Update app.py

app.py

@@ -1,17 +1,31 @@
 import gradio as gr
 from transformers import T5ForConditionalGeneration, T5Tokenizer
 import pandas as pd
+import pdfplumber
+from io import BytesIO
+import difflib
+import time
 
 # Load the model
 model_name = "ramsrigouthamg/t5_paraphraser"
 tokenizer = T5Tokenizer.from_pretrained(model_name)
 model = T5ForConditionalGeneration.from_pretrained(model_name)
 
+def extract_text_from_file(file):
+    if file.name.endswith('.pdf'):
+        with pdfplumber.open(file) as pdf:
+            return "\n".join([page.extract_text() for page in pdf.pages])
+    elif file.name.endswith(('.txt', '.docx')):
+        return open(file.name, 'r', encoding='utf-8').read()
+    return ""
+
+def calculate_similarity(original, paraphrased):
+    return round(difflib.SequenceMatcher(None, original, paraphrased).ratio() * 100)
+
 def paraphrase_text(text, creativity_level=3, tone="neutral"):
     if not text.strip():
-        return "", "", 0, 0
+        return "", "", 0, 0, 0
     
-    # Adjust generation parameters
     num_beams = 3 + creativity_level
     temperature = 0.7 + (creativity_level * 0.15)
     
@@ -37,161 +51,187 @@ def paraphrase_text(text, creativity_level=3, tone="neutral"):
 
     paraphrased_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True)
     
-    # Calculate word counts
     original_count = len(text.split())
     paraphrased_count = len(paraphrased_text.split())
+    similarity = calculate_similarity(text, paraphrased_text)
     
-    return paraphrased_text, original_count, paraphrased_count
+    return paraphrased_text, original_count, paraphrased_count, similarity
+
+def process_input(text, file, creativity, tone):
+    if file is not None:
+        text = extract_text_from_file(file)
+    return paraphrase_text(text, creativity, tone)
 
-def create_comparison(original, paraphrased, orig_count, para_count):
-    comparison = pd.DataFrame({
-        "Metric": ["Text", "Word Count"],
-        "Original": [original, orig_count],
-        "Paraphrased": [paraphrased, para_count]
+def create_comparison(original, paraphrased, orig_count, para_count, similarity):
+    return pd.DataFrame({
+        "Metric": ["Text", "Word Count", "Similarity"],
+        "Original": [original, orig_count, f"{similarity}%"],
+        "Paraphrased": [paraphrased, para_count, f"{similarity}%"]
     })
-    return comparison
 
-# Custom HTML with Tailwind CSS
-custom_html = """
-<script src="https://cdn.tailwindcss.com"></script>
-<style>
-  .gradio-container {
+custom_css = """
+:root {
+    --primary: #4f46e5;
+    --primary-dark: #4338ca;
+    --text: #1f2937;
+    --light-bg: #f9fafb;
+}
+.gradio-container {
     font-family: 'Inter', sans-serif;
-    max-width: 100% !important;
-  }
-  .gradio-interface {
-    padding: 0 !important;
-  }
-  .gradio-input, .gradio-output {
-    border: 1px solid #e5e7eb !important;
-    border-radius: 0.5rem !important;
-  }
-  .gradio-input textarea, .gradio-output textarea {
-    padding: 1rem !important;
-  }
-  .gradio-button {
-    background-color: #4f46e5 !important;
-    color: white !important;
-    border-radius: 0.375rem !important;
-    padding: 0.5rem 1rem !important;
-    font-weight: 500 !important;
-  }
-  .gradio-button:hover {
-    background-color: #4338ca !important;
-  }
-  .gradio-slider .gradio-slider-handle {
-    background: #4f46e5 !important;
-  }
-  .gradio-slider .gradio-slider-bar {
-    background: #c7d2fe !important;
-  }
-</style>
+    max-width: 1200px !important;
+    margin: 0 auto !important;
+}
+.header {
+    background: linear-gradient(135deg, var(--primary) 0%, var(--primary-dark) 100%);
+    border-radius: 12px 12px 0 0;
+}
+.card {
+    background: white;
+    border-radius: 12px;
+    box-shadow: 0 4px 24px rgba(0,0,0,0.08);
+    padding: 24px;
+    margin-bottom: 24px;
+}
+.stats-card {
+    background: var(--light-bg);
+    border-radius: 12px;
+    padding: 16px;
+}
+.gradio-button {
+    background: var(--primary) !important;
+    transition: all 0.2s ease !important;
+}
+.gradio-button:hover {
+    background: var(--primary-dark) !important;
+    transform: translateY(-1px);
+}
+.file-upload {
+    border: 2px dashed #d1d5db !important;
+    border-radius: 8px !important;
+    padding: 20px !important;
+}
 """
 
-with gr.Blocks(theme=gr.themes.Base(), css=custom_html) as demo:
-    # Header Section
-    gr.HTML("""
-    <div class="bg-indigo-600 text-white py-6 px-4 rounded-t-lg">
-      <div class="max-w-4xl mx-auto text-center">
-        <h1 class="text-3xl font-bold mb-2">Professional AI Paraphraser</h1>
-        <p class="text-indigo-100">Transform your text while preserving meaning with advanced T5 AI</p>
-      </div>
-    </div>
-    """)
+with gr.Blocks(theme=gr.themes.Soft(), css=custom_css) as demo:
+    # Header
+    with gr.Column(elem_classes=["header"]):
+        gr.HTML("""
+        <div class="text-center py-8 text-white">
+            <h1 class="text-3xl font-bold mb-2">Advanced AI Paraphraser Pro</h1>
+            <p class="text-indigo-100 opacity-90">Enterprise-grade text transformation with semantic analysis</p>
+        </div>
+        """)
     
     # Main Content
-    with gr.Row().style(equal_height=True):
+    with gr.Row():
+        # Input Panel
         with gr.Column(scale=1):
-            # Input Section
-            gr.HTML("""<div class="p-4 bg-white rounded-lg shadow-sm">""")
-            input_text = gr.Textbox(
-                lines=8,
-                placeholder="Enter text to paraphrase...",
-                label="Original Text",
-                elem_classes=["border", "rounded-lg", "p-4", "w-full"]
-            ).style(container=False)
-            
-            with gr.Row():
-                creativity = gr.Slider(
-                    1, 5, value=3,
-                    label="Creativity Level",
-                    info="1 = Conservative, 5 = Highly Creative",
-                    interactive=True
+            with gr.Column(elem_classes=["card"]):
+                gr.Markdown("### Input Text")
+                input_text = gr.Textbox(
+                    lines=8,
+                    placeholder="Paste your content here...",
+                    label="Original Text",
+                    elem_id="input-text"
                 )
-                tone = gr.Dropdown(
-                    ["neutral", "formal", "casual", "academic"],
-                    value="neutral",
-                    label="Output Tone",
-                    interactive=True
+                
+                # File Upload
+                file_upload = gr.File(
+                    label="Or upload document (PDF/TXT)",
+                    file_types=[".pdf", ".txt", ".docx"],
+                    elem_classes=["file-upload"]
+                )
+                
+                with gr.Row():
+                    creativity = gr.Slider(
+                        1, 5, value=3,
+                        label="Creativity Level",
+                        interactive=True,
+                        info="Higher values produce more creative rewrites"
+                    )
+                    tone = gr.Dropdown(
+                        ["neutral", "formal", "casual", "academic", "professional"],
+                        value="professional",
+                        label="Output Tone",
+                        interactive=True
+                    )
+                
+                submit_btn = gr.Button(
+                    "Paraphrase Content",
+                    variant="primary",
+                    size="lg"
                 )
-            
-            submit_btn = gr.Button(
-                "Paraphrase",
-                variant="primary"
-            ).style(full_width=True)
-            gr.HTML("""</div>""")
         
+        # Output Panel
         with gr.Column(scale=1):
-            # Output Section
-            gr.HTML("""<div class="p-4 bg-white rounded-lg shadow-sm h-full">""")
-            output_text = gr.Textbox(
-                lines=8,
-                label="Paraphrased Text",
-                interactive=False,
-                elem_classes=["border", "rounded-lg", "p-4", "w-full"]
-            ).style(container=False)
-            
-            # Stats Section
-            gr.HTML("""
-            <div class="mt-4 p-4 bg-gray-50 rounded-lg">
-              <h3 class="font-medium text-gray-900 mb-3">Statistics</h3>
-              <div class="grid grid-cols-3 gap-4">
-            """)
-            
-            with gr.Row():
-                orig_count = gr.Number(
-                    label="Original Words",
-                    interactive=False
-                ).style(container=False)
-                para_count = gr.Number(
-                    label="Paraphrased Words",
-                    interactive=False
-                ).style(container=False)
-                similarity = gr.Number(
-                    label="Similarity Score",
+            with gr.Column(elem_classes=["card"]):
+                gr.Markdown("### Paraphrased Output")
+                output_text = gr.Textbox(
+                    lines=8,
+                    label="Rewritten Text",
+                    interactive=True,
+                    elem_id="output-text"
+                )
+                
+                # Action Buttons
+                with gr.Row():
+                    copy_btn = gr.Button("Copy to Clipboard", size="sm")
+                    download_btn = gr.Button("Download Result", size="sm")
+                
+                # Stats Card
+                with gr.Column(elem_classes=["stats-card"]):
+                    gr.Markdown("**Text Analysis**")
+                    with gr.Row():
+                        orig_count = gr.Number(label="Original Words", precision=0)
+                        para_count = gr.Number(label="Paraphrased Words", precision=0)
+                        similarity = gr.Number(label="Similarity Score", suffix="%")
+    
+    # Advanced Analysis
+    with gr.Column(elem_classes=["card"]):
+        with gr.Tabs():
+            with gr.TabItem("Detailed Comparison"):
+                comparison_table = gr.DataFrame(
+                    headers=["Metric", "Original", "Paraphrased"],
+                    datatype=["str", "str", "str"],
                     interactive=False,
-                    visible=False  # Placeholder for future feature
-                ).style(container=False)
+                    wrap=True
+                )
             
-            gr.HTML("""</div></div></div>""")
-    
-    # Comparison Section
-    gr.HTML("""<div class="mt-4 p-4 bg-white rounded-lg shadow-sm">""")
-    with gr.Accordion("Detailed Comparison", open=False):
-        comparison_table = gr.DataFrame(
-            headers=["Metric", "Original", "Paraphrased"],
-            datatype=["str", "str", "str"],
-            interactive=False,
-            elem_classes=["w-full"]
-        )
-    gr.HTML("""</div>""")
+            with gr.TabItem("Text Diff"):
+                diff_display = gr.HTML()
     
     # Footer
     gr.HTML("""
-    <div class="mt-6 py-4 text-center text-gray-500 text-sm">
-      <p>Powered by T5 Transformer | Embeddable in any website</p>
+    <div class="text-center py-4 text-gray-500 text-sm">
+        <p>© 2024 AI Paraphraser Pro | Scholar Writer</p>
     </div>
     """)
     
-    # Event handlers
+    # Event Handlers
     submit_btn.click(
-        fn=paraphrase_text,
-        inputs=[input_text, creativity, tone],
-        outputs=[output_text, orig_count, para_count]
+        fn=process_input,
+        inputs=[input_text, file_upload, creativity, tone],
+        outputs=[output_text, orig_count, para_count, similarity]
     ).then(
         fn=create_comparison,
-        inputs=[input_text, output_text, orig_count, para_count],
+        inputs=[input_text, output_text, orig_count, para_count, similarity],
         outputs=comparison_table
+    ).then(
+        fn=lambda o, p: difflib.HtmlDiff().make_table(o.splitlines(), p.splitlines()),
+        inputs=[input_text, output_text],
+        outputs=diff_display
+    )
+    
+    copy_btn.click(
+        fn=None,
+        inputs=output_text,
+        js="(text) => { navigator.clipboard.writeText(text); return 'Copied!'; }"
+    )
+    
+    download_btn.click(
+        fn=lambda t: (t, "paraphrased_result.txt"),
+        inputs=output_text,
+        outputs=gr.File(label="Downloading...", visible=False)
     )
 
 if __name__ == "__main__":