feat(redshift): column compression encodings and comments can now be customised

packages/@aws-cdk/aws-redshift/README.md

@@ -200,17 +200,32 @@ new Table(this, 'Table', {
 });
 ```
 
-Tables can also be configured with a comment:
+Tables and their respective columns can be configured to contain comments:
 
 ```ts fixture=cluster
 new Table(this, 'Table', {
   tableColumns: [
-    { name: 'col1', dataType: 'varchar(4)' }, 
-    { name: 'col2', dataType: 'float' }
+    { name: 'col1', dataType: 'varchar(4)', comment: 'This is a column comment' }, 
+    { name: 'col2', dataType: 'float', comment: 'This is a another column comment' }
+  ],
+  cluster: cluster,
+  databaseName: 'databaseName',
+  tableComment: 'This is a table comment',
+});
+```
+
+Table columns can be configured to use a specific compression encoding:
+
+```ts fixture=cluster
+import { ColumnEncoding } from '@aws-cdk/aws-redshift';
+
+new Table(this, 'Table', {
+  tableColumns: [
+    { name: 'col1', dataType: 'varchar(4)', encoding: ColumnEncoding.TEXT32K },
+    { name: 'col2', dataType: 'float', encoding: ColumnEncoding.DELTA32K },
   ],
   cluster: cluster,
   databaseName: 'databaseName',
-  comment: 'This is a comment',
 });
 ```
 
@@ -369,6 +384,8 @@ cluster.addToParameterGroup('enable_user_activity_logging', 'true');
 In most cases, existing clusters [must be manually rebooted](https://docs.aws.amazon.com/redshift/latest/mgmt/working-with-parameter-groups.html) to apply parameter changes. You can automate parameter related reboots by setting the cluster's `rebootForParameterChanges` property to `true` , or by using `Cluster.enableRebootForParameterChanges()`.
 
 ```ts
+import * as ec2 from '@aws-cdk/aws-ec2';
+import * as cdk from '@aws-cdk/core';
 declare const vpc: ec2.Vpc;
 
 const cluster = new Cluster(this, 'Cluster', {
@@ -451,14 +468,16 @@ Some Amazon Redshift features require Amazon Redshift to access other AWS servic
 When you create an IAM role and set it as the default for the cluster using console, you don't have to provide the IAM role's Amazon Resource Name (ARN) to perform authentication and authorization.
 
 ```ts
+import * as ec2 from '@aws-cdk/aws-ec2';
+import * as iam from '@aws-cdk/aws-iam';
 declare const vpc: ec2.Vpc;
 
 const defaultRole = new iam.Role(this, 'DefaultRole', {
   assumedBy: new iam.ServicePrincipal('redshift.amazonaws.com'),
 },
 );
 
-new Cluster(stack, 'Redshift', {
+new Cluster(this, 'Redshift', {
     masterUser: {
       masterUsername: 'admin',
     },
@@ -471,14 +490,16 @@ new Cluster(stack, 'Redshift', {
 A default role can also be added to a cluster using the `addDefaultIamRole` method.
 
 ```ts
+import * as ec2 from '@aws-cdk/aws-ec2';
+import * as iam from '@aws-cdk/aws-iam';
 declare const vpc: ec2.Vpc;
 
 const defaultRole = new iam.Role(this, 'DefaultRole', {
   assumedBy: new iam.ServicePrincipal('redshift.amazonaws.com'),
 },
 );
 
-const redshiftCluster = new Cluster(stack, 'Redshift', {
+const redshiftCluster = new Cluster(this, 'Redshift', {
     masterUser: {
       masterUsername: 'admin',
     },
@@ -494,6 +515,8 @@ redshiftCluster.addDefaultIamRole(defaultRole);
 Attaching IAM roles to a Redshift Cluster grants permissions to the Redshift service to perform actions on your behalf.
 
 ```ts
+import * as ec2 from '@aws-cdk/aws-ec2';
+import * as iam from '@aws-cdk/aws-iam';
 declare const vpc: ec2.Vpc
 
 const role = new iam.Role(this, 'Role', {
@@ -511,6 +534,8 @@ const cluster = new Cluster(this, 'Redshift', {
 Additional IAM roles can be attached to a cluster using the `addIamRole` method.
 
 ```ts
+import * as ec2 from '@aws-cdk/aws-ec2';
+import * as iam from '@aws-cdk/aws-iam';
 declare const vpc: ec2.Vpc
 
 const role = new iam.Role(this, 'Role', {

packages/@aws-cdk/aws-redshift/lib/private/database-query-provider/table.ts

@@ -42,7 +42,7 @@ async function createTable(
   tableAndClusterProps: TableAndClusterProps,
 ): Promise<string> {
   const tableName = tableNamePrefix + tableNameSuffix;
-  const tableColumnsString = tableColumns.map(column => `${column.name} ${column.dataType}`).join();
+  const tableColumnsString = tableColumns.map(column => `${column.name} ${column.dataType}${getEncodingColumnString(column)}`).join();
 
   let statement = `CREATE TABLE ${tableName} (${tableColumnsString})`;
 
@@ -63,6 +63,11 @@ async function createTable(
 
   await executeStatement(statement, tableAndClusterProps);
 
+  for (const column of tableColumns) {
+    if (column.comment) {
+      await executeStatement(`COMMENT ON COLUMN ${tableName}.${column.name} IS '${column.comment}'`, tableAndClusterProps);
+    }
+  }
   if (tableAndClusterProps.tableComment) {
     await executeStatement(`COMMENT ON TABLE ${tableName} IS '${tableAndClusterProps.tableComment}'`, tableAndClusterProps);
   }
@@ -120,6 +125,20 @@ async function updateTable(
     alterationStatements.push(...columnAdditions.map(addition => `ALTER TABLE ${tableName} ${addition}`));
   }
 
+  const columnEncoding = tableColumns.filter(column => {
+    return oldTableColumns.some(oldColumn => column.name === oldColumn.name && column.encoding !== oldColumn.encoding);
+  }).map(column => `ALTER COLUMN ${column.name} ENCODE ${column.encoding || 'AUTO'}`);
+  if (columnEncoding.length > 0) {
+    alterationStatements.push(`ALTER TABLE ${tableName} ${columnEncoding.join(', ')}`);
+  }
+
+  const columnComments = tableColumns.filter(column => {
+    return oldTableColumns.some(oldColumn => column.name === oldColumn.name && column.comment !== oldColumn.comment);
+  }).map(column => `COMMENT ON COLUMN ${tableName}.${column.name} IS ${column.comment ? `'${column.comment}'` : 'NULL'}`);
+  if (columnComments.length > 0) {
+    alterationStatements.push(...columnComments);
+  }
+
   if (useColumnIds) {
     const columnNameUpdates = tableColumns.reduce((updates, column) => {
       const oldColumn = oldTableColumns.find(oldCol => oldCol.id && oldCol.id === column.id);
@@ -190,3 +209,10 @@ async function updateTable(
 function getSortKeyColumnsString(sortKeyColumns: Column[]) {
   return sortKeyColumns.map(column => column.name).join();
 }
+
+function getEncodingColumnString(column: Column): string {
+  if (column.encoding) {
+    return ` ENCODE ${column.encoding}`;
+  }
+  return '';
+}

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

@@ -90,6 +90,20 @@ export interface Column {
    * @default - column is not a SORTKEY
    */
   readonly sortKey?: boolean;
+
+  /**
+   * The encoding to use for the column.
+   *
+   * @default - Amazon Redshift determines the encoding based on the data type.
+   */
+  readonly encoding?: ColumnEncoding;
+
+  /**
+   * A comment to attach to the column.
+   *
+   * @default - no comment
+   */
+  readonly comment?: string;
 }
 
 /**
@@ -371,3 +385,112 @@ export enum TableSortStyle {
    */
   INTERLEAVED = 'INTERLEAVED',
 }
+
+/**
+ * The compression encoding of a column.
+ *
+ * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Compression_encodings.html
+ */
+export enum ColumnEncoding {
+  /**
+   * Amazon Redshift assigns an optimal encoding based on the column data.
+   * This is the default.
+   */
+  AUTO = 'AUTO',
+
+  /**
+   * The column is not compressed.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Raw_encoding.html
+   */
+  RAW = 'RAW',
+
+  /**
+   * The column is compressed using the AZ64 algorithm.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/az64-encoding.html
+   */
+  AZ64 = 'AZ64',
+
+  /**
+   * The column is compressed using a separate dictionary for each block column value on disk.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Byte_dictionary_encoding.html
+   */
+  BYTEDICT = 'BYTEDICT',
+
+  /**
+   * The column is compressed based on the difference between values in the column.
+   * This records differences as 1-byte values.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Delta_encoding.html
+   */
+  DELTA = 'DELTA',
+
+  /**
+   * The column is compressed based on the difference between values in the column.
+   * This records differences as 2-byte values.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Delta_encoding.html
+   */
+  DELTA32K = 'DELTA32K',
+
+  /**
+   * The column is compressed using the LZO algorithm.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/lzo-encoding.html
+   */
+  LZO = 'LZO',
+
+  /**
+   * The column is compressed to a smaller storage size than the original data type.
+   * The compressed storage size is 1 byte.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_MostlyN_encoding.html
+   */
+  MOSTLY8 = 'MOSTLY8',
+
+  /**
+   * The column is compressed to a smaller storage size than the original data type.
+   * The compressed storage size is 2 bytes.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_MostlyN_encoding.html
+   */
+  MOSTLY16 = 'MOSTLY16',
+
+  /**
+   * The column is compressed to a smaller storage size than the original data type.
+   * The compressed storage size is 4 bytes.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_MostlyN_encoding.html
+   */
+  MOSTLY32 = 'MOSTLY32',
+
+  /**
+   * The column is compressed by recording the number of occurrences of each value in the column.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Runlength_encoding.html
+   */
+  RUNLENGTH = 'RUNLENGTH',
+
+  /**
+   * The column is compressed by recording the first 245 unique words and then using a 1-byte index to represent each word.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Text255_encoding.html
+   */
+  TEXT255 = 'TEXT255',
+
+  /**
+   * The column is compressed by recording the first 32K unique words and then using a 2-byte index to represent each word.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/c_Text255_encoding.html
+   */
+  TEXT32K = 'TEXT32K',
+
+  /**
+   * The column is compressed using the ZSTD algorithm.
+   *
+   * @see https://docs.aws.amazon.com/redshift/latest/dg/zstd-encoding.html
+   */
+  ZSTD = 'ZSTD',
+}