All Issues
24,993 verified issues
Who's using axios?
I would like to get an idea of who's using axios. If your company or project uses axios, please add it here. Would be great to know if you're using it for browser, node, react-native, or other as well.
ReferenceError: fetch is not defined in import OpenAI from 'openai'
Hello there! There seems to have been a few issues around this that have been resolved recently, but I'm still getting it so thought I would share just in case it's something different. Confirm this is a Node library issue and not an underlying OpenAI API issue - [X] This is an issue with the Node library Describe the bug When building and running the OpenAI library in our NestJS API, we can do so locally with no issue ; we can communicate with it and get responses etc. However, we have a Jest testing suite, and when trying to test, it fails with: ReferenceError: fetch is not defined 1 | import {Injectable} from "@nestjs/common"; 2 | import {ChatCompletionMessageParam} from "openai/resources/chat"; > 3 | import OpenAI from 'openai'; at Object.<anonymous> (../../../../node_modules/openai/_shims/fetch.js:8:17) at Object.<anonymous> (../../../../node_modules/openai/core.js:64:17) at Object.<anonymous> (../../../../node_modules/openai/src/index.ts:116:7) To Reproduce 1. Create a NestJS API 2. Install the openai library via npm 3. Create a controller that returns a response when a message is passed through via POST 4. Create a test suite in Jest to test this endpoint 5. Error occurs Code snippets _No response_ OS macOS Node version Node v18.17.1 Library version openai 4.6.0
Occasionally getting 429 Too many requests error
Describe the bug I have an app that makes 8 text completion requests to OpenAI using the GPT3.5 turbo in the back-end. The requests are chained together and all together I'm using around 5000 tokens. I have a pay as you go account (older than 48 hours) which according to the documentation about rate limits means that I can make 3500 requests per minute or do 90000 tokens per minute. Now considering the limitations, I believe I don't really hit any rate limits here but somehow I get a 429 Too many requests error back on some occasions. Currently the app doesn't have any active users than me. To Reproduce It happens randomly, so it's hard to say how to reproduce this issue other than chaining multiple createChatCompletions back to back. Code snippets [code block] And it's used like so in the backend: [code block] ``` OS macOs Node version Node v16.14.2 Library version "openai": "^3.2.1",
@apply doesn't work inside <style jsx></style>
Hi @adamwathan , I was working on a React component in NextJS and was using Tailwind to customize the designs. Putting the tailwind classes in `className` works fine while using like : [code block] But, since I am using too many classes for the component, and it will be a common style. I wanted to make it work by using @apply into another custom class. Something like : [code block] and CSS goes like : [code block] BTW I am using Styles as JSX. So, finally I want my component to look like : [code block] But this doesn't works out. Error I get is while inspecting the element Can you help me out in understanding if I am doing anything wrong, or does it support working under <style jsx> or not? An alternate solution would be appreciated. Thanks
The `punycode` module is deprecated in Node.js 21 (type = module)
Confirm this is a Node library issue and not an underlying OpenAI API issue - [X] This is an issue with the Node library Describe the bug When importing the `openai` package in Node.js 21 it throws an error: [code block] To Reproduce `package.json`: [code block] `index.js`: [code block] `.env.dev`: [code block] Code snippets _No response_ OS macOS Node version v21.2.0 Library version openai 4.19.1
Is 'functions' parameter not supported?
Describe the bug I'm trying to use the functions parameter to help force JSON response. Is this not supported? To Reproduce Call createChatCompletion with 'functions' array. Code snippets [code block] Results in the error: [code block] ``` OS macOS Node version 18.7.0 Library version 3.3.0
bug: SSE Subscription Never Aborts
Provide environment information System: OS: Linux 5.0 undefined CPU: (6) x64 Intel(R) Core(TM) i9-9880H CPU @ 2.30GHz Memory: 0 Bytes / 0 Bytes Shell: 1.0 - /bin/jsh Binaries: Node: 18.20.3 - /usr/local/bin/node Yarn: 1.22.19 - /usr/local/bin/yarn npm: 10.2.3 - /usr/local/bin/npm pnpm: 8.15.6 - /usr/local/bin/pnpm npmPackages: @tanstack/react-query: ^5.59.17 => 5.59.20 @trpc/client: ^11.0.0-rc.608 => 11.0.0-rc.608+f75de97b3 @trpc/next: ^11.0.0-rc.608 => 11.0.0-rc.608+f75de97b3 @trpc/react-query: ^11.0.0-rc.608 => 11.0.0-rc.608+f75de97b3 @trpc/server: ^11.0.0-rc.608 => 11.0.0-rc.608+f75de97b3 next: 15.0.2 => 15.0.2 react: ^19.0.0-rc-7c8e5e7a-20241101 => 19.0.0-rc-fb9a90fa48-20240614 typescript: ^5 => 5.6.3 Describe the bug Subscription never gets aborted Link to reproduction https://stackblitz.com/~/github.com/yspreen/trpc-bug To reproduce https://stackblitz.com/~/github.com/yspreen/trpc-bug Additional information _No response_ 👨👧👦 Contributing - [X] 🙋♂️ Yes, I'd be down to file a PR fixing this bug!
.findOne() is different from .findOne().lean()
Prerequisites - [X] I have written a descriptive issue title - [X] I have searched existing issues to ensure the bug has not already been reported Mongoose version 6.8.2 Node.js version 16.18.0 MongoDB server version 6.0.1 Typescript version (if applicable) 4.9.4 Description related to https://github.com/Automattic/mongoose/issues/12905 nested paths are not working well when using just .findOne, but works well with lean example this breaks [code block] this do not breaks [code block] we can see nested path in mongosh and robo 3t [code block] it does not work on 6.8.2, 6.8.0 it works on 6.7.0 I will try to find the exactly version that breaks this is a regression Steps to Reproduce [code block] Expected Behavior it should resolve properly with and without .lean()
React Native app shows blank screen after Expo SDK upgrade
After upgrading Expo SDK, the app shows a blank white screen with no error visible. The root cause is usually a Metro bundler cache containing stale pre-upgrade build artifacts, or a native module that requires a fresh `npx expo prebuild` to regenerate platform-specific code after the SDK version change.
TanStack Query v5 shows stale data after mutation despite invalidateQueries
After a successful mutation (create/update/delete), calling queryClient.invalidateQueries() causes a refetch that is visible in the network tab returning fresh data — but the UI component continues displaying the old data. The issue occurs when multiple query keys share the same base key prefix and the invalidation targets only the parent key while child queries remain stale.
Calling populate on embedded doc of embedded doc
Hey, I have the following schemas ( simplified for issue) User = new Schema login:String A = new Schema emb: [B] comments: [Comments] //and following embedded schemas B = new Schema comments: [Comments] Comments = new Schema creator: type: ObjectId ref: 'User' message: String Now while populate works on A eg A.find().populate('comments.creator') it doesn't work on the double nested embedded doc eg A.find().populate('emb.comments.creator') Any ideas?
Async view support
This is a somewhat controversial topic for views. Here are my opinions. The good: - we could pass "promise"-like objects to the view and forget about them (db queries etc) - async helpers - early head flush The bad: - complicates, and in turn slows down the view system - graceful error reporting becomes much more difficult - promotes the use of a view for non-view-like computation
Arrays as paths
[code block] As suggested by @71104 in #1369.
response.resume error: Cannot resume() closed Socket.
Hey, I'm having this sporadic message, I haven't dig too much into it, but it started to show up once I Implemented a file uploading feature using connect-form. Any ideas?
http-proxy doesn't proxy the latest websocket specifications
Howdy, A few days ago, we added FF6 and Chrome 14 websocket support to Socket.IO. But http-proxy doesn't seem to be able to proxy these upgrade events. If I run a plain Socket.IO 0.8.2 chat example it works fine in FF6 and latest chrome, but if I add node-http-proxy infront of it only the Draft 76 specification receives a upgrade event. After a bit digging I found out that https://github.com/nodejitsu/node-http-proxy/blob/master/lib/node-http-proxy.js#L862 doesn't get fired for the new websocket specifications (it does work for older websocket specifications like draft 76)
Check for connection:close
Reverse proxies like to turn a connection:keep-alive to a connection:close for the backend. This means a connection CAN'T be reused, it must be closed with res.end(). Else the reverse proxy thinks that the request isn't fully answered, and keeps it in the internal buffer. This is what I watched at lighttpd and nginx. For nginx there is a easy work around with just disabling the buffer, but there is no such setting for lighttpd. So please: close the connection if the connection:close header is set
redirect for http server when behind https proxy
Found some interesting behavior in the absolute res.redirect(...) when serving an express http server behind a https nginx proxy. I looked at the code and see that for absolute redirects it creates the url with http:// host .. etc etc ... this causes the nginx proxy to redirect to https again (since all my traffic must be https). What would be the best way to make the force the url to be: https ? Note that the express server itself is http but it sits behind a https proxy for all requests.
broadcast doesn't work in namespace
[code block] message event is fired... but it can't broadcast. and `broadcast.emit` too.. I use socket.io .0.7.2 and node. 0.4.8
Firefox support
I can not get the simple chat example working with firefox 3.5.9 (linux) or 3.6.10 (win7) Using live http headers, nothing is failing and WebSocketMain.swf is being loaded but it never connects. Yes I am running server.js as root and port 843 is open and returning: <cross-domain-policy> <allow-access-from domain="" to-ports=""/> </cross-domain-policy> I have the main chat body with "connecting..." and no text input of submit button. It fails after a few minutes with this.socket is undefined http://localhost:8080/client/socket.io.js line 386. I can supply the http header log and a screen shot if you like. Works in: IE 8.0.7600 (win7) Chrome 6.0.472.63 (linux) iPad iOS 3.2.2 / iPhone iOS 4.0 Safari
socket.io does not work on Firefox 6 beta
When using socket.io in Firefox 6 beta the connection will hang after performing authorization. The connect event will not be fired on the server. The Firefox debugging console provides the following information: Error: Syntax-Error Sourcefile: http://host/socket.io/1/?t=1310553989400 Line: 1, Column: 1 Sourcecode: 190006406397831442:15:25:websocket,htmlfile,xhr-polling,jsonp-polling Since firefox claims a syntax error maybe a correct mimetype will help (there's none set in the response)?