prep-composer is a lightweight JavaScript utility that facilitates convenient and safe composition of SQL statements by automatically managing and escaping parameters.
- Inline Parameter Values: Simplifies SQL writing by automatically managing the order and number of parameter values. No placeholders and separate parameter list.
- Modular SQL Composition: Allows the creation of complex SQL statements through independent, reusable segments.
- Identifier Escaping: Supports escaping of identifiers for enhanced security.
- Flexible Integration: Compatible with established escaping utilities like sqlstring, and supports
?
placeholders for prepared statements. - Dialect Compatibility: Designed to work with all SQL dialects; identifier escaping is currently optimized for MySQL and compatible databases using `backtick` escaping.
const {sql, $} = require('prep-composer');
const SqlString = require('sqlstring'); // optional
const name = "O'Brien";
const newJobTitle = 'Senior Developer';
const updateTable = sql('UPDATE', $['my_db']['employees']); // identifier escaping is optional
const fieldsToUpdate = {
job_title: 'Designer',
salary: 200000,
};
const condition = sql('name =', $(name));
const query = sql(updateTable, 'SET', $(fieldsToUpdate), 'WHERE', condition);
console.log(query.toString());
// UPDATE `my_db`.`employees` SET `job_title` = ?, `salary` = ? WHERE name = ?
console.log(query.parameters);
// [ "Designer", 200000, "O'Brien" ]
console.log(query.toString(SqlString.escape));
// UPDATE `my_db`.`employees` SET `job_title` = 'Designer', `salary` = 200000 WHERE name = 'O\'Brien'
You can install prep-composer using npm:
npm install prep-composer
Creates a new SqlSegment
composed of the provided fragments. Fragments can be strings, literals,
identifiers, or values. Raw SQL is typically represented by a string
fragment. It will be turned into a Literal
internally. An array of fragments is flattened into the SQL segment, separated by commas.
Type definition for an SQL fragment. A fragment can be a Literal
, a string
(will be turned into a Literal
),
a Value
, an Identifier
or an SqlSegment
.
Helper function to create a new Value
fragment for the given value. Values represent the variable fragments of
a statement that are later escaped or replaced with placeholders.
If the passed value is an array, it is turned into
an array of SqlSegment
s by converting all the elements to values if they are not of type Literal
or SqlSegment
.
This is practical for creating IN
conditions.
If an object is passed it will be turned into a series of escaped-key=escaped-value assignments. If values are already
literals, values or identifiers, they are used as is. So to set a field to a static raw SQL value, you have to wrap
the value in an sql
function call (e.g. { hire_date: sql('NOW()') }
) otherwise it is interpreted as a parameter
that has to be escaped.
Helper to create a new Identifier
fragment for the given identifier. The identifier is immediately escaped
with backticks when passed to the sql
function. Identifiers can be chained: $['db']['field']
(yields `db`.`field`
). It is recommended to use the square
brackets syntax $[<identifier>]
for consistency. Dots within identifiers are preserved ($['db.field']
yields `db.field`
).
Represents a composite SQL segment composed of multiple fragments. A segment can be used as a part of another segment, allowing the creation of complex SQL statements.
Members:
toString(escapeFn?: (value: any) => string): string
: Returns the SQL statement as a string. IfescapeFn
is provided, it will be used to escape values. Otherwise?
placeholders are output.parameters: any[]
: An array of parameter values extracted from the SQL segment.append(fragment: Fragment | Fragment[]): SqlSegment
: Returns a newSqlSegment
by appending the provided fragment(s) to the current segment. As in thesql
function, an array of fragments is flattened into the SQL and separated by commas. Spaces are automatically added between fragments if required.
This project is licensed under the MIT License. See the LICENSE file for details.
Feel free to contribute to this project by submitting issues or pull requests. Your feedback and contributions are welcome!