prettier-sql/lib/core/Tokenizer.d.ts

Format whitespace in a SQL query to make it more readable

import * as regexFactory from './regexFactory';
import { Token, TokenType } from './token';
/** Struct that defines how a SQL language can be broken into tokens */
interface TokenizerOptions {
    reservedKeywords: string[];
    reservedCommands: string[];
    reservedLogicalOperators: string[];
    reservedDependentClauses: string[];
    reservedBinaryCommands: string[];
    stringTypes: regexFactory.StringPatternType[];
    blockStart: string[];
    blockEnd: string[];
    indexedPlaceholderTypes?: string[];
    namedPlaceholderTypes: string[];
    lineCommentTypes: string[];
    specialWordChars?: {
        prefix?: string;
        any?: string;
        suffix?: string;
    };
    operators?: string[];
}
/** Converts SQL language string into a token stream */
export default class Tokenizer {
    WHITESPACE_REGEX: RegExp;
    REGEX_MAP: {
        [tokenType in TokenType]: RegExp;
    };
    INDEXED_PLACEHOLDER_REGEX?: RegExp;
    IDENT_NAMED_PLACEHOLDER_REGEX?: RegExp;
    STRING_NAMED_PLACEHOLDER_REGEX?: RegExp;
    /**
     * @param {TokenizerOptions} cfg
     *  @param {string[]} cfg.reservedKeywords - Reserved words in SQL
     *  @param {string[]} cfg.reservedDependentClauses - Words that following a specific Statement and must have data attached
     *  @param {string[]} cfg.reservedLogicalOperators - Words that are set to newline
     *  @param {string[]} cfg.reservedCommands - Words that are set to new line separately
     *  @param {string[]} cfg.reservedBinaryCommands - Words that are top level but have no indentation
     *  @param {string[]} cfg.stringTypes - string types to enable - "", '', ``, [], N''
     *  @param {string[]} cfg.blockStart - Opening parentheses to enable, like (, [
     *  @param {string[]} cfg.blockEnd - Closing parentheses to enable, like ), ]
     *  @param {string[]} cfg.indexedPlaceholderTypes - Prefixes for indexed placeholders, like ?
     *  @param {string[]} cfg.namedPlaceholderTypes - Prefixes for named placeholders, like @ and :
     *  @param {string[]} cfg.lineCommentTypes - Line comments to enable, like # and --
     *  @param {string[]} cfg.specialWordChars - Special chars that can be found inside of words, like @ and #
     *  @param {string[]} cfg.operators - Additional operators to recognize
     */
    constructor(cfg: TokenizerOptions);
    /**
     * Takes a SQL string and breaks it into tokens.
     * Each token is an object with type and value.
     *
     * @param {string} input - The SQL string
     * @returns {Token[]} output token stream
     */
    tokenize(input: string): Token[];
    /** Matches preceding whitespace if present */
    getWhitespace(input: string): string;
    /** Curried function of `getTokenOnFirstMatch` that allows token type to be passed first */
    matchToken: (tokenType: TokenType) => (input: string) => Token | undefined;
    /** Attempts to match next token from input string, tests RegExp patterns in decreasing priority */
    getNextToken(input: string, previousToken?: Token): Token;
    /**
     * Attempts to match a placeholder token pattern
     * @return {Token | undefined} - The placeholder token if found, otherwise undefined
     */
    getPlaceholderToken(input: string): Token | undefined;
    getEscapedPlaceholderKey({ key, quoteChar }: {
        key: string;
        quoteChar: string;
    }): string;
    /**
     * Attempts to match a Reserved word token pattern, avoiding edge cases of Reserved words within string tokens
     * @return {Token | undefined} - The Reserved word token if found, otherwise undefined
     */
    getReservedWordToken(input: string, previousToken?: Token): Token | undefined;
    /**
     * Attempts to match RegExp from head of input, returning undefined if not found
     * @param {string} _.input - The string to match
     * @param {TokenType} _.type - The type of token to match against
     * @param {RegExp} _.regex - The regex to match
     * @return {Token | undefined} - The matched token if found, otherwise undefined
     */
    getTokenOnFirstMatch({ input, type, regex, }: {
        input: string;
        type: TokenType;
        regex: RegExp;
    }): Token | undefined;
}
export {};