Where to place js files that aren't serverless

I have a few js files used in my serverless functions (they are like libraries) and I have reached maximum serverless functions count. I'm looking for a simple way that will let me keep them unaccessible from users but accessible to functions for require("foo.js")

Vercel Alum

Hey can you explain a little more about your setup:

Libraries or frameworks being used
Any current errors you are seeing
Are you in a monorepo setup or a single monolithic repository?

More than happy to dive in, I think a little more context will help me steer the conversation to find a solution faster.

I'm not using any frameworks, I have a small repo with no actual errors the files I said about look like this and I have used them in other places

1//--ai.js--//2const tokens = (i) => {3  let t = 0;4  for (const m of i) {5    t += m.content.split(" ").length;6  }7  return t;8};9const limit_tokens = (i,x) => {10  let t = 0;11  let l = [];12  for (let n=i.length-1;n>=0;n--) {13    const m = i[n];14    const c = m.content.split(" ").length;15    if (t+c<=x) {16      l.unshift(m);17      t += c;18    } else {19      l.unshift({ ...m , content: m.content.split(" ").slice(-x+t).join(" ") });20      break;21    }22  }23  return l;24};25const groq_gen = async (messages,model="gemma2-9b-it",temperature=1,top_p=1,max_completion_tokens=1024) => {26  const res = await fetch("https://api.groq.com/openai/v1/chat/completions",{27    method: "POST",28    headers: {29      "Content-Type": "application/json",30      "Authorization": "Bearer ******"31    },32    body: JSON.stringify({33      messages,34      model,35      temperature,36      top_p,37      max_completion_tokens,38      stop: null39    })40  });41  return await res.json();42};43module.exports = {44  tokens,45  limit_tokens,46  groq_gen47}

Vercel Alum

Ah ok, no problem! Most common reason is probably putting everything in an /api. If you go to your project and select your deployment, there is a Sources tab that you should see to look at the input and the output of your project.

I created a test project just now that is this input:

This is the output

Notice the lib/test.js is not marked as a lambda. I think that is what you are looking for.

Let me know if I'm off, happy to keep digging in 👍

Thank you! That's what I'm looking for but as I said there is 1 more problem that I need to fix, The lib folder and it's files are accessible within the deployment url but I don't want it to be accessible

reply to #5

Vercel Alum

Can you elaborate more?

In my example, the lib folder I'm unable to access directly from the client/deployment (ex. browser), but the /api/test can use that lib/test and process a request.

There may also be some project config that could help you here?

Very interested, let me know!

Accepted answer

Can't find why but mine doesn't work I used

1{2  "routes": [3    {4      "src": "/lib/(.*)",5      "dest": "/404.html"6    }7  ]8}

and I think it's fixed for now

This topic is closed. It no longer accepts new replies.

I'm not using any frameworks, I have a small repo with no actual errors the files I said about look like this and I have used them in other places

1//--ai.js--//2const tokens = (i) => {3  let t = 0;4  for (const m of i) {5    t += m.content.split(" ").length;6  }7  return t;8};9const limit_tokens = (i,x) => {10  let t = 0;11  let l = [];12  for (let n=i.length-1;n>=0;n--) {13    const m = i[n];14    const c = m.content.split(" ").length;15    if (t+c<=x) {16      l.unshift(m);17      t += c;18    } else {19      l.unshift({ ...m , content: m.content.split(" ").slice(-x+t).join(" ") });20      break;21    }22  }23  return l;24};25const groq_gen = async (messages,model="gemma2-9b-it",temperature=1,top_p=1,max_completion_tokens=1024) => {26  const res = await fetch("https://api.groq.com/openai/v1/chat/completions",{27    method: "POST",28    headers: {29      "Content-Type": "application/json",30      "Authorization": "Bearer ******"31    },32    body: JSON.stringify({33      messages,34      model,35      temperature,36      top_p,37      max_completion_tokens,38      stop: null39    })40  });41  return await res.json();42};43module.exports = {44  tokens,45  limit_tokens,46  groq_gen47}