Missing apiUrl documentation for running UnstructuredLoader locally about langchainjs HOT 3 CLOSED

Hello, @andrewdoro! I'm here to help you with any bugs, questions, or contributions. Let's work together to resolve your issue.

To address the issue of the default API option when running UnstructuredLoader locally, you can add the following code snippet to the local setup section of the documentation:

/**
 * Local setup for UnstructuredLoader with default API option.
 * Ensure the `apiUrl` is set to "http://localhost:8000/general/v0/general".
 * @example
 * ```typescript
 * const loader = new UnstructuredLoader("path/to/file", {
 *   apiUrl: "http://localhost:8000/general/v0/general",
 *   apiKey: "YOUR_API_KEY", // Optional if your local setup requires it
 * });
 * const docs = await loader.load();
 * ```
 */

This snippet ensures that the apiUrl points to the local server, avoiding the default API and preventing the "API key is malformed" error ^[1][2].

To continue the conversation, mention @dosu.

from langchainjs.

You're still required to provide an API key even if running locally. Sign up for one here (free), and they'll send you an email with a key.

from langchainjs.

I understand that I need the key to use their API.

For the local instance I was able to run it locally without an API Key by using the docker setup. I literally just used stuff, it's not a placeholder.

 const loader = new UnstructuredLoader(`pdfs/${randomName}.pdf`, {
   apiUrl: "http://localhost:8000/general/v0/general",
   apiKey: "stuff",
   strategy: "hi_res",
 });

This issue was related to the local setup of Unstructured. According to docs you can run this docker command to start a local instance.

docker run -p 8000:8000 -d --rm --name unstructured-api quay.io/unstructured-io/unstructured-api:latest --port 8000 --host 0.0.0.0

To actually consume this instance you also have to add the apiUrl. I'm proposing adding this to the docs for people wanting to run unstructured locally:
apiUrl: "http://localhost:8000/general/v0/general",

from langchainjs.