Skip to content

Commit

Permalink
fix!: columns table in information_schema misses some columns (#3639)
Browse files Browse the repository at this point in the history 
* fix: columns table in information_schema misses some columns

* fix: test_information_schema_dot_columns

* fix: fuzz test

* feat: adds srs_id and refactor some columns with constant vector

* fix: test_information_schema_dot_columns

* chore: update comment

Co-authored-by: JeremyHi <[email protected]>

* build(deps): bump h2 from 0.3.24 to 0.3.26 (#3642)

Bumps [h2](https://github.com/hyperium/h2) from 0.3.24 to 0.3.26.
- [Release notes](https://github.com/hyperium/h2/releases)
- [Changelog](https://github.com/hyperium/h2/blob/v0.3.26/CHANGELOG.md)
- [Commits](hyperium/h2@v0.3.24...v0.3.26)

---
updated-dependencies:
- dependency-name: h2
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* build(deps): bump whoami from 1.4.1 to 1.5.1 (#3643)

Bumps [whoami](https://github.com/ardaku/whoami) from 1.4.1 to 1.5.1.
- [Changelog](https://github.com/ardaku/whoami/blob/v1/CHANGELOG.md)
- [Commits](ardaku/whoami@v1.4.1...v1.5.1)

---
updated-dependencies:
- dependency-name: whoami
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* feat: adding victoriametrics remote write (#3641)

* feat: adding victoria metrics remote write

* test: add e2e tests for prom and vm remote writes

* fix: construct correct pk list with pre-existing pk (#3614)

* fix: construct correct pk list with pre-existing pk

Signed-off-by: Ruihang Xia <[email protected]>

* update UT

Signed-off-by: Ruihang Xia <[email protected]>

---------

Signed-off-by: Ruihang Xia <[email protected]>

* test(sqlness): release databases after tests (#3648)

* refactor: rename Greptime_Type to Greptime_type

---------

Signed-off-by: dependabot[bot] <[email protected]>
Signed-off-by: Ruihang Xia <[email protected]>
Co-authored-by: JeremyHi <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Ning Sun <[email protected]>
Co-authored-by: Ruihang Xia <[email protected]>
Co-authored-by: Weny Xu <[email protected]>
  • Loading branch information
@killme2008 @fengjiachun
@dependabot @sunng87 @waynexia @WenyXu
6 people authored Apr 8, 2024
1 parent 7e8e9ab commit 87e0189
Show file tree
Hide file tree
Showing 10 changed files with 697 additions and 446 deletions.
1 change: 1 addition & 0 deletions Cargo.lock
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

1 change: 1 addition & 0 deletions src/catalog/Cargo.toml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,7 @@ prometheus.workspace = true
serde_json.workspace = true
session.workspace = true
snafu.workspace = true
sql.workspace = true
store-api.workspace = true
table.workspace = true
tokio.workspace = true
Expand Down
191 changes: 185 additions & 6 deletions src/catalog/src/information_schema/columns.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,13 +26,16 @@ use common_recordbatch::{RecordBatch, SendableRecordBatchStream};
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, DataType};
use datatypes::prelude::{ConcreteDataType, DataType, MutableVector};
use datatypes::scalars::ScalarVectorBuilder;
use datatypes::schema::{ColumnSchema, Schema, SchemaRef};
use datatypes::value::Value;
use datatypes::vectors::{StringVectorBuilder, VectorRef};
use datatypes::vectors::{
ConstantVector, Int64Vector, Int64VectorBuilder, StringVector, StringVectorBuilder, VectorRef,
};
use futures::TryStreamExt;
use snafu::{OptionExt, ResultExt};
use sql::statements;
use store_api::storage::{ScanRequest, TableId};

use super::{InformationTable, COLUMNS};
Expand All @@ -52,14 +55,38 @@ pub const TABLE_CATALOG: &str = "table_catalog";
pub const TABLE_SCHEMA: &str = "table_schema";
pub const TABLE_NAME: &str = "table_name";
pub const COLUMN_NAME: &str = "column_name";
const ORDINAL_POSITION: &str = "ordinal_position";
const CHARACTER_MAXIMUM_LENGTH: &str = "character_maximum_length";
const CHARACTER_OCTET_LENGTH: &str = "character_octet_length";
const NUMERIC_PRECISION: &str = "numeric_precision";
const NUMERIC_SCALE: &str = "numeric_scale";
const DATETIME_PRECISION: &str = "datetime_precision";
const CHARACTER_SET_NAME: &str = "character_set_name";
pub const COLLATION_NAME: &str = "collation_name";
pub const COLUMN_KEY: &str = "column_key";
pub const EXTRA: &str = "extra";
pub const PRIVILEGES: &str = "privileges";
const GENERATION_EXPRESSION: &str = "generation_expression";
// Extension field to keep greptime data type name
pub const GREPTIME_DATA_TYPE: &str = "greptime_data_type";
pub const DATA_TYPE: &str = "data_type";
pub const SEMANTIC_TYPE: &str = "semantic_type";
pub const COLUMN_DEFAULT: &str = "column_default";
pub const IS_NULLABLE: &str = "is_nullable";
const COLUMN_TYPE: &str = "column_type";
pub const COLUMN_COMMENT: &str = "column_comment";
const SRS_ID: &str = "srs_id";
const INIT_CAPACITY: usize = 42;

// The maximum length of string type
const MAX_STRING_LENGTH: i64 = 2147483647;
const UTF8_CHARSET_NAME: &str = "utf8";
const UTF8_COLLATE_NAME: &str = "utf8_bin";
const PRI_COLUMN_KEY: &str = "PRI";
const TIME_INDEX_COLUMN_KEY: &str = "TIME INDEX";
const DEFAULT_PRIVILEGES: &str = "select,insert";
const EMPTY_STR: &str = "";

impl InformationSchemaColumns {
pub(super) fn new(catalog_name: String, catalog_manager: Weak<dyn CatalogManager>) -> Self {
Self {
Expand All @@ -75,12 +102,46 @@ impl InformationSchemaColumns {
ColumnSchema::new(TABLE_SCHEMA, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(TABLE_NAME, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(COLUMN_NAME, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(ORDINAL_POSITION, ConcreteDataType::int64_datatype(), false),
ColumnSchema::new(
CHARACTER_MAXIMUM_LENGTH,
ConcreteDataType::int64_datatype(),
true,
),
ColumnSchema::new(
CHARACTER_OCTET_LENGTH,
ConcreteDataType::int64_datatype(),
true,
),
ColumnSchema::new(NUMERIC_PRECISION, ConcreteDataType::int64_datatype(), true),
ColumnSchema::new(NUMERIC_SCALE, ConcreteDataType::int64_datatype(), true),
ColumnSchema::new(DATETIME_PRECISION, ConcreteDataType::int64_datatype(), true),
ColumnSchema::new(
CHARACTER_SET_NAME,
ConcreteDataType::string_datatype(),
true,
),
ColumnSchema::new(COLLATION_NAME, ConcreteDataType::string_datatype(), true),
ColumnSchema::new(COLUMN_KEY, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(EXTRA, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(PRIVILEGES, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(
GENERATION_EXPRESSION,
ConcreteDataType::string_datatype(),
false,
),
ColumnSchema::new(
GREPTIME_DATA_TYPE,
ConcreteDataType::string_datatype(),
false,
),
ColumnSchema::new(DATA_TYPE, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(SEMANTIC_TYPE, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(COLUMN_DEFAULT, ConcreteDataType::string_datatype(), true),
ColumnSchema::new(IS_NULLABLE, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(COLUMN_TYPE, ConcreteDataType::string_datatype(), false),
ColumnSchema::new(COLUMN_COMMENT, ConcreteDataType::string_datatype(), true),
ColumnSchema::new(SRS_ID, ConcreteDataType::int64_datatype(), true),
]))
}

Expand Down Expand Up @@ -136,9 +197,18 @@ struct InformationSchemaColumnsBuilder {
schema_names: StringVectorBuilder,
table_names: StringVectorBuilder,
column_names: StringVectorBuilder,
ordinal_positions: Int64VectorBuilder,
character_maximum_lengths: Int64VectorBuilder,
character_octet_lengths: Int64VectorBuilder,
numeric_precisions: Int64VectorBuilder,
numeric_scales: Int64VectorBuilder,
datetime_precisions: Int64VectorBuilder,
character_set_names: StringVectorBuilder,
collation_names: StringVectorBuilder,
column_keys: StringVectorBuilder,
greptime_data_types: StringVectorBuilder,
data_types: StringVectorBuilder,
semantic_types: StringVectorBuilder,

column_defaults: StringVectorBuilder,
is_nullables: StringVectorBuilder,
column_types: StringVectorBuilder,
Expand All @@ -159,6 +229,16 @@ impl InformationSchemaColumnsBuilder {
schema_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
table_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
column_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
ordinal_positions: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
character_maximum_lengths: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
character_octet_lengths: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
numeric_precisions: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
numeric_scales: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
datetime_precisions: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
character_set_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
collation_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
column_keys: StringVectorBuilder::with_capacity(INIT_CAPACITY),
greptime_data_types: StringVectorBuilder::with_capacity(INIT_CAPACITY),
data_types: StringVectorBuilder::with_capacity(INIT_CAPACITY),
semantic_types: StringVectorBuilder::with_capacity(INIT_CAPACITY),
column_defaults: StringVectorBuilder::with_capacity(INIT_CAPACITY),
Expand Down Expand Up @@ -194,6 +274,7 @@ impl InformationSchemaColumnsBuilder {
};

self.add_column(
idx,
&predicates,
&catalog_name,
&schema_name,
Expand All @@ -208,16 +289,27 @@ impl InformationSchemaColumnsBuilder {
self.finish()
}

#[allow(clippy::too_many_arguments)]
fn add_column(
&mut self,
index: usize,
predicates: &Predicates,
catalog_name: &str,
schema_name: &str,
table_name: &str,
semantic_type: &str,
column_schema: &ColumnSchema,
) {
let data_type = &column_schema.data_type.name();
// Use sql data type name
let data_type = statements::concrete_data_type_to_sql_data_type(&column_schema.data_type)
.map(|dt| dt.to_string().to_lowercase())
.unwrap_or_else(|_| column_schema.data_type.name());

let column_key = match semantic_type {
SEMANTIC_TYPE_PRIMARY_KEY => PRI_COLUMN_KEY,
SEMANTIC_TYPE_TIME_INDEX => TIME_INDEX_COLUMN_KEY,
_ => EMPTY_STR,
};

let row = [
(TABLE_CATALOG, &Value::from(catalog_name)),
Expand All @@ -226,6 +318,8 @@ impl InformationSchemaColumnsBuilder {
(COLUMN_NAME, &Value::from(column_schema.name.as_str())),
(DATA_TYPE, &Value::from(data_type.as_str())),
(SEMANTIC_TYPE, &Value::from(semantic_type)),
(ORDINAL_POSITION, &Value::from((index + 1) as i64)),
(COLUMN_KEY, &Value::from(column_key)),
];

if !predicates.eval(&row) {
Expand All @@ -236,7 +330,63 @@ impl InformationSchemaColumnsBuilder {
self.schema_names.push(Some(schema_name));
self.table_names.push(Some(table_name));
self.column_names.push(Some(&column_schema.name));
self.data_types.push(Some(data_type));
// Starts from 1
self.ordinal_positions.push(Some((index + 1) as i64));

if column_schema.data_type.is_string() {
self.character_maximum_lengths.push(Some(MAX_STRING_LENGTH));
self.character_octet_lengths.push(Some(MAX_STRING_LENGTH));
self.numeric_precisions.push(None);
self.numeric_scales.push(None);
self.datetime_precisions.push(None);
self.character_set_names.push(Some(UTF8_CHARSET_NAME));
self.collation_names.push(Some(UTF8_COLLATE_NAME));
} else if column_schema.data_type.is_numeric() || column_schema.data_type.is_decimal() {
self.character_maximum_lengths.push(None);
self.character_octet_lengths.push(None);

self.numeric_precisions.push(
column_schema
.data_type
.numeric_precision()
.map(|x| x as i64),
);
self.numeric_scales
.push(column_schema.data_type.numeric_scale().map(|x| x as i64));

self.datetime_precisions.push(None);
self.character_set_names.push(None);
self.collation_names.push(None);
} else {
self.character_maximum_lengths.push(None);
self.character_octet_lengths.push(None);
self.numeric_precisions.push(None);
self.numeric_scales.push(None);

match &column_schema.data_type {
ConcreteDataType::DateTime(datetime_type) => {
self.datetime_precisions
.push(Some(datetime_type.precision() as i64));
}
ConcreteDataType::Timestamp(ts_type) => {
self.datetime_precisions
.push(Some(ts_type.precision() as i64));
}
ConcreteDataType::Time(time_type) => {
self.datetime_precisions
.push(Some(time_type.precision() as i64));
}
_ => self.datetime_precisions.push(None),
}

self.character_set_names.push(None);
self.collation_names.push(None);
}

self.column_keys.push(Some(column_key));
self.greptime_data_types
.push(Some(&column_schema.data_type.name()));
self.data_types.push(Some(&data_type));
self.semantic_types.push(Some(semantic_type));
self.column_defaults.push(
column_schema
Expand All @@ -249,23 +399,52 @@ impl InformationSchemaColumnsBuilder {
} else {
self.is_nullables.push(Some("No"));
}
self.column_types.push(Some(data_type));
self.column_types.push(Some(&data_type));
self.column_comments
.push(column_schema.column_comment().map(|x| x.as_ref()));
}

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

let privileges = Arc::new(ConstantVector::new(
Arc::new(StringVector::from(vec![DEFAULT_PRIVILEGES])),
rows_num,
));
let empty_string = Arc::new(ConstantVector::new(
Arc::new(StringVector::from(vec![EMPTY_STR])),
rows_num,
));
let srs_ids = Arc::new(ConstantVector::new(
Arc::new(Int64Vector::from(vec![None])),
rows_num,
));

let columns: Vec<VectorRef> = vec![
Arc::new(self.catalog_names.finish()),
Arc::new(self.schema_names.finish()),
Arc::new(self.table_names.finish()),
Arc::new(self.column_names.finish()),
Arc::new(self.ordinal_positions.finish()),
Arc::new(self.character_maximum_lengths.finish()),
Arc::new(self.character_octet_lengths.finish()),
Arc::new(self.numeric_precisions.finish()),
Arc::new(self.numeric_scales.finish()),
Arc::new(self.datetime_precisions.finish()),
Arc::new(self.character_set_names.finish()),
Arc::new(self.collation_names.finish()),
Arc::new(self.column_keys.finish()),
empty_string.clone(),
privileges,
empty_string,
Arc::new(self.greptime_data_types.finish()),
Arc::new(self.data_types.finish()),
Arc::new(self.semantic_types.finish()),
Arc::new(self.column_defaults.finish()),
Arc::new(self.is_nullables.finish()),
Arc::new(self.column_types.finish()),
Arc::new(self.column_comments.finish()),
srs_ids,
];

RecordBatch::new(self.schema.clone(), columns).context(CreateRecordBatchSnafu)
Expand Down
32 changes: 32 additions & 0 deletions src/datatypes/src/data_type.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -266,6 +266,38 @@ impl ConcreteDataType {
}
}

/// Try to get numeric precision, returns `None` if it's not numeric type
pub fn numeric_precision(&self) -> Option<u8> {
match self {
ConcreteDataType::Int8(_) | ConcreteDataType::UInt8(_) => Some(3),
ConcreteDataType::Int16(_) | ConcreteDataType::UInt16(_) => Some(5),
ConcreteDataType::Int32(_) | ConcreteDataType::UInt32(_) => Some(10),
ConcreteDataType::Int64(_) => Some(19),
ConcreteDataType::UInt64(_) => Some(20),
ConcreteDataType::Float32(_) => Some(12),
ConcreteDataType::Float64(_) => Some(22),
ConcreteDataType::Decimal128(decimal_type) => Some(decimal_type.precision()),
_ => None,
}
}

/// Try to get numeric scale, returns `None` if it's float or not numeric type
pub fn numeric_scale(&self) -> Option<i8> {
match self {
ConcreteDataType::Int8(_)
| ConcreteDataType::UInt8(_)
| ConcreteDataType::Int16(_)
| ConcreteDataType::UInt16(_)
| ConcreteDataType::Int32(_)
| ConcreteDataType::UInt32(_)
| ConcreteDataType::Int64(_)
| ConcreteDataType::UInt64(_) => Some(0),
ConcreteDataType::Float32(_) | ConcreteDataType::Float64(_) => None,
ConcreteDataType::Decimal128(decimal_type) => Some(decimal_type.scale()),
_ => None,
}
}

/// Try to cast data type as a [`TimeType`].
pub fn as_time(&self) -> Option<TimeType> {
match self {
Expand Down
7 changes: 7 additions & 0 deletions src/datatypes/src/types/datetime_type.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -23,10 +23,17 @@ use crate::prelude::{LogicalTypeId, MutableVector, ScalarVectorBuilder, Value, V
use crate::types::LogicalPrimitiveType;
use crate::vectors::{DateTimeVector, DateTimeVectorBuilder, PrimitiveVector};

const MILLISECOND_VARIATION: u64 = 3;
/// Data type for [`DateTime`].
#[derive(Debug, Default, Clone, PartialEq, Eq, Hash, PartialOrd, Ord, Serialize, Deserialize)]
pub struct DateTimeType;

impl DateTimeType {
pub fn precision(&self) -> u64 {
MILLISECOND_VARIATION
}
}

impl DataType for DateTimeType {
fn name(&self) -> String {
"DateTime".to_string()
Expand Down
Loading

0 comments on commit 87e0189

Please sign in to comment.