# Process a PDF for text extraction and embedding

**POST /message/pdf**

Takes a URL to a PDF and a fileId (CUID of the file_uploads record). Checks the PDF size, downloads it, extracts text, chunks it, generates embeddings, and stores them.



## Servers
- Local development server: http://localhost:3000/api/v1 (Local development server)
- Production server: https://referspecs-backend.onrender.com/api/v1 (Production server)



## Parameters



### Body: application/json (object)

- **url** (string(url))
  The publicly accessible URL of the PDF file (from Vercel Blob).
- **fileId** (string)
  The CUID of the corresponding record in the 'file_uploads' table.


## Responses
### 200
PDF processed and embedded successfully

#### Body: application/json (object)
- **success** (boolean)

- **statusCode** (integer)

- **message** (string)

- **data** (object)


### 400
Bad Request (e.g., missing URL/fileId, file too large, invalid URL, failed size check)

#### Body: application/json (object)
- **success** (boolean)

- **statusCode** (integer)

- **message** (string)


### 500
Internal Server Error (e.g., failed to download/process PDF, embedding error)

#### Body: application/json (object)
- **success** (boolean)

- **statusCode** (integer)

- **message** (string)



[Powered by Bump.sh](https://bump.sh)