feat: Athena Databases and Tables

packages/@aws-cdk/aws-athena/lib/database.ts

@@ -0,0 +1,120 @@
+import { PolicyStatement } from '@aws-cdk/aws-iam';
+import { IBucket } from '@aws-cdk/aws-s3';
+import { Aws, Construct, IConstruct } from '@aws-cdk/core';
+import { AwsCustomResource } from '@aws-cdk/custom-resources';
+
+/**
+ * Creates a Table with the name and the parameters that you specify.
+ */
+export interface IDatabase extends IConstruct {
+  /**
+   * The Database name.
+   * @attribute
+   */
+  readonly databaseName: string;
+}
+
+/**
+ * Properties used to define an Athena Database
+ */
+export interface DatabaseProps {
+  /**
+   * The name for the Database.
+   * @default none
+   */
+  readonly databaseName: string;
+
+  /**
+   * Establishes the metadata value for the built-in metadata property named `comment` and the value
+   * you provide for comment.
+   * @default none
+   */
+  readonly comment: string;
+
+  /**
+   * Specifies the location where database files and metastore will exist. The location must be an
+   * Amazon S3 location.
+   */
+   readonly locationBucket: IBucket;
+
+  /**
+   * The path part of the S3 Location.
+   * @default none
+   */
+  readonly locationPath?: string;
+
+  /**
+   * Allows you to specify custom metadata properties for the database definition.
+   * @default none
+   */
+  readonly databaseProperties?: Map<string, string>;
+}
+
+/**
+ * An Athena Database
+ */
+export class Database extends Construct implements IDatabase {
+  /**
+   * some docstring
+   */
+  public readonly databaseName: string;
+
+  constructor(scope: Construct, id: string, props: DatabaseProps) {
+  super(scope, id);
+  this.databaseName = props.databaseName;
+
+  const s3Policy = new PolicyStatement();
+  s3Policy.addActions('s3:*');
+  s3Policy.addResources(props.locationBucket.arnForObjects('*'));
+  s3Policy.addResources(props.locationBucket.bucketArn);
+
+  const athenaPolicy = new PolicyStatement();
+  athenaPolicy.addActions('athena:StartQueryExecution');
+  athenaPolicy.addAllResources();
+
+  const gluePolicy = new PolicyStatement();
+  gluePolicy.addActions('glue:GetDatabase', 'glue:GetTable');
+  athenaPolicy.addResources(`arn:aws:glue:${Aws.REGION}:${Aws.ACCOUNT_ID}:catalog`);
+
+  new AwsCustomResource(this, 'CreateAthenaDatabase', {
+    onCreate: {
+      service: 'Athena',
+      action: 'startQueryExecution',
+      parameters: {
+        QueryString: queryStringBuilder(props),
+        ResultConfiguration: {
+          OutputLocation: `s3://${props.locationBucket.bucketName}/`
+        },
+      },
+      physicalResourceIdPath: 'QueryExecutionId'
+    },
+    policyStatements: [s3Policy, athenaPolicy, gluePolicy]
+  });
+  }
+}
+
+function queryStringBuilder(props: DatabaseProps) {
+  let queryString = `CREATE DATABASE ${props.databaseName}`;
+  if (props.comment) {
+    queryString += ` COMMENT ${props.comment}`;
+  }
+  queryString += ` LOCATION 's3://${props.locationBucket.bucketName}/`;
+  if (props.locationPath) {
+    queryString += `${props.locationPath}/`;
+  }
+  queryString += "'";
+  if (props.databaseProperties) {
+    const propString = propertiesStringBuilder(props.databaseProperties);
+    queryString += ` ${propString}`;
+  }
+  queryString += ";";
+  return queryString;
+}
+
+function propertiesStringBuilder(databaseProperties: Map<string, string>) {
+  const tempStrings: string[] = [];
+  databaseProperties.forEach((value: string, key: string) => {
+    tempStrings.push(`'${key}'='${value}'`);
+  });
+  return `(${tempStrings.join(", ")})`;
+}

packages/@aws-cdk/aws-athena/lib/datatypes.ts

@@ -0,0 +1,87 @@
+/**
+ * Supported Datatypes in Athena
+ */
+export enum DataType {
+  /**
+   * Values are true and false.
+   */
+  BOOLEAN = 'BOOLEAN',
+  /**
+   * A 8-bit signed INTEGER in two’s complement format, with a minimum value of -2^7 and a maximum value of 2^7-1.
+   */
+  TINYINT = 'TINYINT',
+  /**
+   * A 16-bit signed INTEGER in two’s complement format, with a minimum value of -2^15 and a maximum value of 2^15-1.
+   */
+  SMALLINT = 'SMALLINT',
+  /**
+   * Athena combines two different implementations of the INTEGER data type. In Data Definition Language (DDL)
+   * queries, Athena uses the INT data type. In all other queries, Athena uses the INTEGER data type, where INTEGER
+   * is represented as a 32-bit signed value in two's complement format, with a minimum value of-2^31 and a maximum
+   * value of 2^31-1. In the JDBC driver, INTEGER is returned, to ensure compatibility with business analytics
+   * applications.
+   */
+  INT = 'INT',
+  /**
+   * A 64-bit signed INTEGER in two’s complement format, with a minimum value of -2^63 and a maximum value of 2^63-1.
+   */
+  BIGINT = 'BIGINT',
+  /**
+   * Floating-point types
+   */
+  DOUBLE = 'DOUBLE',
+  /**
+   * Floating-point types
+   */
+  FLOAT = 'FLOAT',
+  /**
+   * TODO
+   * [ (precision, scale) ], where precision is the total number of digits, and scale (optional) is the number of
+   * digits in fractional part, the default is 0. For example, use these type definitions: DECIMAL(11,5), DECIMAL(15).
+   * To specify decimal values as literals, such as when selecting rows with a specific decimal value in a query DDL
+   * expression, specify the DECIMAL type definition, and list the decimal value as a literal (in single quotes) in
+   * your query, as in this example: decimal_value = DECIMAL '0.12'.
+   */
+  DECIMAL = 'DECIMALTODO',
+  /**
+   * Fixed length character data, with a specified length between 1 and 255, such as char(10).
+   * For more information, see CHAR Hive Data Type.
+   * https://cwiki.apache.org/confluence/display/Hive/LanguageManual+Types#LanguageManualTypes-char
+   */
+  CHAR = 'CHAR',
+  /**
+   * Variable length character data, with a specified length between 1 and 65535, such as varchar(10).
+   * For more information, see VARCHAR Hive Data Type.
+   * https://cwiki.apache.org/confluence/display/Hive/LanguageManual+Types#LanguageManualTypes-varchar
+   */
+  VARCHAR = 'VARCHAR',
+  /**
+   * Not Officially documented, but clearly supported by Athena :shrug emoji:
+   */
+  STRING = 'STRING',
+  /**
+   * (for data in Parquet)
+   */
+  BINARY = 'BINARY',
+  /**
+   * DATE, in the UNIX format, such as YYYY-MM-DD.
+   */
+  DATE = 'DATE',
+  /**
+   * Instant in time and date in the UNiX format, such as yyyy-mm-dd hh:mm:ss[.f...].
+   * For example, TIMESTAMP '2008-09-15 03:04:05.324'. This format uses the session time zone.
+   */
+  TIMESTAMP = 'TIMESTAMP',
+  /**
+   * < data_type >
+   */
+  ARRAY = 'ARRAYTODO',
+  /**
+   * < primitive_type, data_type >
+   */
+  MAP = 'MAPTODO',
+  /**
+   * < col_name : data_type [COMMENT col_comment] [, ...] >
+   */
+  STRUCT = 'STRUCTTODO'
+}

packages/@aws-cdk/aws-athena/lib/index.ts

@@ -1,2 +1,4 @@
+export * from './table';
+export * from './datatypes';
 // AWS::Athena CloudFormation Resources:
-export * from './athena.generated';
+export * from './athena.generated';

packages/@aws-cdk/aws-athena/lib/table.ts

@@ -0,0 +1,102 @@
+import { PolicyStatement } from '@aws-cdk/aws-iam';
+import { IBucket } from '@aws-cdk/aws-s3';
+import { Aws, Construct, IConstruct } from '@aws-cdk/core';
+import { AwsCustomResource } from '@aws-cdk/custom-resources';
+import { DataType } from './datatypes';
+
+/**
+ * Creates a Table with the name and the parameters that you specify.
+ */
+export interface ITable extends IConstruct {
+  /**
+   * The Table name.
+   * @attribute
+   */
+  readonly tableName: string;
+}
+
+/**
+ * Properties used to define an Athena Table
+ */
+export interface TableProps {
+  /**
+   * The database to add the Table to.
+   * @default none
+   */
+  readonly databaseName: string;
+
+  /**
+   * The name for the Table.
+   * @default none
+   */
+  readonly tableName: string;
+
+  /**
+   * Specifies the name for each column to be created, along with the column's data type.
+   * Column names do not allow special characters other than underscore (_).
+   * If col_name begins with an underscore, enclose the column name in backticks, for example `_mycolumn`.
+   */
+   readonly schema: {[key: string]: DataType}
+
+  /**
+   * Table comment.
+   * @default none
+   */
+  readonly comment?: string;
+
+  /**
+   * The Bucket that this Table databases.
+   * @default none
+   */
+  readonly queryBucket: IBucket;
+}
+
+/**
+ * An Athena Table
+ */
+export class Table extends Construct implements ITable {
+  /**
+   * some docstring
+   */
+  public readonly tableName: string;
+
+  constructor(scope: Construct, id: string, props: TableProps) {
+  super(scope, id);
+  this.tableName = props.tableName;
+
+  const s3Policy = new PolicyStatement();
+  s3Policy.addActions('s3:*');
+  s3Policy.addResources(props.queryBucket.arnForObjects('*'));
+  s3Policy.addResources(props.queryBucket.bucketArn);
+
+  const athenaPolicy = new PolicyStatement();
+  athenaPolicy.addActions('athena:StartQueryExecution');
+  athenaPolicy.addAllResources();
+
+  const gluePolicy = new PolicyStatement();
+  gluePolicy.addActions('glue:GetDatabase', 'glue:GetTable');
+  athenaPolicy.addResources(`arn:aws:glue:${Aws.REGION}:${Aws.ACCOUNT_ID}:catalog`);
+
+  new AwsCustomResource(this, 'CreateAthenaTable', {
+    onCreate: {
+      service: 'Athena',
+      action: 'startQueryExecution',
+      parameters: {
+        QueryString: `CREATE EXTERNAL TABLE IF NOT EXISTS ${props.databaseName}.${props.tableName} (${schemaStringBuilder(props.schema)}) LOCATION 's3://${props.queryBucket.bucketName}'/`,
+        ResultConfiguration: {
+        OutputLocation: `s3://${props.queryBucket.bucketName}/`
+        },
+      },
+      physicalResourceIdPath: 'QueryExecutionId'
+    },
+      policyStatements: [s3Policy, athenaPolicy, gluePolicy]
+  });
+  }
+}
+
+function schemaStringBuilder(schema: {[key: string]: DataType}) {
+  const tempStrings: string[] = [];
+  tempStrings.concat(Object.keys(schema)
+    .map(key => `'${key}' '${schema[key]}'`));
+  return `(${tempStrings.join(", ")})`;
+}

packages/@aws-cdk/aws-athena/package.json

@@ -68,14 +68,18 @@
     "pkglint": "1.18.0"
   },
   "dependencies": {
-    "@aws-cdk/core": "1.18.0"
+    "@aws-cdk/aws-s3": "1.18.0",
+    "@aws-cdk/core": "1.18.0",
+    "@aws-cdk/custom-resources": "1.18.0"
   },
   "homepage": "https://github.com/aws/aws-cdk",
   "peerDependencies": {
-    "@aws-cdk/core": "1.18.0"
+    "@aws-cdk/aws-s3": "1.18.0",
+    "@aws-cdk/core": "1.18.0",
+    "@aws-cdk/custom-resources": "1.18.0"
   },
   "engines": {
     "node": ">= 10.3.0"
   },
   "stability": "experimental"
-}
+}