Update app.py

app.py

@@ -1,138 +1,99 @@
 import gradio as gr
-import cv2
-import numpy as np
-from PIL import Image, ImageDraw, ImageFont
-import textwrap
-import moviepy.editor as mp
-import moviepy.video.fx.all as vfx
+from openai import OpenAI
+import os
+from fpdf import FPDF
+import docx
+
+# Fetching the access token from environment variables
+ACCESS_TOKEN = os.getenv("HF_TOKEN")
+
+# Initialize the OpenAI client
+client = OpenAI(
+    base_url="https://api-inference.huggingface.co/v1/",
+    api_key=ACCESS_TOKEN,
+)
 
-css = """
-#col-container {
-    margin: 0 auto;
-    max-width: 290px;
+css = '''
+.gradio-container{max-width: 1000px !important}
+h1{text-align:center}
+footer {
+    visibility: hidden
 }
-"""
-
-def create_typing_video(code_text, format_choice, line_spacing, width_choice, height_choice, font_name="arial.ttf", font_size=18, frame_rate=10, sound_choice=None, custom_audio=None, background_color="black", text_color="white", enhance_quality=False, video_speed="1.0"):
-    font_path = f"font/{font_name}"
-    
-    # Convert font_size to integer
-    font_size = int(font_size)
-    font = ImageFont.truetype(font_path, font_size)
-    
-    video_frames = []
-    
-    # Setup initial parameters
-    image_width, image_height = int(width_choice), int(height_choice)
-    max_width = image_width - 40  # Margin of 20 pixels on each side
-    current_text = ""
-    
-    # Create the background
-    background = Image.new("RGB", (image_width, image_height), color=background_color)
-    
-    # Calculate the maximum width and adjust font size if necessary
-    while True:
-        wrapped_lines = textwrap.wrap(code_text, width=max_width // font.getlength(' '))
-        text_height = sum([font.getbbox(line)[3] - font.getbbox(line)[1] for line in wrapped_lines])
-        
-        if text_height <= image_height - 40:
-            break
-        font_size -= 1
-        font = ImageFont.truetype(font_path, font_size)
-    
-    # Generate frames for the typing effect
-    for char in code_text:
-        current_text += char
-
-        if format_choice == "Paragraph":
-            wrapped_lines = textwrap.wrap(current_text, width=max_width // font.getlength(' '))
-        else:  # Programming
-            wrapped_lines = current_text.splitlines()
-
-        # Copy the background image for each frame
-        image = background.copy()
-        draw = ImageDraw.Draw(image)
-        
-        y_position = 20
-        for line in wrapped_lines:
-            draw.text((20, y_position), line, font=font, fill=text_color)
-            line_height = font.getbbox(line)[3] - font.getbbox(line)[1]
-            y_position += line_height * line_spacing
-
-        # Convert to numpy array for OpenCV
-        frame = np.array(image)
-        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
-        video_frames.append(frame)
-    
-    # Create a video writer
-    video_filename = "typed_code_video.mp4"
-    out = cv2.VideoWriter(video_filename, cv2.VideoWriter_fourcc(*"mp4v"), frame_rate, (image_width, image_height))
+'''
+
+# Function to format and respond to the user's message using the Llama model
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
+
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
+
+    messages.append({"role": "user", "content": message})
+
+    response = ""
     
-    for frame in video_frames:
-        out.write(frame)
-    out.release()
-
-    # Adjust video speed
-    speed_factor = {
-        "1x": 1.0,
-        "1.25x": 1.25,
-        "1.5x": 1.5,
-        "1.75x": 1.75,
-        "2x": 2.0
-    }.get(video_speed, 1.0)  # Default to 1.0 if video_speed is not found in the dictionary
-    
-    video = mp.VideoFileClip(video_filename).fx(vfx.speedx, factor=speed_factor)
-    video.write_videofile("speed_adjusted_video.mp4", codec="libx264")
-    video_filename = "speed_adjusted_video.mp4"
-
-    # Add sound if a sound choice is selected
-    if sound_choice and sound_choice != "No Sound":
-        video = mp.VideoFileClip(video_filename)
-        audio = mp.AudioFileClip(f"type-sounds/{sound_choice}")
-        
-        # Loop the audio to match the duration of the video
-        audio = audio.fx(mp.afx.audio_loop, duration=video.duration)
-        video = video.set_audio(audio)
-        video.write_videofile("typed_code_video_with_sound.mp4", codec="libx264")
-        video_filename = "typed_code_video_with_sound.mp4"
-    
-    # Add custom audio if provided
-    if custom_audio:
-        video = mp.VideoFileClip(video_filename)
-        audio = mp.AudioFileClip(custom_audio)
-        
-        # Loop the custom audio to match the duration of the video
-        audio = audio.fx(mp.afx.audio_loop, duration=video.duration)
-        video = video.set_audio(audio)
-        video.write_videofile("typed_code_video_with_custom_audio.mp4", codec="libx264")
-        video_filename = "typed_code_video_with_custom_audio.mp4"
-    
-    # Apply video quality enhancement if enabled
-    if enhance_quality:
-        video = mp.VideoFileClip(video_filename)
-        video = video.fx(vfx.resize, height=720)  # Resize video to enhance quality
-        video = video.fx(vfx.colorx, 1.2)  # Increase contrast
-        video.write_videofile("enhanced_" + video_filename, codec="libx264")
-        video_filename = "enhanced_" + video_filename
-    
-    return video_filename
+    for message in client.chat.completions.create(
+        model="meta-llama/Meta-Llama-3.1-70B-Instruct",
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+        messages=messages,
+    ):
+        token = message.choices[0].delta.content
+        response += token
+        yield response
+
+# Save the generated content to a file with the specified font, font size, and line spacing
+def save_file(content, filename, file_format, font_name, font_size, line_spacing):
+    font_path = f"font/{font_name}"
     
-def generate_video(code_text, format_choice, line_spacing, width_choice, height_choice, font_choice, font_size, sound_choice, custom_audio, background_color, text_color, enhance_quality, video_speed):
-    return create_typing_video(code_text, format_choice, line_spacing, width_choice, height_choice, font_name=font_choice, font_size=font_size, sound_choice=sound_choice, custom_audio=custom_audio, background_color=background_color, text_color=text_color, enhance_quality=enhance_quality, video_speed=video_speed)
-
-# Create Gradio interface
-format_choice = gr.Dropdown(
-    choices=["Paragraph", "Programming"],
-    value="Paragraph",
-    label="Text Format"
-)
-
-line_spacing = gr.Dropdown(
-    choices=[1.0, 1.15, 1.5, 2.0, 2.5, 3.0],
-    value=1.5,
-    label="Line Spacing"
-)
-
+    if file_format == "pdf":
+        pdf = FPDF()
+        pdf.add_page()
+        pdf.set_auto_page_break(auto=True, margin=15)
+        pdf.add_font("CustomFont", "", font_path, uni=True)
+        pdf.set_font("CustomFont", size=int(font_size))
+        for line in content.split("\n"):
+            pdf.multi_cell(0, 10 * float(line_spacing), line)
+        pdf.output(f"{filename}.pdf")
+        return f"{filename}.pdf"
+    elif file_format == "docx":
+        doc = docx.Document()
+        paragraph = doc.add_paragraph(content)
+        run = paragraph.runs[0]
+        run.font.name = font_name.split(".")[0]  # Use the font name without extension
+        run.font.size = docx.shared.Pt(int(font_size))  # Set the font size
+        paragraph_format = paragraph.paragraph_format
+        paragraph_format.line_spacing = float(line_spacing)  # Set the line spacing
+        doc.save(f"{filename}.docx")
+        return f"{filename}.docx"
+    elif file_format == "txt":
+        with open(f"{filename}.txt", "w") as f:
+            f.write(content)
+        return f"{filename}.txt"
+    else:
+        raise ValueError("Unsupported file format")
+
+# Combine respond and save file functions
+def respond_and_save(message, history, system_message, filename="output", file_format="pdf", font_name="arial.ttf", font_size="18", line_spacing="1.5", max_tokens=512, temperature=0.7, top_p=0.95):
+    generated_text = ""
+    for output in respond(message, history, system_message, max_tokens, temperature, top_p):
+        generated_text = output
+    saved_file = save_file(generated_text, filename, file_format, font_name, font_size, line_spacing)
+    return generated_text, history + [(message, generated_text)], saved_file
+
+# Create the font dropdown
 font_choice = gr.Dropdown(
     choices=[
         "DejaVuMathTeXGyre.ttf", 
@@ -152,7 +113,7 @@ font_choice = gr.Dropdown(
         "COOPBL.TTF",
         "Rockwell-Bold.ttf",
         "Candara Light.TTF",
-        "Carlito-Regular.ttf Carlito-Regular.ttf",
+        "Carlito-Regular.ttf",
         "Castellar.ttf",
         "Courier New.ttf",
         "LSANS.TTF",
@@ -167,99 +128,48 @@ font_choice = gr.Dropdown(
         "youyuan.TTF",
         "TfPonetoneExpanded-7BJZA.ttf",
     ],
-    value="SourceCodePro-Black.ttf",
-    label="Currently, it is recommended to use the default font."
+    value="arial.ttf",
+    label="Font Style"
 )
 
+# Create the font size dropdown
 font_size = gr.Dropdown(
-    choices=["16", "18", "20", "22", "24"],
+    choices=["12", "14", "16", "18", "20", "22", "24"],
     value="18",
     label="Font Size"
 )
 
-width_choice = gr.Dropdown(
-    choices=["400","800", "1024", "1280", "1920"],
-    value="800",
-    label="Video Width"
-)
-
-height_choice = gr.Dropdown(
-    choices=["400", "720", "1080", "1440", "2160"],
-    value="400",
-    label="Video Height"
-)
-
-sound_choice = gr.Dropdown(
-    choices=["No Sound",
-             "Mediumspeed Typing.mp3", 
-             "Speed Typing.mp3",
-             "Bass Typing.mp3",
-             "Bay Typing.mp3",
-             "Crack Typing.mp3",
-             "Deep Sence Typing.mp3",
-             "Flacking Typing.mp3",
-             "Flaw Typing.mp3",
-             "Focused Typing.mp3",
-             "K55 Typing.mp3",
-             "Laptop Typing.mp3",
-             "NDC Typing.mp3",
-             "RedMECH Typing.mp3",
-             "Smooth Typing.mp3",
-             "Stop Tpying.mp3",
-            ],
-    value="No Sound",
-    label="Typing Sound"
-)
-custom_audio = gr.File(
-    label="Upload Custom Audio SFX🔊",
-    type="filepath"
-)
-
-background_color = gr.Dropdown(
-    choices=["black", "white", "darkblue", "orange", "green"],
-    value="black",
-    label="Background Color"
-)
-
-text_color = gr.Dropdown(
-    choices=["black", "white", "darkblue", "orange", "green"],
-    value="white",
-    label="Text Color"
-)
-
-enhance_quality = gr.Checkbox(
-    label="Enhance Video Quality"
-)
-
-video_speed = gr.Dropdown(
-    choices=["1x", "1.25x", "1.5x", "1.75x", "2x"],
-    value="1x",
-    label="Video Speed"
+# Create the line spacing dropdown
+line_spacing = gr.Dropdown(
+    choices=[1.0, 1.15, 1.5, 2.0, 2.5, 3.0],
+    value=1.5,
+    label="Line Spacing"
 )
 
-iface = gr.Interface(
-    fn=generate_video,
+# Create Gradio interface
+demo = gr.Interface(
+    fn=respond_and_save,
     inputs=[
-        gr.Textbox(label="Enter Content", lines=10, placeholder="Enter the text to be displayed in the video..."),
-        format_choice,
-        line_spacing,
-        width_choice,
-        height_choice,
+        gr.Textbox(placeholder="Type your message here...", label="Chatbot", lines=5),
+        gr.State(value=[]),  
+        gr.Textbox(placeholder="System message", label="System message", value="", visible=False),
+        gr.Textbox(placeholder="Filename (default: output)", label="Filename", value="output"),
+        gr.Radio(["pdf", "docx", "txt"], label="File Format", value="pdf"),
         font_choice,
-        font_size,
-        sound_choice,
-        custom_audio,
-        background_color,
-        text_color,
-        enhance_quality,
-        video_speed
+        font_size,  # Add the font size dropdown to the interface
+        line_spacing,  # Add the line spacing dropdown to the interface
+        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature", visible=False),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-P", visible=False),
+    ],
+    outputs=[
+        gr.Textbox(label="Generated Text", lines=5),
+        gr.State(value=[]), 
+        gr.File(label="Download File")
     ],
-    
-    outputs=gr.Video(label="Typing Video"),
-    title="Type Bytes🐧",
     css=css,
-    theme="bethecloud/storj_theme",
+    title="GRABDOC PRO",
+    theme="bethecloud/storj_theme"
 )
 
-if __name__ == "__main__": 
-    iface.launch(share=True)
+demo.queue().launch(show_api=False)