Predicates;
use crate::CatalogManager;

/// The `TABLE_CONSTRAINTS` table describes which tables have constraints.
pub(super) struct InformationSchemaTableConstraints {
    schema: SchemaRef,
    catalog_name: String,
    catalog_manager: Weak<dyn CatalogManager>,
}

const CONSTRAINT_CATALOG: &str = "constraint_catalog";
const CONSTRAINT_SCHEMA: &str = "constraint_schema";
const CONSTRAINT_NAME: &str = "constraint_name";
const TABLE_SCHEMA: &str = "table_schema";
const TABLE_NAME: &str = "table_name";
const CONSTRAINT_TYPE: &str = "constraint_type";
const ENFORCED: &str = "enforced";

const INIT_CAPACITY: usize = 42;

const TIME_INDEX_CONSTRAINT_TYPE: &str = "TIME INDEX";
const PRI_KEY_CONSTRAINT_TYPE: &str = "PRIMARY KEY";

impl InformationSchemaTableConstraints {
    pub(super) fn new(catalog_name: String, catalog_manager: Weak<dyn CatalogManager>) -> Self {
        Self {
            schema: Self::schema(),
            catalog_name,
            catalog_manager,
        }
    }

    fn schema() -> SchemaRef {
        Arc::new(Schema::new(vec![
            ColumnSchema::new(
                CONSTRAINT_CATALOG,
                ConcreteDataType::string_datatype(),
                false,
            ),
            ColumnSchema::new(
                CONSTRAINT_SCHEMA,
                ConcreteDataType::string_datatype(),
                false,
            ),
            ColumnSchema::new(CONSTRAINT_NAME, ConcreteDataType::string_datatype(), false),
            ColumnSchema::new(TABLE_SCHEMA, ConcreteDataType::string_datatype(), false),
            ColumnSchema::new(TABLE_NAME, ConcreteDataType::string_datatype(), false),
            ColumnSchema::new(CONSTRAINT_TYPE, ConcreteDataType::string_datatype(), false),
            ColumnSchema::new(ENFORCED, ConcreteDataType::string_datatype(), false),
        ]))
    }

    fn builder(&self) -> InformationSchemaTableConstraintsBuilder {
        InformationSchemaTableConstraintsBuilder::new(
            self.schema.clone(),
            self.catalog_name.clone(),
            self.catalog_manager.clone(),
        )
    }
}

impl InformationTable for InformationSchemaTableConstraints {
    fn table_id(&self) -> TableId {
        INFORMATION_SCHEMA_TABLE_CONSTRAINTS_TABLE_ID
    }

    fn table_name(&self) -> &'static str {
        TABLE_CONSTRAINTS
    }

    fn schema(&self) -> SchemaRef {
        self.schema.clone()
    }

    fn to_stream(&self, request: ScanRequest) -> Result<SendableRecordBatchStream> {
        let schema = self.schema.arrow_schema().clone();
        let mut builder = self.builder();
        let stream = Box::pin(DfRecordBatchStreamAdapter::new(
            schema,
            futures::stream::once(async move {
                builder
                    .make_table_constraints(Some(request))
                    .await
                    .map(|x| x.into_df_record_batch())
                    .map_err(Into::into)
            }),
        ));
        Ok(Box::pin(
            RecordBatchStreamAdapter::try_new(stream)
                .map_err(BoxedError::new)
                .context(InternalSnafu)?,
        ))
    }
}

struct InformationSchemaTableConstraintsBuilder {
    schema: SchemaRef,
    catalog_name: String,
    catalog_manager: Weak<dyn CatalogManager>,

    constraint_schemas: StringVectorBuilder,
    constraint_names: StringVectorBuilder,
    table_schemas: StringVectorBuilder,
    table_names: StringVectorBuilder,
    constraint_types: StringVectorBuilder,
}

impl InformationSchemaTableConstraintsBuilder {
    fn new(
        schema: SchemaRef,
        catalog_name: String,
        catalog_manager: Weak<dyn CatalogManager>,
    ) -> Self {
        Self {
            schema,
            catalog_name,
            catalog_manager,
            constraint_schemas: StringVectorBuilder::with_capacity(INIT_CAPACITY),
            constraint_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
            table_schemas: StringVectorBuilder::with_capacity(INIT_CAPACITY),
            table_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
            constraint_types: StringVectorBuilder::with_capacity(INIT_CAPACITY),
        }
    }

    /// Construct the `information_schema.table_constraints` virtual table
    async fn make_table_constraints(
        &mut self,
        request: Option<ScanRequest>,
    ) -> Result<RecordBatch> {
        let catalog_name = self.catalog_name.clone();
        let catalog_manager = self
            .catalog_manager
            .upgrade()
            .context(UpgradeWeakCatalogManagerRefSnafu)?;
        let predicates = Predicates::from_scan_request(&request);

        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, None);

            while let Some(table) = stream.try_next().await? {
                let keys = &table.table_info().meta.primary_key_indices;
                let schema = table.schema();

                if schema.timestamp_index().is_some() {
                    self.add_table_constraint(
                        &predicates,
                        &schema_name,
                        TIME_INDEX_CONSTRAINT_NAME,
                        &schema_name,
                        &table.table_info().name,
                        TIME_INDEX_CONSTRAINT_TYPE,
                    );
                }

                if !keys.is_empty() {
                    self.add_table_constraint(
                        &predicates,
                        &schema_name,
                        PRI_CONSTRAINT_NAME,
                        &schema_name,
                        &table.table_info().name,
                        PRI_KEY_CONSTRAINT_TYPE,
                    );
                }
            }
        }

        self.finish()
    }

    fn add_table_constraint(
        &mut self,
        predicates: &Predicates,
        constraint_schema: &str,
        constraint_name: &str,
        table_schema: &str,
        table_name: &str,
        constraint_type: &str,
    ) {
        let row = [
            (CONSTRAINT_SCHEMA, &Value::from(constraint_schema)),
            (CONSTRAINT_NAME, &Value::from(constraint_name)),
            (TABLE_SCHEMA, &Value::from(table_schema)),
            (TABLE_NAME, &Value::from(table_name)),
            (CONSTRAINT_TYPE, &Value::from(constraint_type)),
        ];

        if !predicates.eval(&row) {
            return;
        }

        self.constraint_schemas.push(Some(constraint_schema));
        self.constraint_names.push(Some(constraint_name));
        self.table_schemas.push(Some(table_schema));
        self.table_names.push(Some(table_name));
        self.constraint_types.push(Some(constraint_type));
    }

    fn finish(&mut self) -> Result<RecordBatch> {
        let rows_num = self.constraint_names.len();

        let constraint_catalogs = Arc::new(ConstantVector::new(
            Arc::new(StringVector::from(vec!["def"])),
            rows_num,
        ));
        let enforceds = Arc::new(ConstantVector::new(
            Arc::new(StringVector::from(vec!["YES"])),
            rows_num,
        ));

        let columns: Vec<VectorRef> = vec![
            constraint_catalogs,
            Arc::new(self.constraint_schemas.finish()),
            Arc::new(self.constraint_names.finish()),
            Arc::new(self.table_schemas.finish()),
            Arc::new(self.table_names.finish()),
            Arc::new(self.constraint_types.finish()),
            enforceds,
        ];

        RecordBatch::new(self.schema.clone(), columns).context(CreateRecordBatchSnafu)
    }
}

impl DfPartitionStream for InformationSchemaTableConstraints {
    fn schema(&self) -> &ArrowSchemaRef {
        self.schema.arrow_schema()
    }

    fn execute(&self, _: Arc<TaskContext>) -> DfSendableRecordBatchStream {
        let schema = self.schema.arrow_schema().clone();
        let mut builder = self.builder();
        Box::pin(DfRecordBatchStreamAdapter::new(
            schema,
            futures::stream::once(async move {
                builder
                    .make_table_constraints(None)
                    .await
                    .map(|x| x.into_df_record_batch())
                    .map_err(Into::into)
            }),
        ))
    }
}