Add experimental dataTypeCase and functionCase options

README.md

@@ -151,6 +151,8 @@ All fields are optional and all fields that are not specified will be filled wit
 - [**`useTabs`**](docs/useTabs.md) to use tabs for indentation.
 - [**`keywordCase`**](docs/keywordCase.md) uppercases or lowercases keywords.
 - [**`identifierCase`**](docs/identifierCase.md) uppercases or lowercases identifiers. (**experimental!**)
+- [**`dataTypeCase`**](docs/dataTypeCase.md) uppercases or lowercases data types. (**experimental!**)
+- [**`functionCase`**](docs/functionCase.md) uppercases or lowercases function names. (**experimental!**)
 - [**`indentStyle`**](docs/indentStyle.md) defines overall indentation style.
 - [**`logicalOperatorNewline`**](docs/logicalOperatorNewline.md) newline before or after boolean operator (AND, OR, XOR).
 - [**`expressionWidth`**](docs/expressionWidth.md) maximum number of characters in parenthesized expressions to be kept on single line.

docs/dataTypeCase.md

@@ -0,0 +1,52 @@
+# dataTypeCase (experimental)
+
+Converts data types to upper- or lowercase.
+
+Caveat: Only supported by languages which export `dataTypes` from their `.keywords.ts` file (eg. `bigquery`, `postgresql` and others)
+
+Note: Casing of function names like `VARCHAR(30)` are not modified - instead rely on the `functionCase` option for this.
+
+## Options
+
+- `"preserve"` (default) preserves the original case.
+- `"upper"` converts to uppercase.
+- `"lower"` converts to lowercase.
+
+### preserve
+
+```sql
+CREATE TABLE
+  users (
+    id InTeGeR PRIMARY KEY GENERATED ALWAYS AS IDENTITY,
+    first_name VarChaR(30) NOT NULL,
+    bio teXT,
+    is_email_verified BooL NOT NULL DEFAULT FALSE,
+    created_timestamp timestamPtz NOT NULL DEFAULT NOW()
+  )
+```
+
+### upper
+
+```sql
+CREATE TABLE
+  users (
+    id INTEGER PRIMARY KEY GENERATED ALWAYS AS IDENTITY,
+    first_name VarChaR(30) NOT NULL,
+    bio TEXT,
+    is_email_verified BOOL NOT NULL DEFAULT FALSE,
+    created_timestamp TIMESTAMPTZ NOT NULL DEFAULT NOW()
+  )
+```
+
+### lower
+
+```sql
+CREATE TABLE
+  users (
+    id integer PRIMARY KEY GENERATED ALWAYS AS IDENTITY,
+    first_name VarChaR(30) NOT NULL,
+    bio text,
+    is_email_verified bool NOT NULL DEFAULT FALSE,
+    created_timestamp timestamptz NOT NULL DEFAULT NOW()
+  )
+```

docs/functionCase.md

@@ -0,0 +1,48 @@
+# functionCase (experimental)
+
+Converts function names to upper- or lowercase.
+
+## Options
+
+- `"preserve"` (default) preserves the original case.
+- `"upper"` converts to uppercase.
+- `"lower"` converts to lowercase.
+
+### preserve
+
+```sql
+CREATE TABLE
+  users (
+    id iNtegeR PRIMARY KEY GENERATED ALWAYS AS IDENTITY,
+    first_name VarChaR(30) NOT NULL,
+    bio TEXT,
+    is_email_verified BOOL NOT NULL DEFAULT FALSE,
+    created_timestamp TIMESTAMPTZ NOT NULL DEFAULT NoW()
+  )
+```
+
+### upper
+
+```sql
+CREATE TABLE
+  users (
+    id iNtegeR PRIMARY KEY GENERATED ALWAYS AS IDENTITY,
+    first_name VARCHAR(30) NOT NULL,
+    bio TEXT,
+    is_email_verified BOOL NOT NULL DEFAULT FALSE,
+    created_timestamp TIMESTAMPTZ NOT NULL DEFAULT NOW()
+  )
+```
+
+### lower
+
+```sql
+CREATE TABLE
+  users (
+    id iNtegeR PRIMARY KEY GENERATED ALWAYS AS IDENTITY,
+    first_name varchar(30) NOT NULL,
+    bio TEXT,
+    is_email_verified BOOL NOT NULL DEFAULT FALSE,
+    created_timestamp TIMESTAMPTZ NOT NULL DEFAULT now()
+  )
+```

src/FormatOptions.ts

@@ -8,13 +8,19 @@ export type KeywordCase = 'preserve' | 'upper' | 'lower';
 
 export type IdentifierCase = 'preserve' | 'upper' | 'lower';
 
+export type DataTypeCase = 'preserve' | 'upper' | 'lower';
+
+export type FunctionCase = 'preserve' | 'upper' | 'lower';
+
 export type LogicalOperatorNewline = 'before' | 'after';
 
 export interface FormatOptions {
   tabWidth: number;
   useTabs: boolean;
   keywordCase: KeywordCase;
   identifierCase: IdentifierCase;
+  dataTypeCase: DataTypeCase;
+  functionCase: FunctionCase;
   indentStyle: IndentStyle;
   logicalOperatorNewline: LogicalOperatorNewline;
   expressionWidth: number;

src/formatter/ExpressionFormatter.ts

@@ -28,6 +28,7 @@ import {
   CaseExpressionNode,
   CaseWhenNode,
   CaseElseNode,
+  DataTypeNode,
 } from '../parser/ast.js';
 
 import Layout, { WS } from './Layout.js';
@@ -130,26 +131,39 @@ export default class ExpressionFormatter {
         return this.formatLineComment(node);
       case NodeType.block_comment:
         return this.formatBlockComment(node);
+      case NodeType.data_type:
+        return this.formatDataType(node);
       case NodeType.keyword:
         return this.formatKeywordNode(node);
     }
   }
 
   private formatFunctionCall(node: FunctionCallNode) {
     this.withComments(node.nameKw, () => {
-      this.layout.add(this.showKw(node.nameKw));
+      this.layout.add(this.showFunctionKw(node.nameKw));
     });
     this.formatNode(node.parenthesis);
   }
 
   private formatArraySubscript(node: ArraySubscriptNode) {
+    let formattedArray: string;
+
+    switch (node.array.type) {
+      case NodeType.data_type:
+        formattedArray = this.showDataType(node.array);
+        break;
+      case NodeType.keyword:
+        formattedArray = this.showKw(node.array);
+        break;
+      default:
+        formattedArray = this.showIdentifier(node.array);
+        break;
+    }
+
     this.withComments(node.array, () => {
-      this.layout.add(
-        node.array.type === NodeType.keyword
-          ? this.showKw(node.array)
-          : this.showIdentifier(node.array)
-      );
+      this.layout.add(formattedArray);
     });
+
     this.formatNode(node.parenthesis);
   }
 
@@ -489,6 +503,10 @@ export default class ExpressionFormatter {
     }
   }
 
+  private formatDataType(node: DataTypeNode) {
+    this.layout.add(this.showDataType(node), WS.SPACE);
+  }
+
   private showKw(node: KeywordNode): string {
     if (isTabularToken(node.tokenType)) {
       return toTabularFormat(this.showNonTabularKw(node), this.cfg.indentStyle);
@@ -509,6 +527,26 @@ export default class ExpressionFormatter {
     }
   }
 
+  private showFunctionKw(node: KeywordNode): string {
+    if (isTabularToken(node.tokenType)) {
+      return toTabularFormat(this.showNonTabularFunctionKw(node), this.cfg.indentStyle);
+    } else {
+      return this.showNonTabularFunctionKw(node);
+    }
+  }
+
+  // Like showFunctionKw(), but skips tabular formatting
+  private showNonTabularFunctionKw(node: KeywordNode): string {
+    switch (this.cfg.functionCase) {
+      case 'preserve':
+        return equalizeWhitespace(node.raw);
+      case 'upper':
+        return node.text;
+      case 'lower':
+        return node.text.toLowerCase();
+    }
+  }
+
   private showIdentifier(node: IdentifierNode): string {
     if (node.quoted) {
       return node.text;
@@ -523,4 +561,15 @@ export default class ExpressionFormatter {
       }
     }
   }
+
+  private showDataType(node: DataTypeNode): string {
+    switch (this.cfg.dataTypeCase) {
+      case 'preserve':
+        return equalizeWhitespace(node.raw);
+      case 'upper':
+        return node.text;
+      case 'lower':
+        return node.text.toLowerCase();
+    }
+  }
 }

src/languages/bigquery/bigquery.formatter.ts

@@ -163,9 +163,8 @@ export const bigquery: DialectOptions = {
     reservedSetOperations,
     reservedJoins,
     reservedPhrases,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     extraParens: ['[]'],
     stringTypes: [

src/languages/db2/db2.formatter.ts

@@ -265,9 +265,8 @@ export const db2: DialectOptions = {
     reservedSetOperations,
     reservedJoins,
     reservedPhrases,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     extraParens: ['[]'],
     stringTypes: [

src/languages/db2/db2.keywords.ts

@@ -403,12 +403,17 @@ export const keywords: string[] = [
 
 export const dataTypes: string[] = [
   // https://www.ibm.com/docs/en/db2-for-zos/12?topic=columns-data-types
+  'ARRAY',
+  'BIGINT',
   'CCSID',
   'CHAR',
   'CHARACTER',
   'DATE',
   'DOUBLE',
+  'INT',
+  'INTEGER',
   'LONG',
+  'SMALLINT',
   'TIME',
   'TIMESTAMP',
 ];

src/languages/db2i/db2i.formatter.ts

@@ -161,9 +161,8 @@ export const db2i: DialectOptions = {
     reservedSetOperations,
     reservedJoins,
     reservedPhrases,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     nestedBlockComments: true,
     extraParens: ['[]'],

src/languages/db2i/db2i.keywords.ts

@@ -498,6 +498,7 @@ export const keywords: string[] = [
 
 export const dataTypes: string[] = [
   // https://www.ibm.com/docs/en/i/7.2?topic=iaodsd-odbc-data-types-how-they-correspond-db2-i-database-types
+  'ARRAY',
   'BIGINT',
   'BINARY',
   'BIT',
@@ -516,6 +517,7 @@ export const dataTypes: string[] = [
   'DOUBLE',
   'FLOAT',
   'GRAPHIC',
+  'INT',
   'INTEGER',
   'LONG',
   'NUMERIC',

src/languages/hive/hive.formatter.ts

@@ -90,9 +90,8 @@ export const hive: DialectOptions = {
     reservedSetOperations,
     reservedJoins,
     reservedPhrases,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     extraParens: ['[]'],
     stringTypes: ['""-bs', "''-bs"],

src/languages/mariadb/mariadb.formatter.ts

@@ -273,9 +273,8 @@ export const mariadb: DialectOptions = {
     reservedJoins,
     reservedPhrases,
     supportsXor: true,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     // TODO: support _ char set prefixes such as _utf8, _latin1, _binary, _utf8mb4, etc.
     stringTypes: [

src/languages/mysql/mysql.formatter.ts

@@ -240,9 +240,8 @@ export const mysql: DialectOptions = {
     reservedJoins,
     reservedPhrases,
     supportsXor: true,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     // TODO: support _ char set prefixes such as _utf8, _latin1, _binary, _utf8mb4, etc.
     stringTypes: [

src/languages/mysql/mysql.keywords.ts

@@ -235,8 +235,12 @@ export const dataTypes: string[] = [
   'BIGINT', // (R)
   'BINARY', // (R)
   'BLOB', // (R)
+  'BOOL', // (R)
+  'BOOLEAN', // (R)
   'CHAR', // (R)
   'CHARACTER', // (R)
+  'DATE', // (R)
+  'DATETIME', // (R)
   'DEC', // (R)
   'DECIMAL', // (R)
   'DOUBLE', // (R)
@@ -260,6 +264,7 @@ export const dataTypes: string[] = [
   'PRECISION', // (R)
   'REAL', // (R)
   'SMALLINT', // (R)
+  'TIMESTAMP', // (R)
   'TINYBLOB', // (R)
   'TINYINT', // (R)
   'TINYTEXT', // (R)

src/languages/n1ql/n1ql.formatter.ts

@@ -92,9 +92,8 @@ export const n1ql: DialectOptions = {
     reservedJoins,
     reservedPhrases,
     supportsXor: true,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     // NOTE: single quotes are actually not supported in N1QL,
     // but we support them anyway as all other SQL dialects do,

src/languages/plsql/plsql.formatter.ts

@@ -90,9 +90,8 @@ export const plsql: DialectOptions = {
     reservedJoins,
     reservedPhrases,
     supportsXor: true,
-    reservedKeywords:
-      // Temporary, will be replaced by reservedDataTypes
-      [...new Set(keywords.concat(dataTypes))],
+    reservedKeywords: keywords,
+    reservedDataTypes: dataTypes,
     reservedFunctionNames: functions,
     stringTypes: [
       { quote: "''-qq", prefixes: ['N'] },