# Error Propagation
Funkcia offers a concise and convenient way to write your code in a more imperative style that utilizes the native scope provided by the generator syntax. This syntax is more linear and resembles normal synchronous code.
Drawing primarily from Rust's `?` operator for error propagation, and inspired by Gleam's [`use` expressions](https://gleam.run/news/v0.25-introducing-use-expressions/), [neverthrow's `safeTry`](https://github.com/supermacro/neverthrow?tab=readme-ov-file#safetry), and [Effect's `gen`](https://effect.website/docs/getting-started/using-generators/#_top) functions, the following functions provide a clean way to handle sequential operations while maintaining proper error handling and type safety.
#### use
Evaluates an *async* generator early returning when a `Result.Error` is propagated or returning the `ResultAsync` returned by the generator.
* Each `yield*` automatically awaits and unwraps the `ResultAsync` value or propagates `Error`.
* If any operation resolves to `Result.Error`, the entire generator exits early.
import { ResultAsync } from 'funkcia';
declare const safeReadFile: (path: string) => ResultAsync<string, NodeFSError>;
declare const safeWriteFile: (path: string, content: string) => ResultAsync<string, NodeFSError>;
// ┌─── ResultAsync<string, NodeFSError>
// ▼
const mergedContent = ResultAsync.use(async function* () {
const fileA = yield* safeReadFile('data.txt');
const fileB = yield* safeReadFile('non-existent-file.txt'); // returns ResultAsync.Error immediately
return safeWriteFile('output.txt', `${fileA}\n${fileB}`); // doesn't run
});
// Output: ResultAsync<string, NodeFSError>
#### fn
Returns a function that evaluates an *async* generator when called with the defined arguments, early returning when a `Result.Error` is propagated or returning the `ResultAsync` returned by the generator.
* Each `yield*` automatically awaits and unwraps the `ResultAsync` value or propagates `Error`.
* If any operation resolves to `Result.Error`, the entire generator exits early.
import { ResultAsync } from 'funkcia';
declare const safeReadFile: (path: string) => ResultAsync<string, NodeFSError>;
declare const safeWriteFile: (path: string, content: string) => ResultAsync<string, NodeFSError>;
// ┌─── (output: string, pathA: string, pathB: string) => ResultAsync<string, NodeFSError>
// ▼
const safeMergeFiles = ResultAsync.fn(async function* (output: string, pathA: string, pathB: string) {
const fileA = yield* safeReadFile(pathA);
const fileB = yield* safeReadFile(pathB);
return safeWriteFile(output, `${fileA}\n${fileB}`);
});
const mergedContent = safeMergeFiles('output.txt', 'data.txt', 'updated-data.txt');
// Output: ResultAsync<string, NodeFSError>
### Understanding the use method
The `use` method provides a way to write sequential operations that might fail, similar to Rust's `?` operator. It lets you write code that looks synchronous while safely handling potential failures.
It essentially creates a "safe context" where you can work with values as if they were guaranteed to exist, while maintaining all the safety guarantees of `ResultAsync`. If anything fails, the failure propagates automatically. Like an electronic relay that controls current flow, relay controls computation flow: `Result.Ok` continues, `Result.Error` breaks the circuit.
Here's a practical example:
```typescript
import { ResultAsync } from 'funkcia';
declare function rateLimit(clientId: ClientId, ip: IpAddress): ResultAsync;
declare function findUserByEmail(email: Email): ResultAsync;
const userPreferences = ResultAsync.use(function* () {
// First, check if API rate limit is allowed
yield* rateLimit(req.headers['x-client-id'], req.ip);
// If rate-limit is not blocked, get the user
const user = yield* findUserByEmail(req.query.email);
// If all steps succeed, we can use the accumulated context to get user preferences
return ResultAsync.ok(user.preferences);
});
```
The equivalent code without `use` would be much more nested:
```typescript
import { ResultAsync } from 'funkcia';
declare function rateLimit(clientId: ClientId, ip: IpAddress): ResultAsync;
declare function findUserByEmail(email: Email): ResultAsync;
const userPreferences = rateLimit(req.headers['x-client-id'], req.ip)
.andThen(() =>
findUserByEmail(req.query.email)
.map(user => user.preferences)
);
```
Or with intermediate variables:
{% code fullWidth="false" %}
```typescript
import { ResultAsync } from 'funkcia';
declare function rateLimit(clientId: ClientId, ip: IpAddress): ResultAsync;
declare function findUserByEmail(email: Email): ResultAsync;
const rateLimitResult = rateLimit(req.headers['x-client-id'], req.ip);
const user = rateLimitResult.andThen(() => findUserByEmail(req.query.email));
const userPreferences = user.map(user => user.preferences);
```
{% endcode %}
[^1]: const fileA: string
[^2]: const fileB: string
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://funkcia.lukemorales.io/data-types/resultasync/error-propagation.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.