File size: 14,566 Bytes
fa4e803
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
import streamlit as st
import os
import requests
import hashlib
from typing import List, Dict, Any
from datetime import datetime
import json
import re
from urllib.parse import quote
import time
import random
import functools

# Import required libraries
from crewai import Agent, Task, Crew, Process
from crewai.tools import BaseTool
import nltk
from textstat import flesch_reading_ease, flesch_kincaid_grade
from bs4 import BeautifulSoup
import concurrent.futures
from duckduckgo_search import DDGS

# Import Ollama and LangChain components
from langchain_community.chat_models import ChatOllama
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

# Download NLTK data
try:
    nltk.download('punkt', quiet=True)
    nltk.download('stopwords', quiet=True)
    nltk.download('wordnet', quiet=True)
except:
    pass

# Custom Tools for CrewAI
class WebSearchTool(BaseTool):
    name: str = "web_search"
    description: str = "Search the web for content to check plagiarism"

    def _run(self, query: str) -> str:
        """Search the web using DuckDuckGo with rate limiting"""
        try:
            # Add delay to avoid overwhelming the search API
            time.sleep(1)

            with DDGS() as ddgs:
                results = list(ddgs.text(query, max_results=5))  # Reduced from 10 to 5
                search_results = []
                for result in results:
                    search_results.append({
                        'title': result.get('title', ''),
                        'body': result.get('body', ''),
                        'url': result.get('href', '')
                    })
                return json.dumps(search_results)
        except Exception as e:
            return f"Search failed: {str(e)}"

class TextAnalysisTool(BaseTool):
    name: str = "text_analysis"
    description: str = "Analyze text for readability and quality metrics"

    def _run(self, text: str) -> str:
        """Analyze text quality"""
        try:
            # Calculate readability scores
            flesch_score = flesch_reading_ease(text)
            fk_grade = flesch_kincaid_grade(text)

            # Word count and sentence analysis
            words = text.split()
            sentences = text.split('.')

            analysis = {
                'word_count': len(words),
                'sentence_count': len(sentences),
                'avg_words_per_sentence': len(words) / max(len(sentences), 1),
                'flesch_reading_ease': flesch_score,
                'flesch_kincaid_grade': fk_grade,
                'readability_level': self._get_readability_level(flesch_score)
            }

            return json.dumps(analysis)
        except Exception as e:
            return f"Analysis failed: {str(e)}"

    def _get_readability_level(self, score):
        if score >= 90: return "Very Easy"
        elif score >= 80: return "Easy"
        elif score >= 70: return "Fairly Easy"
        elif score >= 60: return "Standard"
        elif score >= 50: return "Fairly Difficult"
        elif score >= 30: return "Difficult"
        else: return "Very Difficult"

class PlagiarismChecker(BaseTool):
    name: str = "plagiarism_checker"
    description: str = "Check text for potential plagiarism by comparing with web content"

    def _run(self, text: str, search_results: str) -> str:
        """Check for plagiarism by comparing text with search results"""
        try:
            results = json.loads(search_results)
            text_sentences = [s.strip() for s in text.split('.') if s.strip()]

            plagiarism_results = []
            total_sentences = len(text_sentences)
            flagged_sentences = 0

            for sentence in text_sentences:
                if len(sentence.split()) < 5:  # Skip very short sentences
                    continue

                similarity_found = False
                for result in results:
                    content = result.get('body', '') + ' ' + result.get('title', '')

                    # Simple similarity check
                    if self._calculate_similarity(sentence, content) > 0.7:
                        similarity_found = True
                        flagged_sentences += 1
                        plagiarism_results.append({
                            'sentence': sentence,
                            'source': result.get('url', 'Unknown'),
                            'similarity_score': self._calculate_similarity(sentence, content)
                        })
                        break

            plagiarism_score = (flagged_sentences / max(total_sentences, 1)) * 100

            return json.dumps({
                'plagiarism_score': plagiarism_score,
                'total_sentences': total_sentences,
                'flagged_sentences': flagged_sentences,
                'flagged_content': plagiarism_results[:3]  # Return top 3 matches
            })
        except Exception as e:
            return f"Plagiarism check failed: {str(e)}"

    def _calculate_similarity(self, text1: str, text2: str) -> float:
        """Calculate basic similarity between two texts"""
        words1 = set(text1.lower().split())
        words2 = set(text2.lower().split())

        if not words1 or not words2:
            return 0.0

        intersection = words1.intersection(words2)
        union = words1.union(words2)

        return len(intersection) / len(union) if union else 0.0

# Rate limit handling decorator (can be kept for other potential API calls, though not strictly needed for local Ollama)
def rate_limit_handler(max_retries=5, base_delay=2, max_delay=60):
    def decorator(func):
        @functools.wraps(func)
        def wrapper(*args, **kwargs):
            for attempt in range(max_retries):
                try:
                    return func(*args, **kwargs)
                except Exception as e:
                    error_message = str(e).lower()
                    if "rate_limit" in error_message or "429" in error_message:
                        if attempt < max_retries - 1:
                            delay = min(max_delay, base_delay * (2 ** attempt) + random.uniform(0, 1))
                            st.warning(f"Rate limit hit. Retrying in {delay:.1f} seconds... (Attempt {attempt + 1}/{max_retries})")
                            time.sleep(delay)
                        else:
                            st.error(f"Max retries reached for rate limit: {e}")
                            raise e
                    else:
                        raise e
            return None
        return wrapper
    return decorator

# Custom LLM class for CrewAI with Ollama
# Removed GroqLLM and replaced with direct ChatOllama usage

# Simplified agents for better token management
def create_agents(llm):
    """Create specialized agents for different tasks"""

    # Combined Analysis Agent (combines plagiarism and analysis)
    analysis_agent = Agent(
        role="Content Analyzer",
        goal="Analyze text for plagiarism and quality metrics",
        backstory="You are an expert in content analysis and plagiarism detection.",
        tools=[WebSearchTool(), PlagiarismChecker(), TextAnalysisTool()],
        verbose=True,
        allow_delegation=False,
        llm=llm
    )

    # Paraphrasing Agent
    paraphrasing_agent = Agent(
        role="Content Rewriter",
        goal="Rewrite text to be original while maintaining meaning",
        backstory="You are an expert writer who creates original content.",
        verbose=True,
        allow_delegation=False,
        llm=llm
    )

    return analysis_agent, paraphrasing_agent

def create_tasks(input_text, agents):
    """Create simplified tasks for the agents"""
    analysis_agent, paraphrasing_agent = agents

    # Truncate input text if too long
    if len(input_text.split()) > 350:
        words = input_text.split()
        input_text = ' '.join(words[:350]) + "..."

    # Task 1: Combined Analysis
    analysis_task = Task(
        description=f"""
        Analyze this text briefly:

        Text: {input_text}

        Provide:
        1. Basic plagiarism check
        2. Readability score
        3. Word count

        Keep response under 200 words.
        """,
        agent=analysis_agent,
        expected_output="Brief analysis with plagiarism score and readability metrics"
    )

    # Task 2: Paraphrasing
    paraphrasing_task = Task(
        description=f"""
        Rewrite this text to be original:

        Original: {input_text}

        Requirements:
        1. Maintain meaning
        2. Use different words
        3. Keep it clear and readable

        Provide only the rewritten text.
        """,
        agent=paraphrasing_agent,
        expected_output="Paraphrased text that maintains original meaning",
        dependencies=[analysis_task]
    )

    return [analysis_task, paraphrasing_task]

def run_crew_analysis(input_text, selected_model):
    """Run the simplified CrewAI analysis"""
    try:
        # Initialize LLM with Ollama
        # Ensure Ollama server is running and the model is pulled (e.g., ollama run llama2)
        llm = ChatOllama(model=selected_model)

        # Create agents
        agents = create_agents(llm)

        # Create tasks
        tasks = create_tasks(input_text, agents)

        # Create crew
        crew = Crew(
            agents=list(agents),
            tasks=tasks,
            process=Process.sequential,
            verbose=True
        )

        # Execute the crew with progress tracking
        with st.spinner("Analyzing text with AI agents..."):
            result = crew.kickoff()

        return result
    except Exception as e:
        st.error(f"Error in crew analysis: {str(e)}")
        return None

# Streamlit UI
def main():
    st.set_page_config(
        page_title="AI Paraphrasing & Plagiarism Checker",
        page_icon="πŸ€–",
        layout="wide"
    )

    st.title("πŸ€– AI-Powered Paraphrasing & Plagiarism Checker")
    st.markdown("*Built with CrewAI Multi-Agent Framework and Ollama (Local LLM)*")

    # Sidebar for configuration
    with st.sidebar:
        st.header("πŸ”§ Configuration")

        # Removed Groq API Key input

        # Model selection for Ollama
        st.markdown("**Ollama Setup:**\n\n1.  Download and install Ollama from [ollama.ai](https://ollama.ai/).\n2.  Run `ollama run <model_name>` in your terminal (e.g., `ollama run llama2` or `ollama run mistral`).\n3.  Ensure the Ollama server is running before using this app.")

        model_options = [
            "llama2", # A good general-purpose model
            "mistral", # Another strong contender
            "phi3", # Smaller, faster model for local use
            # Add other Ollama models as needed
        ]

        selected_model = st.selectbox(
            "Select Ollama Model",
            model_options,
            index=0,  # Default to llama2
            help="Choose an Ollama model you have pulled locally."
        )

        st.markdown("---")
        st.markdown("### πŸ“Š Features")
        st.markdown("- Smart plagiarism detection")
        st.markdown("- Intelligent paraphrasing")
        st.markdown("- Readability analysis")
        st.markdown("- Local LLM support (Ollama)")

    # Main content area
    col1, col2 = st.columns([1, 1])

    with col1:
        st.header("πŸ“ Input Text")

        # Text length warning
        st.info("πŸ’‘ For best results, keep text under 400 words")

        # Text input
        input_text = st.text_area(
            "Enter text to analyze and paraphrase:",
            height=300,
            placeholder="Paste your text here (max 400 words recommended)..."
        )

        # Show word count
        if input_text:
            word_count = len(input_text.split())
            if word_count > 400:
                st.warning(f"⚠️ Text has {word_count} words. Consider shortening for optimal results.")
            else:
                st.success(f"βœ… Text has {word_count} words!")

        # Analysis button
        if st.button("πŸš€ Analyze & Paraphrase", type="primary", use_container_width=True):
            if not input_text.strip():
                st.error("Please enter some text to analyze!")
            else:
                # Run analysis with selected Ollama model
                result = run_crew_analysis(input_text, selected_model)

                if result:
                    st.session_state.analysis_result = result
                    st.session_state.original_text = input_text
                    st.success("βœ… Analysis completed!")

    with col2:
        st.header("πŸ“Š Analysis Results")

        if "analysis_result" in st.session_state:
            result = st.session_state.analysis_result

            # Display results in tabs
            tab1, tab2 = st.tabs(["πŸ“ Paraphrased Text", "πŸ“ˆ Analysis"])

            with tab1:
                st.subheader("πŸ“ Paraphrased Text")

                # Display paraphrased text
                paraphrased_text = str(result)

                st.text_area(
                    "Paraphrased version:",
                    value=paraphrased_text,
                    height=300,
                    help="This is the AI-generated paraphrased version"
                )

                # Download button
                st.download_button(
                    label="πŸ“₯ Download Paraphrased Text",
                    data=paraphrased_text,
                    file_name="paraphrased_text.txt",
                    mime="text/plain"
                )

            with tab2:
                st.subheader("πŸ“ˆ Analysis Summary")

                # Display quick stats
                original_words = len(st.session_state.original_text.split())
                paraphrased_words = len(str(result).split())

                col_a, col_b = st.columns(2)
                with col_a:
                    st.metric("Original Words", original_words)
                    st.metric("Processing Status", "βœ… Complete")

                with col_b:
                    st.metric("Paraphrased Words", paraphrased_words)
                    st.metric("Model Used", selected_model)

                # Simple comparison chart
                st.bar_chart({
                    "Original": [original_words],
                    "Paraphrased": [paraphrased_words]
                })
        else:
            st.info("πŸ‘ˆ Enter text and click 'Analyze & Paraphrase' to see results")



if __name__ == "__main__":
    main()