sql-query-identifier (Go)

This is a Go port of the NPM sql-query-identifier library. It identifies the type of each statement in a SQL query string, providing the start and end position of each statement.

Features

• Multi-Statement Support: Parses a string containing multiple SQL statements separated by semicolons.
• Dialect-Aware: Supports multiple SQL dialects, including MySQL, PostgreSQL, MSSQL, and more.
• Statement Identification: Identifies a wide range of statement types (e.g., SELECT, INSERT, CREATE TABLE).
• Execution Type Analysis: Classifies statements by their behavior (LISTING, MODIFICATION).
• Parameter Extraction: Identifies positional (?, $1) and named (:name) parameters.
• Strict & Non-Strict Modes: Choose whether to error on unknown statement types or classify them as UNKNOWN.

Installation

go get github.com/Infisical/sql-query-identifier

Usage

The following example demonstrates how to parse a query string and print the identified statements as a JSON object.

package main

import (
	"encoding/json"
	"fmt"
	"log"

	sqlqueryidentifier "github.com/Infisical/sql-query-identifier"
)

func main() {
	query := `
		INSERT INTO Persons (PersonID, Name) VALUES (1, 'Jack');
		SELECT * FROM Persons;
`

	// Use Dialect constants for type safety
	dialect := sqlqueryidentifier.DialectMySQL
	strict := false

	options := sqlqueryidentifier.IdentifyOptions{
		Dialect: &dialect,
		Strict:  &strict,
	}

	results, err := sqlqueryidentifier.Identify(query, options)
	if err != nil {
		log.Fatalf("Failed to identify query: %v", err)
	}

	// Marshal the results to JSON for clear output
	jsonOutput, err := json.MarshalIndent(results, "", "  ")
	if err != nil {
		log.Fatalf("Failed to marshal results to JSON: %v", err)
	}

	fmt.Println(string(jsonOutput))
}

API

Identify(query string, options IdentifyOptions) ([]IdentifyResult, error)

• query (string): The raw SQL string to be processed.
• options (IdentifyOptions): Configuration for the parser.
  • Strict (*bool): If false, will classify unknown statements as UNKNOWN instead of returning an error. Defaults to true.
  • Dialect (*Dialect): The SQL dialect to use for parsing. Defaults to generic.

Supported Dialects

• mssql
• sqlite
• mysql
• oracle
• psql
• bigquery
• generic (default)

Supported Statement Types

Data Manipulation

• SELECT
• INSERT
• UPDATE
• DELETE
• TRUNCATE

Data Definition

• CREATE_DATABASE
• CREATE_SCHEMA
• CREATE_TABLE
• CREATE_VIEW
• CREATE_TRIGGER
• CREATE_FUNCTION
• CREATE_INDEX
• CREATE_PROCEDURE
• DROP_DATABASE
• DROP_SCHEMA
• DROP_TABLE
• DROP_VIEW
• DROP_TRIGGER
• DROP_FUNCTION
• DROP_INDEX
• DROP_PROCEDURE
• ALTER_DATABASE
• ALTER_SCHEMA
• ALTER_TABLE
• ALTER_VIEW
• ALTER_TRIGGER
• ALTER_FUNCTION
• ALTER_INDEX
• ALTER_PROCEDURE

SHOW (MySQL and generic dialects)

• SHOW_BINARY
• SHOW_BINLOG
• SHOW_CHARACTER
• SHOW_COLLATION
• SHOW_COLUMNS
• SHOW_CREATE
• SHOW_DATABASES
• SHOW_ENGINE
• SHOW_ENGINES
• SHOW_ERRORS
• SHOW_EVENTS
• SHOW_FUNCTION
• SHOW_GRANTS
• SHOW_INDEX
• SHOW_MASTER
• SHOW_OPEN
• SHOW_PLUGINS
• SHOW_PRIVILEGES
• SHOW_PROCEDURE
• SHOW_PROCESSLIST
• SHOW_PROFILE
• SHOW_PROFILES
• SHOW_RELAYLOG
• SHOW_REPLICAS
• SHOW_SLAVE
• SHOW_REPLICA
• SHOW_STATUS
• SHOW_TABLE
• SHOW_TABLES
• SHOW_TRIGGERS
• SHOW_VARIABLES
• SHOW_WARNINGS

Other

• ANON_BLOCK (BigQuery and Oracle dialects only)
• UNKNOWN (only available if strict mode is disabled)

Execution Types

Execution types classify the behavior of a query.

• LISTING: The query lists or retrieves data.
• MODIFICATION: The query modifies the database structure or data.
• INFORMATION: The query shows information, such as profiling data.
• ANON_BLOCK: The query is an anonymous block which may contain multiple statements.
• UNKNOWN: The query type could not be determined (only available if strict mode is disabled).

How It Works

This library uses AST and parser techniques to identify the SQL query type. It does not validate the entire query; instead, it validates only the required tokens to identify the statement type.

The identification process is:

1. Tokenizing: The input string is broken down into tokens (keywords, strings, comments, etc.).
2. Parsing: The stream of tokens is parsed to identify the statement boundaries and types.
   • Comments and string contents are ignored to prevent false positives.
   • Keywords are expected at the beginning of a statement.
   • Semicolons are used to identify the end of a statement.

Because the library does not perform a full SQL validation, it is recommended to use it on queries that have already been successfully executed by a SQL client.

License

This project is licensed under the MIT License.