Pragyan

AI-Powered DSA Question Solver with Video Explanations

---

🚀 Quick Install

pip install pragyan

---

⚡ Quick Start

In Your Python Script

from pragyan import Pragyan

# Initialize with your API key
solver = Pragyan(api_key="your-gemini-api-key")

# Solve from LeetCode URL
result = solver.solve("https://leetcode.com/problems/two-sum/", language="python")
print(result.solution.code)

From Terminal

# Interactive mode (recommended)
pragyan interactive

# Quick solve from URL
pragyan solve "https://leetcode.com/problems/two-sum/" --video

---

Pragyan is a Python package that helps developers understand and solve Data Structures and Algorithms (DSA) problems. It leverages AI to analyze problems, generate optimized solutions, and create animated explanation videos.

---

Table of Contents

• Features
• Architecture
• Installation
• Quick Start
• CLI Reference
• Python API
• Configuration
• Supported Languages
• Video Generation
• Examples
• Troubleshooting
• Contributing
• License

---

Features

Feature	Description
Multi-Platform Scraping	Extract problems from LeetCode, GeeksforGeeks, Codeforces, HackerRank
AI-Powered Analysis	Identify patterns, data structures, and optimal approaches
Multi-Language Solutions	Generate code in 11+ programming languages
Video Explanations	Animated visualizations using Manim
Step-by-Step Breakdown	Detailed algorithm walkthroughs
Test Case Generation	Auto-generate test cases for validation
Verbose Logging	Real-time progress updates during execution

---

Architecture

System Overview

+------------------------------------------------------------------+
|                         PRAGYAN SYSTEM                            |
+------------------------------------------------------------------+
|                                                                   |
|  +-------------+     +-------------+     +------------------+     |
|  |   INPUT     |     |   CORE      |     |    OUTPUT        |     |
|  +-------------+     +-------------+     +------------------+     |
|  | - URL       |     | - Scraper   |     | - Solution Code  |     |
|  | - Text      |---->| - Analyzer  |---->| - Explanation    |     |
|  | - API Key   |     | - Solver    |     | - Video (MP4)    |     |
|  +-------------+     | - Generator |     | - Test Cases     |     |
|                      +-------------+     +------------------+     |
|                                                                   |
+------------------------------------------------------------------+

Component Architecture

pragyan/
|
+-- cli.py              Command Line Interface (Click + Rich)
|      |
|      +-- Interactive prompts
|      +-- Progress logging
|      +-- Result display
|
+-- main.py             Orchestration Layer
|      |
|      +-- Pragyan class (main entry point)
|      +-- Component coordination
|      +-- Pipeline management
|
+-- scraper.py          Web Scraping Module
|      |
|      +-- LeetCode adapter
|      +-- GFG adapter
|      +-- Generic HTML parser
|      +-- Selenium for dynamic content
|
+-- llm_client.py       AI Integration Layer
|      |
|      +-- Gemini client (Google AI)
|      +-- Groq client (open models)
|      +-- Response parsing
|      +-- Error handling
|
+-- solver.py           Solution Generation
|      |
|      +-- Problem analysis
|      +-- Code generation
|      +-- Complexity analysis
|      +-- Test case creation
|
+-- video_generator.py  Video Creation Module
|      |
|      +-- Manim scene generation
|      +-- Animation templates
|      +-- Rendering pipeline
|
+-- models.py           Data Models (Pydantic)
|      |
|      +-- Question, Solution
|      +-- VideoConfig, LLMConfig
|      +-- ProgrammingLanguage enum
|
+-- logger.py           Logging System
       |
       +-- Rich console output
       +-- Progress tracking
       +-- User engagement

Processing Pipeline

+--------+    +----------+    +----------+    +--------+    +-------+
| INPUT  |--->| SCRAPE   |--->| ANALYZE  |--->| SOLVE  |--->| VIDEO |
+--------+    +----------+    +----------+    +--------+    +-------+
    |              |               |              |             |
    v              v               v              v             v
  URL or       Question        Analysis       Solution      MP4 File
  Text         Object          Dict           Object        
                 |               |              |
                 v               v              v
            - Title         - Topics       - Code
            - Description   - Concept      - Explanation
            - Examples      - Approach     - Complexity
            - Constraints   - Edge Cases   - Steps

Data Flow Diagram

                    +------------------+
                    |    User Input    |
                    | (URL/Text/API)   |
                    +--------+---------+
                             |
                             v
                    +------------------+
                    |  QuestionScraper |
                    |  - Fetch HTML    |
                    |  - Parse content |
                    |  - Extract data  |
                    +--------+---------+
                             |
                             v
                    +------------------+
                    |   LLM Client     |
                    |  - Gemini API    |
                    |  - Groq API      |
                    +--------+---------+
                             |
              +--------------+--------------+
              |                             |
              v                             v
     +----------------+            +----------------+
     |   Analyzer     |            |    Solver      |
     | - Identify     |            | - Generate     |
     |   patterns     |            |   code         |
     | - Find concept |            | - Explain      |
     | - Edge cases   |            |   steps        |
     +-------+--------+            +-------+--------+
              |                             |
              +-------------+---------------+
                            |
                            v
                   +------------------+
                   | Video Generator  |
                   | - Build scenes   |
                   | - Render Manim   |
                   | - Export MP4     |
                   +--------+---------+
                            |
                            v
                   +------------------+
                   |     Output       |
                   | - Code + Explain |
                   | - Video file     |
                   +------------------+

---

Installation

Basic Installation

pip install pragyan

Video Generation Dependencies

For video generation with Manim:

Windows:

pip install manim
choco install ffmpeg

macOS:

pip install manim
brew install ffmpeg

Linux (Ubuntu/Debian):

pip install manim
sudo apt-get install ffmpeg

Verify Installation

pragyan version

---

Quick Start

Interactive Mode (Recommended)

pragyan interactive

This launches a guided session that:

1. Prompts for problem URL or text
2. Selects programming language
3. Configures API provider
4. Displays real-time progress
5. Generates solution and video

Command Line

# Solve from URL
pragyan solve -u https://leetcode.com/problems/two-sum -p gemini -k YOUR_API_KEY

# Solve from text
pragyan solve -t "Find two numbers that add up to target" -l python -p gemini -k KEY

# Analyze only (no solution)
pragyan analyze https://leetcode.com/problems/two-sum -p gemini -k KEY

---

CLI Reference

Command	Description
pragyan interactive	Interactive mode with prompts
pragyan solve	Solve a problem
pragyan analyze	Analyze without solving
pragyan languages	List supported languages
pragyan version	Show version

Solve Options

Option	Short	Description
--url	-u	Problem URL
--text	-t	Problem text
--language	-l	Programming language
--provider	-p	AI provider (gemini/groq)
--api-key	-k	API key
--no-video		Skip video generation
--output-dir	-o	Video output directory
--quality	-q	Video quality (low/medium/high)

---

Python API

Basic Usage

from pragyan import Pragyan

# Initialize
pragyan = Pragyan(
    provider="gemini",
    api_key="YOUR_API_KEY"
)

# Solve from URL
result = pragyan.process(
    "https://leetcode.com/problems/two-sum",
    language="python",
    generate_video=True
)

# Access results
print(result["solution"].code)
print(result["solution"].time_complexity)
print(f"Video: {result['video_path']}")

Advanced Usage

from pragyan import Pragyan, ProgrammingLanguage, VideoConfig

# Custom configuration
video_config = VideoConfig(
    output_dir="./videos",
    video_quality="high_quality",
    resolution="1080p"
)

pragyan = Pragyan(
    provider="gemini",
    api_key="YOUR_API_KEY",
    video_config=video_config
)

# Step-by-step control
question = pragyan.scrape_question("https://leetcode.com/problems/binary-search")
analysis = pragyan.analyze(question)
solution = pragyan.solve(question, ProgrammingLanguage.JAVA, analysis)
video_path = pragyan.generate_video(question, solution, analysis)

# Print details
print(f"Title: {question.title}")
print(f"Topics: {analysis['topics']}")
print(f"Concept: {analysis['main_concept']}")
print(f"Approach: {solution.approach}")
print(f"Time: {solution.time_complexity}")
print(f"Space: {solution.space_complexity}")

Class Reference

Pragyan

Method	Description	Returns
scrape_question(url)	Fetch problem from URL	Question
parse_question(text)	Parse problem from text	Question
analyze(question)	Analyze problem	Dict
solve(question, lang, analysis)	Generate solution	Solution
generate_video(question, solution, analysis)	Create video	Path
process(input, lang, video)	Complete pipeline	Dict
generate_test_cases(question, n)	Generate test cases	List

---

Configuration

Environment Variables

export PRAGYAN_API_KEY="your_api_key"

API Keys

Google Gemini (Recommended):

1. Visit Google AI Studio
2. Create API key
3. Use with provider="gemini"

Groq:

1. Visit Groq Console
2. Create API key
3. Use with provider="groq"

VideoConfig Options

Parameter	Type	Default	Description
output_dir	Path	~/Downloads	Video output directory
video_quality	str	medium_quality	low/medium/high_quality
fps	int	30	Frames per second
pixel_width	int	1920	Video width
pixel_height	int	1080	Video height
background_color	str	#1e1e1e	Background hex color

---

Supported Languages

Language	Aliases
Python	python, py
Java	java
C++	cpp, c++
C	c
JavaScript	javascript, js
TypeScript	typescript, ts
Go	go, golang
Rust	rust, rs
Kotlin	kotlin, kt
Swift	swift
C#	csharp, c#, cs

---

Video Generation

Video Structure

1. Introduction - Problem title and topics
2. Problem Overview - Description and key points
3. Concept Explanation - Core algorithm/technique
4. Step-by-Step Approach - Algorithm walkthrough
5. Code Display - Syntax-highlighted solution
6. Example Walkthrough - Trace through example
7. Complexity Analysis - Time and space complexity
8. Summary - Key takeaways

Quality Settings

Quality	Resolution	Render Time
low	480p	Fast
medium	720p	Moderate
high	1080p	Slow

---

Examples

Two Sum Problem

from pragyan import Pragyan

pragyan = Pragyan(provider="gemini", api_key="KEY")
result = pragyan.process("https://leetcode.com/problems/two-sum")

print(result["solution"].code)
# def twoSum(nums, target):
#     seen = {}
#     for i, num in enumerate(nums):
#         complement = target - num
#         if complement in seen:
#             return [seen[complement], i]
#         seen[num] = i
#     return []

Custom Problem Text

from pragyan import Pragyan

pragyan = Pragyan(provider="groq", api_key="KEY")

problem = """
Given an array of integers, find the maximum subarray sum.

Example:
Input: [-2, 1, -3, 4, -1, 2, 1, -5, 4]
Output: 6 (subarray [4, -1, 2, 1])

Constraints:
- 1 <= length <= 10^5
- -10^4 <= array[i] <= 10^4
"""

question = pragyan.parse_question(problem)
solution = pragyan.solve(question, "cpp")
print(solution.code)

Multiple Languages

from pragyan import Pragyan, ProgrammingLanguage

pragyan = Pragyan(provider="gemini", api_key="KEY")
question = pragyan.scrape_question("https://leetcode.com/problems/binary-search")
analysis = pragyan.analyze(question)

for lang in [ProgrammingLanguage.PYTHON, ProgrammingLanguage.JAVA, ProgrammingLanguage.CPP]:
    solution = pragyan.solve(question, lang, analysis)
    print(f"\n--- {lang.value} ---")
    print(solution.code)

---

Troubleshooting

Common Issues

API Key Error:

Error: Invalid API key

Solution: Verify your API key is correct and active.

Video Generation Fails:

Error: Manim not found

Solution: Install Manim and FFmpeg:

pip install manim
# Windows: choco install ffmpeg
# macOS: brew install ffmpeg
# Linux: sudo apt install ffmpeg

Scraping Fails:

Error: Could not fetch URL

Solution: Some sites require Selenium. Install Chrome/ChromeDriver:

pip install selenium webdriver-manager

---

Project Structure

pragyan/
├── src/
│   └── pragyan/
│       ├── __init__.py
│       ├── main.py
│       ├── cli.py
│       ├── models.py
│       ├── llm_client.py
│       ├── scraper.py
│       ├── solver.py
│       ├── video_generator.py
│       ├── animations.py
│       ├── logger.py
│       └── utils.py
├── tests/
├── examples/
├── pyproject.toml
├── README.md
└── LICENSE

---

Contributing

Contributions are welcome. Please submit a Pull Request.

1. Fork the repository
2. Create a feature branch
3. Make changes
4. Submit PR

---

License

MIT License - see LICENSE file.

---

Links

• Repository: github.com/Kamalllx/pragyan_pypi
• PyPI: pypi.org/project/pragyan
• Issues: github.com/Kamalllx/pragyan_pypi/issues

---

Built by Kamal