Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/rjjha/chatpdf--backend

A full-stack app for querying and uploading PDFs using OpenAI's RAG system. Built with React, FastAPI, and MongoDB, it ensures seamless interaction and intelligent PDF-based responses.
https://github.com/rjjha/chatpdf--backend

fastapi gemini-api python3

Last synced: about 1 month ago
JSON representation

A full-stack app for querying and uploading PDFs using OpenAI's RAG system. Built with React, FastAPI, and MongoDB, it ensures seamless interaction and intelligent PDF-based responses.

Awesome Lists containing this project

README 

          

# ChatPdf

**PDF Query Assistant**  

A full-stack application enabling users to query and upload PDF files, leveraging a Retrieval-Augmented Generation (RAG) system with OpenAI’s API. Built with a React frontend, FastAPI backend, and MongoDB for query history storage, it provides seamless interaction and intelligent responses based on PDF content.



## Features

---

1. **Upload PDF**: Users can upload a PDF file through an intuitive interface to enable the system to process its content.



2. **Input Query**: Users can type in their questions related to the uploaded PDF directly in the query box on the frontend.



3. **Receive Responses**: The application retrieves relevant information from the PDF using the RAG system and OpenAI’s API, providing accurate and contextual answers.



4. **Query History**: Users can view past queries and responses stored in MongoDB for reference, ensuring continuity and easy access to previously searched information.



----



# Backend Setup Guide



Follow these steps to set up and run the backend for your project.



## Prerequisites



- Python 3.7+ installed

- pip (Python package installer)



---



## Steps to Start



1. **Install `virtualenv`**  

   ```bash

   pip install virtualenv

   ```



2. **Create a Virtual Environment**  

   ```bash

   python3 -m venv venv

   ```



3. **Activate the Virtual Environment**  

   - **Linux/macOS**:  

     ```bash

     source venv/bin/activate

     ```

   - **Windows**:  

     ```bash

     venv\Scripts\activate

     ```



4. **Install Dependencies**  

   ```bash

   pip install -r requirements.txt

   ```



5. **Run the Backend Server**  

   ```bash

   uvicorn app.main:app --reload

   ```



---



## Access the Application



- Open your browser and go to:  

  `http://127.0.0.1:8000`



---



## Notes



- Use `CTRL+C` to stop the server.

- To deactivate the virtual environment, run:  

  ```bash

  deactivate

  ```



# API Documentation



## 1. Root Endpoint



### GET `/`



**Description**:  

Root endpoint to check the status of the service.



### Responses:

- **200 OK**  

  Returns the root status.



---



## 2. Upload PDF



### POST `/api/v1/upload_pdf`



**Description**:  

Uploads a PDF to the server and processes it.



### Parameters:

- **None**  



### Request Body:

- **Content-Type**: `multipart/form-data`

- **File**: 

  - **Field Name**: `file`

  - **Type**: `string($binary)`

  - **Example File**: `Abhisahar Article - Open Src.pdf`



### Example cURL Request:

```bash

curl -X 'POST' \

  'http://127.0.0.1:8000/api/v1/upload_pdf' \

  -H 'accept: application/json' \

  -H 'Content-Type: multipart/form-data' \

  -F 'file=@Abhisahar Article - Open Src.pdf;type=application/pdf'

```



### Responses:



#### 200 OK

- **Description**: PDF uploaded and processed successfully.

- **Response Body**:

```json

{

  "status": "success",

  "message": "PDF 'Abhisahar Article - Open Src.pdf' uploaded and processed successfully.",

  "data": {

    "pdf_id": "67385818c013d1176861687e",

    "pdf_metadata": {

      "id": "67385818c013d1176861687e",

      "filename": "Abhisahar Article - Open Src.pdf",

      "cloudinary_url": "https://res.cloudinary.com/dnu3aojdm/raw/upload/v1731745816/mxvdhu1nxgxtynjdecbw",

      "cloudinary_public_id": "mxvdhu1nxgxtynjdecbw",

      "file_size": 64475,

      "created_at": "2024-11-16T08:30:16",

      "format": "pdf"

    }

  }

}

```



#### 422 Unprocessable Entity

- **Description**: Validation error, typically due to missing or invalid parameters.

- **Response Body**:

```json

{

  "detail": [

    {

      "loc": ["body", "file"],

      "msg": "Field required",

      "type": "value_error.missing"

    }

  ]

}

```



---



## 3. List Uploaded PDFs



### GET `/api/v1/pdfs`



**Description**:  

Retrieves a list of uploaded PDFs with pagination support.



### Parameters:

- **skip**: `integer` (query) - Offset for pagination (default: 0).

- **limit**: `integer` (query) - Number of records to fetch (default: 10).



### Example cURL Request:

```bash

curl -X 'GET' \

  'http://127.0.0.1:8000/api/v1/pdfs?skip=0&limit=10' \

  -H 'accept: application/json'

```



### Responses:



#### 200 OK

- **Description**: Successfully retrieves the list of PDFs.

- **Response Body**:

```json

[

  {

    "id": "67385818c013d1176861687e",

    "filename": "Abhisahar Article - Open Src.pdf",

    "cloudinary_url": "https://res.cloudinary.com/dnu3aojdm/raw/upload/v1731745816/mxvdhu1nxgxtynjdecbw",

    "cloudinary_public_id": "mxvdhu1nxgxtynjdecbw",

    "file_size": 64475,

    "created_at": "2024-11-16T08:30:16",

    "format": "pdf"

  },

  ...

]

```



#### 422 Unprocessable Entity

- **Description**: Validation error for request parameters.

- **Response Body**:

```json

{

  "detail": [

    {

      "loc": ["query", "limit"],

      "msg": "Value is too large",

      "type": "value_error.max_size"

    }

  ]

}

```



---



## 4. Delete PDF



### DELETE `/api/v1/pdfs/{pdf_id}`



**Description**:  

Deletes a PDF from both Cloudinary and the database.



### Parameters:

- **pdf_id**: `string` (path) - The unique identifier of the PDF to be deleted.



### Example cURL Request:

```bash

curl -X 'DELETE' \

  'http://127.0.0.1:8000/api/v1/pdfs/673793c2fcf059dc08dcc4d2' \

  -H 'accept: application/json'

```



### Responses:



#### 200 OK

- **Description**: Successfully deleted the PDF.

- **Response Body**:

```json

{

  "status": "success",

  "message": "PDF deleted successfully",

  "pdf_id": "673793c2fcf059dc08dcc4d2"

}

```



#### 422 Unprocessable Entity

- **Description**: Validation error, typically due to an invalid `pdf_id`.

- **Response Body**:

```json

{

  "detail": [

    {

      "loc": ["path", "pdf_id"],

      "msg": "Invalid PDF ID",

      "type": "value_error"

    }

  ]

}

```



---



## 5. Query PDF



### POST `/api/v1/query`



**Description**:  

Queries a PDF with a specific question.



### Parameters:

- **pdf_id**: `string` (body) - The ID of the PDF to query.

- **query**: `string` (body) - The question to ask the PDF.



### Request Body:

```json

{

  "pdf_id": "67385818c013d1176861687e",

  "query": "what is the topic?"

}

```



### Example cURL Request:

```bash

curl -X 'POST' \

  'http://127.0.0.1:8000/api/v1/query' \

  -H 'accept: application/json' \

  -H 'Content-Type: application/json' \

  -d '{

    "pdf_id": "67385818c013d1176861687e",

    "query": "what is the topic?"

  }'

```



### Responses:



#### 200 OK

- **Description**: Successfully queries the PDF.

- **Response Body**:

```json

{

  "id": "6738587ac013d1176861687f",

  "pdf_id": "67385818c013d1176861687e",

  "query": "what is the topic?",

  "response": "Open source software development",

  "created_at": "2024-11-16T08:31:54.333289"

}

```



#### 422 Unprocessable Entity

- **Description**: Validation error for request parameters.

- **Response Body**:

```json

{

  "detail": [

    {

      "loc": ["body", "query"],

      "msg": "Query is required",

      "type": "value_error.missing"

    }

  ]

}

```

### 6. Query History



**GET** `/api/v1/history/{pdf_id}`



**Description:**  

Retrieves the query history for a specific PDF.



**Parameters:**  

- `pdf_id` (string, path) - The ID of the PDF.  

- `skip` (integer, query) - Offset for pagination (default: 0).  

- `limit` (integer, query) - Number of records to fetch (default: 10).  



**Responses:**  

- **200 OK:**  

  - **Description:** Successfully retrieves the query history.  

  - **Response Body:**  

    ```json

    [

      {

        "id": "6738587ac013d1176861687f",

        "pdf_id": "67385818c013d1176861687e",

        "query": "what is the topic?",

        "response": "Open source software development",

        "created_at": "2024-11-16T08:31:54.333289"

      },

      {

        "id": "67375818c013d1176861688a",

        "pdf_id": "67385818c013d1176861687e",

        "query": "Who contributed to the document?",

        "response": "John Doe and Jane Smith",

        "created_at": "2024-11-16T09:10:23.123456"

      }

    ]

    ```