Skip to main content
Glama
ts-macro-derive.md10 kB
# ts\_macro\_derive The `#[ts_macro_derive]` attribute is a Rust procedural macro that registers your function as a Macroforge derive macro. ## Basic Syntax Rust ``` use macroforge_ts::macros::ts_macro_derive; use macroforge_ts::ts_syn::{TsStream, MacroforgeError}; #[ts_macro_derive(MacroName)] pub fn my_macro(mut input: TsStream) -> Result<TsStream, MacroforgeError> {     // Macro implementation } ``` ## Attribute Options ### Name (Required) The first argument is the macro name that users will reference in `@derive()`: Rust ``` #[ts_macro_derive(JSON)]  // Users write: @derive(JSON) pub fn derive_json(...) ``` ### Description Provides documentation for the macro: Rust ``` #[ts_macro_derive(     JSON,     description = "Generates toJSON() returning a plain object" )] pub fn derive_json(...) ``` ### Attributes Declare which field-level decorators your macro accepts: Rust ``` #[ts_macro_derive(     Debug,     description = "Generates toString()",     attributes(debug)  // Allows @debug({ ... }) on fields )] pub fn derive_debug(...) ``` Note Declared attributes become available as `@attributeName({ options })` decorators in TypeScript. ## Function Signature Rust ``` pub fn my_macro(mut input: TsStream) -> Result<TsStream, MacroforgeError> ``` | Parameter | Description | | ----------------------------------- | ------------------------------------------------------- | | `input: TsStream` | Token stream containing the class/interface AST | | `Result<TsStream, MacroforgeError>` | Returns generated code or an error with source location | ## Parsing Input Use `parse_ts_macro_input!` to convert the token stream: Rust ``` use macroforge_ts::ts_syn::{Data, DeriveInput, parse_ts_macro_input}; #[ts_macro_derive(MyMacro)] pub fn my_macro(mut input: TsStream) -> Result<TsStream, MacroforgeError> {     let input = parse_ts_macro_input!(input as DeriveInput);     // Access class data     match &input.data {         Data::Class(class) => {             let class_name = input.name();             let fields = class.fields();             // ...         }         Data::Interface(interface) => {             // Handle interfaces         }         Data::Enum(_) => {             // Handle enums (if supported)         }     } } ``` ## DeriveInput Structure Rust ``` struct DeriveInput {     pub ident: Ident,           // The type name     pub span: SpanIR,           // Span of the type definition     pub attrs: Vec<Attribute>,  // Decorators (excluding @derive)     pub data: Data,             // The parsed type data     pub context: MacroContextIR, // Macro context with spans     // Helper methods     fn name(&self) -> &str;              // Get the type name     fn decorator_span(&self) -> SpanIR;  // Span of @derive decorator     fn as_class(&self) -> Option<&DataClass>;     fn as_interface(&self) -> Option<&DataInterface>;     fn as_enum(&self) -> Option<&DataEnum>; } enum Data {     Class(DataClass),     Interface(DataInterface),     Enum(DataEnum),     TypeAlias(DataTypeAlias), } impl DataClass {     fn fields(&self) -> &[FieldIR];     fn methods(&self) -> &[MethodSigIR];     fn field_names(&self) -> impl Iterator<Item = &str>;     fn field(&self, name: &str) -> Option<&FieldIR>;     fn body_span(&self) -> SpanIR;      // For inserting code into class body     fn type_params(&self) -> &[String]; // Generic type parameters     fn heritage(&self) -> &[String];    // extends/implements clauses     fn is_abstract(&self) -> bool; } impl DataInterface {     fn fields(&self) -> &[InterfaceFieldIR];     fn methods(&self) -> &[InterfaceMethodIR];     fn field_names(&self) -> impl Iterator<Item = &str>;     fn field(&self, name: &str) -> Option<&InterfaceFieldIR>;     fn body_span(&self) -> SpanIR;     fn type_params(&self) -> &[String];     fn heritage(&self) -> &[String];    // extends clauses } impl DataEnum {     fn variants(&self) -> &[EnumVariantIR];     fn variant_names(&self) -> impl Iterator<Item = &str>;     fn variant(&self, name: &str) -> Option<&EnumVariantIR>; } impl DataTypeAlias {     fn body(&self) -> &TypeBody;     fn type_params(&self) -> &[String];     fn is_union(&self) -> bool;     fn is_object(&self) -> bool;     fn as_union(&self) -> Option<&[TypeMember]>;     fn as_object(&self) -> Option<&[InterfaceFieldIR]>; } ``` ## Accessing Field Data ### Class Fields (FieldIR) Rust ``` struct FieldIR {     pub name: String,               // Field name     pub span: SpanIR,               // Field span     pub ts_type: String,            // TypeScript type annotation     pub optional: bool,             // Whether field has ?     pub readonly: bool,             // Whether field is readonly     pub visibility: Visibility,     // Public, Protected, Private     pub decorators: Vec<DecoratorIR>, // Field decorators } ``` ### Interface Fields (InterfaceFieldIR) Rust ``` struct InterfaceFieldIR {     pub name: String,     pub span: SpanIR,     pub ts_type: String,     pub optional: bool,     pub readonly: bool,     pub decorators: Vec<DecoratorIR>,     // Note: No visibility field (interfaces are always public) } ``` ### Enum Variants (EnumVariantIR) Rust ``` struct EnumVariantIR {     pub name: String,     pub span: SpanIR,     pub value: EnumValue,  // Auto, String(String), or Number(f64)     pub decorators: Vec<DecoratorIR>, } ``` ### Decorator Structure Rust ``` struct DecoratorIR {     pub name: String,      // e.g., "serde"     pub args_src: String,  // Raw args text, e.g., "skip, rename: 'id'"     pub span: SpanIR, } ``` Note To check for decorators, iterate through `field.decorators` and check `decorator.name`. For parsing options, you can write helper functions like the built-in macros do. ## Adding Imports If your macro generates code that requires imports, use the `add_import` method on `TsStream`: Rust ``` // Add an import to be inserted at the top of the file let mut output = body! {     validate(): ValidationResult {         return validateFields(this);     } }; // This will add: import { validateFields, ValidationResult } from "my-validation-lib"; output.add_import("validateFields", "my-validation-lib"); output.add_import("ValidationResult", "my-validation-lib"); Ok(output) ``` Note Imports are automatically deduplicated. If the same import already exists in the file, it won't be added again. ## Returning Errors Use `MacroforgeError` to report errors with source locations: Rust ``` #[ts_macro_derive(ClassOnly)] pub fn class_only(mut input: TsStream) -> Result<TsStream, MacroforgeError> {     let input = parse_ts_macro_input!(input as DeriveInput);     match &input.data {         Data::Class(_) => {             // Generate code...             Ok(body! { /* ... */ })         }         _ => Err(MacroforgeError::new(             input.decorator_span(),             "@derive(ClassOnly) can only be used on classes",         )),     } } ``` ## Complete Example Rust ``` use macroforge_ts::macros::{ts_macro_derive, body}; use macroforge_ts::ts_syn::{     Data, DeriveInput, FieldIR, MacroforgeError, TsStream, parse_ts_macro_input, }; // Helper function to check if a field has a decorator fn has_decorator(field: &FieldIR, name: &str) -> bool {     field.decorators.iter().any(|d| d.name.eq_ignore_ascii_case(name)) } #[ts_macro_derive(     Validate,     description = "Generates a validate() method",     attributes(validate) )] pub fn derive_validate(mut input: TsStream) -> Result<TsStream, MacroforgeError> {     let input = parse_ts_macro_input!(input as DeriveInput);     match &input.data {         Data::Class(class) => {             let validations: Vec<_> = class.fields()                 .iter()                 .filter(|f| has_decorator(f, "validate"))                 .collect();             Ok(body! {                 validate(): string[] {                     const errors: string[] = [];                     {#for field in validations}                         if (!this.@{field.name}) {                             errors.push("@{field.name} is required");                         }                     {/for}                     return errors;                 }             })         }         _ => Err(MacroforgeError::new(             input.decorator_span(),             "@derive(Validate) only works on classes",         )),     } } ``` ## Next Steps * [Learn the template syntax](../../docs/custom-macros/ts-quote)

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/macroforge-ts/mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server