A better fetch API. Works on node, browser, and workers.
Important
You are on v2 (alpha) development branch. See v1 for v1 docs.
Spoiler
Install:
npx nypm i ofetch
Import:
import { ofetch } from "ofetch";
ofetch smartly parse JSON responses.
const { users } = await ofetch("/api/users");
For binary content types, ofetch will instead return a Blob object.
You can optionally provide a different parser than JSON.parse, or specify blob, arrayBuffer, text or stream to force parsing the body with the respective FetchResponse method.
// Return text as is await ofetch("/movie?lang=en", { parseResponse: (txt) => txt }); // Get the blob version of the response await ofetch("/api/generate-image", { responseType: "blob" }); // Get the stream version of the response await ofetch("/api/generate-image", { responseType: "stream" });
If an object or a class with a .toJSON() method is passed to the body option, ofetch automatically stringifies it.
ofetch utilizes JSON.stringify() to convert the passed object. Classes without a .toJSON() method have to be converted into a string value in advance before being passed to the body option.
For PUT, PATCH, and POST request methods, when a string or object body is set, ofetch adds the default "content-type": "application/json" and accept: "application/json" headers (which you can always override).
Additionally, ofetch supports binary responses with Buffer, ReadableStream, Stream, and compatible body types. ofetch will automatically set the duplex: "half" option for streaming support!
Example:
const { users } = await ofetch("/api/users", { method: "POST", body: { some: "json" }, });
ofetch Automatically throws errors when response.ok is false with a friendly error message and compact stack (hiding internals).
A parsed error body is available with error.data. You may also use FetchError type.
await ofetch("https://google.com/404"); // FetchError: [GET] "https://google/404": 404 Not Found // at async main (/project/playground.ts:4:3)
To catch error response:
await ofetch("/url").catch((error) => error.data);
To bypass status error catching you can set ignoreResponseError option:
await ofetch("/url", { ignoreResponseError: true });
ofetch Automatically retries the request if an error happens and if the response status code is included in retryStatusCodes list:
Retry status codes:
408- Request Timeout409- Conflict425- Too Early (Experimental)429- Too Many Requests500- Internal Server Error502- Bad Gateway503- Service Unavailable504- Gateway Timeout
You can specify the amount of retry and delay between them using retry and retryDelay options and also pass a custom array of codes using retryStatusCodes option.
The default for retry is 1 retry, except for POST, PUT, PATCH, and DELETE methods where ofetch does not retry by default to avoid introducing side effects. If you set a custom value for retry it will always retry for all requests.
The default for retryDelay is 0 ms.
await ofetch("http://google.com/404", { retry: 3, retryDelay: 500, // ms retryStatusCodes: [404, 500], // response status codes to retry });
You can specify timeout in milliseconds to automatically abort a request after a timeout (default is disabled).
await ofetch("http://google.com/404", { timeout: 3000, // Timeout after 3 seconds });
The response can be type assisted:
const article = await ofetch<Article>(`/api/article/${id}`); // Auto complete working with article.id
By using baseURL option, ofetch prepends it for trailing/leading slashes and query search params for baseURL using ufo:
await ofetch("/config", { baseURL });
By using query option (or params as alias), ofetch adds query search params to the URL by preserving the query in the request itself using ufo:
await ofetch("/movie?lang=en", { query: { id: 123 } });
Providing async interceptors to hook into lifecycle events of ofetch call is possible.
You might want to use ofetch.create to set shared interceptors.
onRequest is called as soon as ofetch is called, allowing you to modify options or do simple logging.
await ofetch("/api", { async onRequest({ request, options }) { // Log request console.log("[fetch request]", request, options); // Add `?t=1640125211170` to query search params options.query = options.query || {}; options.query.t = new Date(); }, });
onRequestError will be called when the fetch request fails.
await ofetch("/api", { async onRequestError({ request, options, error }) { // Log error console.log("[fetch request error]", request, error); }, });
onResponse will be called after fetch call and parsing body.
await ofetch("/api", { async onResponse({ request, response, options }) { // Log response console.log("[fetch response]", request, response.status, response.body); }, });
onResponseError is the same as onResponse but will be called when fetch happens but response.ok is not true.
await ofetch("/api", { async onResponseError({ request, response, options }) { // Log error console.log( "[fetch response error]", request, response.status, response.body ); }, });
If necessary, it's also possible to pass an array of function that will be called sequentially.
await ofetch("/api", { onRequest: [ () => { /* Do something */ }, () => { /* Do something else */ }, ], });
This utility is useful if you need to use common options across several fetch calls.
Note: Defaults will be cloned at one level and inherited. Be careful about nested options like headers.
const apiFetch = ofetch.create({ baseURL: "/api" }); apiFetch("/test"); // Same as ofetch('/test', { baseURL: '/api' })
By using headers option, ofetch adds extra headers in addition to the request default headers:
await ofetch("/movies", { headers: { Accept: "application/json", "Cache-Control": "no-cache", }, });
If you need to access raw response (for headers, etc), you can use ofetch.raw:
const response = await ofetch.raw("/sushi"); // response._data // response.headers // ...
As a shortcut, you can use ofetch.native that provides native fetch API
const json = await ofetch.native("/sushi").then((r) => r.json());
Example: Handle SSE response:
const stream = await ofetch("/sse"); const reader = stream.getReader(); const decoder = new TextDecoder(); while (true) { const { done, value } = await reader.read(); if (done) break; // Here is the chunked text of the SSE response. const text = decoder.decode(value); }
Important
Environment Variables: Bun and Deno respect HTTP_PROXY and HTTPS_PROXY environment variables. Node.js requires setting NODE_USE_ENV_PROXY=1 to enable built-in proxy support.
In Node.js (>= 18), you can use the dispatcher option with undici's ProxyAgent.
import { ProxyAgent } from "undici"; const proxyAgent = new ProxyAgent("http://localhost:3128"); await ofetch("https://icanhazip.com", { dispatcher: proxyAgent });
Example: Set proxy globally for all requests:
import { ProxyAgent, setGlobalDispatcher } from "undici"; setGlobalDispatcher(new ProxyAgent("http://localhost:3128"));
Example: Allow self-signed certificates (USE AT YOUR OWN RISK!)
import { Agent } from "undici"; // Note: This makes fetch insecure against MITM attacks. USE AT YOUR OWN RISK! const unsecureAgent = new Agent({ connect: { rejectUnauthorized: false } }); await ofetch("https://self-signed.example.com/", { dispatcher: unsecureAgent });
Bun supports the proxy option:
await ofetch("https://icanhazip.com", { proxy: "http://localhost:3128", });
Deno can also use undici with npm specifiers for programmatic configuration.
You can augment the FetchOptions interface to add custom properties.
// Place this in any `.ts` or `.d.ts` file. // Ensure it's included in the project's tsconfig.json "files". declare module "ofetch" { interface FetchOptions { // Custom properties requiresAuth?: boolean; } } export {};
This lets you pass and use those properties with full type safety throughout ofetch calls.
const myFetch = ofetch.create({ onRequest(context) { // ^? { ..., options: {..., requiresAuth?: boolean }} console.log(context.options.requiresAuth); }, }); myFetch("/foo", { requiresAuth: true });
💛 Published under the MIT license.