rawsql-ts 0.1.0-beta.1

package/.vs/CopilotSnapshots/AD2203F1B9FECE44BB6D5DC5D83E87AA/734AB1B2974F9545B7F12AAA5842524B/840203FFA9CCF5B33DFD5C7CC5B13527

@@ -0,0 +1,125 @@
+import { BaseTokenReader } from "./BaseTokenReader";
+import { TokenType } from "../enums/tokenType";
+import { Lexeme } from "../models/Lexeme";
+import { KeywordTrie } from "../models/KeywordTrie";
+import { KeywordParser } from "../KeywordParser";
+
+// 専用のパーサーが必要になるものをコマンドと呼ぶ。
+// また複数語で構成されるキーワードもコマンドとする。
+// 例外は「型」。型はユーザー定義の型もあるため、正確に把握できないため、Identifier として扱っている。
+const trie = new KeywordTrie([
+    ["with"],
+    ["recursive"],
+    ["materialized"],
+    ["not", "materialized"],
+    ["select"],
+    ["from"],
+    ["distinct"],
+    ["distinct", "on"],
+    ["where"],
+    ["group", "by"],
+    ["having"],
+    ["order", "by"],
+    ["limit"],
+    ["for"],
+    ["offset"],
+    // set operations
+    ["union"],
+    ["union", "all"],
+    ["intersect"],
+    ["intersect", "all"],
+    ["except"],
+    ["except", "all"],
+    // between and
+    ["beteen"],
+    // window functions
+    ["window"],
+    ["over"],
+    ["partition", "by"],
+    ["range"],
+    ["range", "between"],
+    ["rows"],
+    ["rows", "between"],
+    ["groups"],
+    ["groups", "between"],
+    // table join commands
+    ["on"],
+    ["using"],
+    ["natural"],
+    ["join"],
+    ["inner", "join"],
+    ["cross", "join"],
+    ["left", "join"],
+    ["left", "outer", "join"],
+    ["right", "join"],
+    ["right", "outer", "join"],
+    ["full", "join"],
+    ["lateral"],
+    // trim
+    ["leading"],
+    ["trailing"],
+    ["both"],
+    ["both", "from"],
+    ["leading", "from"],
+    ["trailing", "from"],
+    ["both", "leading"],
+    ["both", "trailing"],    
+    // date, time, extract
+    ["year", "from"],
+    ["month", "from"],
+    ["day", "from"],
+    ["hour", "from"],
+    ["minute", "from"],
+    ["second", "from"],
+    ["dow", "from"],
+    ["doy", "from"],
+    ["isodow", "from"],
+    ["quarter", "from"],
+    ["week", "from"],
+    ["epoch", "from"],
+    ["at", "time", "zone"],
+    ["interval"],
+    // case 
+    ["case"],
+    ["case", "when"],
+    ["then"],
+    ["else"],
+    ["end"],
+    // escape
+    ["escape"],
+    ["uescape"],
+    // others
+    ["insert", "into"],
+    ["update"],
+    ["delete", "from"],
+    ["merge", "into"],
+    ["matched"],
+    ["not", "matched"],
+    ["update", "set"],
+    ["do", "nothing"],
+    ["values"],
+    ["set"],
+    ["returning"],
+    ["create", "table"],
+    ["create", "temporary", "table"],
+    ["tablesample"],
+]);
+const parser = new KeywordParser(trie);
+
+export class CommandTokenReader extends BaseTokenReader {
+
+    public tryRead(previous: Lexeme | null): Lexeme | null {
+        if (this.isEndOfInput()) {
+            return null;
+        }
+
+        // Check for keyword identifiers
+        const keyword = parser.parse(this.input, this.position);
+        if (keyword !== null) {
+            this.position = keyword.newPosition;
+            return this.createLexeme(TokenType.Identifier, keyword.keyword);
+        }
+
+        return null;
+    }
+}

package/.vs/CopilotSnapshots/AD2203F1B9FECE44BB6D5DC5D83E87AA/state.mpack

@@ -0,0 +1 @@
+��SessionId��$f10322ad-feb9-44ce-bb6d-5dc5d83e87aa��Id�4Microsoft.VisualStudio.Conversations.Chat.HelpWindow�CorrelationStates���CorrelationId�$b2b14a73-4f97-4595-b7f1-2aaa5842524b�Snapshots�� 840203FFA9CCF5B33DFD5C7CC5B13527��SavedFileName١C:\Users\mssgm\Documents\GitHub\ts-sample\.vs\CopilotSnapshots\AD2203F1B9FECE44BB6D5DC5D83E87AA\734AB1B2974F9545B7F12AAA5842524B\840203FFA9CCF5B33DFD5C7CC5B13527�OriginalFilePath�PC:\Users\mssgm\Documents\GitHub\ts-sample\src\tokenReaders\CommandTokenReader.ts�NewFileMarker�

package/.vs/CopilotSnapshots/AEC950C2FAD20147AEE8E325E2CA0A78/89CF3F527D0687479D14274E8CDF3DFF/1CF9939611B0D194AC81951A171DF046

@@ -0,0 +1,30 @@
+import { BaseTokenReader } from './BaseTokenReader';
+import { TokenType } from '../enums/tokenType';
+import { Lexeme } from '../models/Lexeme';
+
+const operatorChars = new Set([
+    '+', '-', '*', '/', '%', // Arithmetic operators
+    '~', '@', '#', '$', '^', '&', // Special symbols
+    ':', // Postgres type cast operator
+    '!', '<', '>', '=', '|'// Other symbols
+]);
+
+export class OperatorTokenReader extends BaseTokenReader {
+    public tryRead(previous: Lexeme | null): Lexeme | null {
+        if (this.isEndOfInput()) {
+            return null;
+        }
+
+        const char = this.input[this.position];
+
+        if (operatorChars.has(char)) {
+            const start = this.position;
+            while (this.canRead() && operatorChars.has(this.input[this.position])) {
+                this.position++;
+            }
+            return this.createLexeme(TokenType.Operator, this.input.slice(start, this.position));
+        }
+
+        return null;
+    }
+}

package/.vs/CopilotSnapshots/AEC950C2FAD20147AEE8E325E2CA0A78/89CF3F527D0687479D14274E8CDF3DFF/BA5D2B145A59265D63FA3B3584344B02

@@ -0,0 +1,158 @@
+import { TokenType } from "../src/enums/tokenType";
+import { SqlTokenizer } from "../src/sqlTokenizer";
+
+test('tokenizes addition operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1+1');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('+');
+});
+
+test('tokenizes multiplication operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1*2');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('*');
+});
+
+test('tokenizes division operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1/3');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('/');
+});
+
+test('tokenizes modulus operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1%4');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('%');
+});
+
+test('tokenizes type convert operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1::text');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('::');
+});
+
+test('tokenizes equality operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1==1');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('==');
+});
+
+test('tokenizes inequality operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1!=2');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('!=');
+});
+
+test('tokenizes inequality operator　<>', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1<>2');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('<>');
+});
+
+test('tokenizes less than operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1<2');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('<');
+});
+
+test('tokenizes greater than operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1>0');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('>');
+});
+
+test('tokenizes less than or equal operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1<=2');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('<=');
+});
+
+test('tokenizes greater than or equal operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1>=0');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('>=');
+});

package/.vs/CopilotSnapshots/AEC950C2FAD20147AEE8E325E2CA0A78/E495EBF64DC77A4BBD470B92DB0391C4/1CF9939611B0D194AC81951A171DF046

@@ -0,0 +1,30 @@
+import { BaseTokenReader } from './BaseTokenReader';
+import { TokenType } from '../enums/tokenType';
+import { Lexeme } from '../models/Lexeme';
+
+const operatorChars = new Set([
+    '+', '-', '*', '/', '%', // Arithmetic operators
+    '~', '@', '#', '$', '^', '&', // Special symbols
+    ':', // Postgres type cast operator
+    '!', '<', '>', '=', '|'// Other symbols
+]);
+
+export class OperatorTokenReader extends BaseTokenReader {
+    public tryRead(previous: Lexeme | null): Lexeme | null {
+        if (this.isEndOfInput()) {
+            return null;
+        }
+
+        const char = this.input[this.position];
+
+        if (operatorChars.has(char)) {
+            const start = this.position;
+            while (this.canRead() && operatorChars.has(this.input[this.position])) {
+                this.position++;
+            }
+            return this.createLexeme(TokenType.Operator, this.input.slice(start, this.position));
+        }
+
+        return null;
+    }
+}

package/.vs/CopilotSnapshots/AEC950C2FAD20147AEE8E325E2CA0A78/E495EBF64DC77A4BBD470B92DB0391C4/BA5D2B145A59265D63FA3B3584344B02

@@ -0,0 +1,67 @@
+import { TokenType } from "../src/enums/tokenType";
+import { SqlTokenizer } from "../src/sqlTokenizer";
+
+test('tokenizes addition operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1+1');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('+');
+});
+
+test('tokenizes multiplication operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1*2');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('*');
+});
+
+test('tokenizes division operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1/3');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('/');
+});
+
+test('tokenizes modulus operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1%4');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('%');
+});
+
+test('tokenizes type convert operator', () => {
+    // Arrange
+    const tokenizer = new SqlTokenizer('1::text');
+
+    // Act
+    const tokens = tokenizer.readLexmes();
+
+    // Assert
+    expect(tokens.length).toBe(3);
+    expect(tokens[1].type).toBe(TokenType.Operator);
+    expect(tokens[1].value).toBe('::');
+});

package/.vs/CopilotSnapshots/AEC950C2FAD20147AEE8E325E2CA0A78/state.mpack

@@ -0,0 +1 @@
+��SessionId��$c250c9ae-d2fa-4701-aee8-e325e2ca0a78��Id�4Microsoft.VisualStudio.Conversations.Chat.HelpWindow�CorrelationStates���CorrelationId�$f6eb95e4-c74d-4b7a-bd47-0b92db0391c4�Snapshots�� BA5D2B145A59265D63FA3B3584344B02��SavedFileName١C:\Users\mssgm\Documents\GitHub\ts-sample\.vs\CopilotSnapshots\AEC950C2FAD20147AEE8E325E2CA0A78\E495EBF64DC77A4BBD470B92DB0391C4\BA5D2B145A59265D63FA3B3584344B02�OriginalFilePath�@C:\Users\mssgm\Documents\GitHub\ts-sample\tests\operator.test.ts�NewFileMarker�� 1CF9939611B0D194AC81951A171DF046��SavedFileName١C:\Users\mssgm\Documents\GitHub\ts-sample\.vs\CopilotSnapshots\AEC950C2FAD20147AEE8E325E2CA0A78\E495EBF64DC77A4BBD470B92DB0391C4\1CF9939611B0D194AC81951A171DF046�OriginalFilePath�QC:\Users\mssgm\Documents\GitHub\ts-sample\src\tokenReaders\OperatorTokenReader.ts�NewFileMarker���CorrelationId�$523fcf89-067d-4787-9d14-274e8cdf3dff�Snapshots�� BA5D2B145A59265D63FA3B3584344B02��SavedFileName١C:\Users\mssgm\Documents\GitHub\ts-sample\.vs\CopilotSnapshots\AEC950C2FAD20147AEE8E325E2CA0A78\89CF3F527D0687479D14274E8CDF3DFF\BA5D2B145A59265D63FA3B3584344B02�OriginalFilePath�@C:\Users\mssgm\Documents\GitHub\ts-sample\tests\operator.test.ts�NewFileMarker�� 1CF9939611B0D194AC81951A171DF046��SavedFileName١C:\Users\mssgm\Documents\GitHub\ts-sample\.vs\CopilotSnapshots\AEC950C2FAD20147AEE8E325E2CA0A78\89CF3F527D0687479D14274E8CDF3DFF\1CF9939611B0D194AC81951A171DF046�OriginalFilePath�QC:\Users\mssgm\Documents\GitHub\ts-sample\src\tokenReaders\OperatorTokenReader.ts�NewFileMarker�

package/.vs/CopilotSnapshots/AF947096A34634478C5D084B8442CD81/1FCDCC65B992954DB78BD1F46892312B/44EB81A1663981E3F52FFFECE3A2918E

@@ -0,0 +1,143 @@
+import { StringUtils } from "./stringUtils";
+
+export enum KeywordMatchResult {
+    NotAKeyword,     // "キーワードではない"
+    PartialOnly,     // "途中（ここで終わることはない）"
+    PartialOrFinal,  // "途中または終端（ここで止まることはある）"
+    Final            // "終端（これ以上長いキーワードはない）"
+}
+
+export class KeywordTrieReader {
+    private input: string;
+    private position: number;
+    private trie: KeywordTrie;
+
+    constructor(input: string, position: number, trie: KeywordTrie) {
+        this.input = input;
+        this.position = position;
+        this.trie = trie;
+        this.trie.reset();
+    }
+
+    private isEndOfInput(shift: number = 0): boolean {
+        return this.position + shift >= this.input.length;
+    }
+
+    private canRead(shift: number = 0): boolean {
+        return !this.isEndOfInput(shift);
+    }
+
+    public readKeyword(): { keyword: string, newPosition: number } | null {
+        if (this.isEndOfInput()) {
+            return null;
+        }
+
+        const result = StringUtils.tryReadRegularIdentifier(this.input, this.position);
+
+        if (result === null) {
+            return null;
+        }
+
+        let matchResult = this.trie.pushLexeme(result.identifier);
+
+        if (matchResult === KeywordMatchResult.NotAKeyword) {
+            return null;
+        }
+
+        if (matchResult === KeywordMatchResult.Final) {
+            this.position = result.newPosition;
+            return {
+                keyword: result.identifier,
+                newPosition: this.position
+            };
+        }
+
+        // multi-word keyword
+        let lexeme = result.identifier;
+
+        this.position = StringUtils.skipWhiteSpacesAndComments(this.input, result.newPosition);
+
+        while (this.canRead()) {
+            const previousMatchResult = matchResult;
+
+            const result = StringUtils.tryReadRegularIdentifier(this.input, this.position);
+
+            if (result !== null) {
+                matchResult = this.trie.pushLexeme(result.identifier);
+
+                if (matchResult === KeywordMatchResult.NotAKeyword) {
+                    if (previousMatchResult === KeywordMatchResult.PartialOrFinal) {
+                        break;
+                    } else {
+                        return null;
+                    }
+                }
+
+                lexeme += ' ' + result.identifier;
+                this.position = StringUtils.skipWhiteSpacesAndComments(this.input, result.newPosition);
+
+                if (matchResult === KeywordMatchResult.Final) {
+                    break;
+                }
+            } else if (previousMatchResult === KeywordMatchResult.PartialOrFinal) {
+                break;
+            } else {
+                return null;
+            }
+        }
+        return {
+            keyword: lexeme,
+            newPosition: this.position
+        };
+    }
+}
+
+export class KeywordTrie {
+    private root: Map<string, any> = new Map();
+    private currentNode: Map<string, any>;
+
+    constructor(keywords: string[][]) {
+        // initialize root node
+        for (const keyword of keywords) {
+            this.addKeyword(keyword);
+        }
+        // set current node to root
+        this.currentNode = this.root;
+    }
+
+    private addKeyword(keyword: string[]) {
+        let node = this.root;
+        for (const word of keyword) {
+            if (!node.has(word)) {
+                node.set(word, new Map());
+            }
+            node = node.get(word);
+        }
+        node.set("__end__", true);
+    }
+
+    public reset(): void {
+        this.currentNode = this.root
+    }
+
+    public pushLexeme(lexeme: string): KeywordMatchResult {
+        if (!this.currentNode.has(lexeme)) {
+            return KeywordMatchResult.NotAKeyword;
+        }
+
+        // move to next node
+        this.currentNode = this.currentNode.get(lexeme);
+
+        const hasEnd = this.currentNode.has("__end__");
+        const hasMore = this.currentNode.size > (hasEnd ? 1 : 0);
+
+        if (hasEnd && !hasMore) {
+            return KeywordMatchResult.Final;
+        }
+        if (hasEnd && hasMore) {
+            return KeywordMatchResult.PartialOrFinal;
+        }
+
+        return KeywordMatchResult.PartialOnly;
+    }
+}