video_mcp / app.py
jomasego's picture
Fix Modal deployment using correct FastAPI endpoint pattern and update frontend
b046b1d
raw
history blame
23.8 kB
import gradio as gr
import os
import requests
import tempfile
import subprocess
import re
import shutil # Added for rmtree
import modal
from typing import Dict, Any, Optional # Added for type hinting
def is_youtube_url(url_string: str) -> bool:
"""Checks if the given string is a YouTube URL."""
# More robust regex to find YouTube video ID, accommodating various URL formats
# and additional query parameters.
youtube_regex = (
r'(?:youtube(?:-nocookie)?\.com/(?:[^/\n\s]+/|watch(?:/|\?(?:[^&\n\s]+&)*v=)|embed(?:/|\?(?:[^&\n\s]+&)*feature=oembed)|shorts/|live/)|youtu\.be/)'
r'([a-zA-Z0-9_-]{11})' # This captures the 11-character video ID
)
# We use re.search because the video ID might not be at the start of the query string part of the URL.
# re.match only matches at the beginning of the string (or beginning of line in multiline mode).
# The regex now directly looks for the 'v=VIDEO_ID' or youtu.be/VIDEO_ID structure.
# The first part of the regex matches the domain and common paths, the second part captures the ID.
return bool(re.search(youtube_regex, url_string))
def download_video(url_string: str, temp_dir: str) -> str | None:
"""Downloads video from a URL (YouTube or direct link) to a temporary directory."""
if is_youtube_url(url_string):
print(f"Attempting to download YouTube video: {url_string}")
# Define a fixed output filename pattern within the temp_dir
output_filename_template = "downloaded_video.%(ext)s" # yt-dlp replaces %(ext)s
output_path_template = os.path.join(temp_dir, output_filename_template)
cmd = [
"yt-dlp",
"-f", "bestvideo[ext=mp4]+bestaudio[ext=m4a]/mp4/best", # Prefer mp4 format
"--output", output_path_template,
url_string
]
print(f"Executing yt-dlp command: {' '.join(cmd)}")
try:
result = subprocess.run(cmd, capture_output=True, text=True, timeout=300, check=False)
print(f"yt-dlp STDOUT:\n{result.stdout}")
print(f"yt-dlp STDERR:\n{result.stderr}")
if result.returncode == 0:
# Find the actual downloaded file based on the template
downloaded_file_path = None
for item in os.listdir(temp_dir):
if item.startswith("downloaded_video."):
potential_path = os.path.join(temp_dir, item)
if os.path.isfile(potential_path):
downloaded_file_path = potential_path
print(f"YouTube video successfully downloaded to: {downloaded_file_path}")
break
if downloaded_file_path:
return downloaded_file_path
else:
print(f"yt-dlp seemed to succeed (exit code 0) but the output file 'downloaded_video.*' was not found in {temp_dir}.")
return None
else:
print(f"yt-dlp failed with return code {result.returncode}.")
return None
except subprocess.TimeoutExpired:
print(f"yt-dlp command timed out after 300 seconds for URL: {url_string}")
return None
except Exception as e:
print(f"An unexpected error occurred during yt-dlp execution for {url_string}: {e}")
return None
elif url_string.startswith(('http://', 'https://')) and url_string.lower().endswith(('.mp4', '.mov', '.avi', '.mkv', '.webm')):
print(f"Attempting to download direct video link: {url_string}")
try:
response = requests.get(url_string, stream=True, timeout=300) # 5 min timeout
response.raise_for_status() # Raises HTTPError for bad responses (4XX or 5XX)
filename = os.path.basename(url_string) or "downloaded_video_direct.mp4"
video_file_path = os.path.join(temp_dir, filename)
with open(video_file_path, 'wb') as f:
for chunk in response.iter_content(chunk_size=8192):
f.write(chunk)
print(f"Direct video downloaded successfully to: {video_file_path}")
return video_file_path
except requests.exceptions.RequestException as e:
print(f"Error downloading direct video link {url_string}: {e}")
return None
except Exception as e:
print(f"An unexpected error occurred during direct video download for {url_string}: {e}")
return None
else:
print(f"Input '{url_string}' is not a recognized YouTube URL or direct video link for download.")
return None
def process_video_input(input_string: str) -> Dict[str, Any]:
"""
Processes the video (from URL or local file path) and returns its transcription status as a JSON object.
"""
if not input_string:
return {
"status": "error",
"error_details": {
"message": "No video URL or file path provided.",
"input_received": input_string
}
}
video_path_to_process = None
# Get base_modal_url and construct modal_endpoint_url
base_modal_url = os.getenv("MODAL_APP_BASE_URL")
if not base_modal_url:
print("ERROR: MODAL_APP_BASE_URL environment variable not set.")
return {
"status": "error",
"error_details": {
"message": "Modal application base URL is not configured. Please set the MODAL_APP_BASE_URL environment variable.",
"input_received": input_string
}
}
modal_endpoint_url = f"{base_modal_url.rstrip('/')}/analyze_video"
print(f"Target Modal endpoint: {modal_endpoint_url}")
response_json = None # Initialize to ensure it's always defined before return
try:
if input_string.startswith(('http://', 'https://')):
print(f"Input is a URL: {input_string}. Sending URL to Modal endpoint as JSON.")
payload = {"video_url": input_string}
headers = {'Content-Type': 'application/json'}
response = requests.post(modal_endpoint_url, json=payload, headers=headers, timeout=1860)
elif os.path.exists(input_string):
print(f"Input is a local file path: {input_string}. Sending file content to Modal endpoint.")
video_path_to_process = input_string # Use input_string as the path
try:
with open(video_path_to_process, "rb") as video_file:
video_bytes_content = video_file.read()
print(f"Read {len(video_bytes_content)} bytes from video file '{video_path_to_process}'.")
files = {'video_file': (os.path.basename(video_path_to_process), video_bytes_content, 'video/mp4')}
response = requests.post(modal_endpoint_url, files=files, timeout=1860)
except FileNotFoundError: # Catch if file disappears just before open
print(f"Error: Video file not found at {video_path_to_process} when trying to read for upload.")
return { # Return immediately
"status": "error",
"error_details": {
"message": "Video file disappeared before it could be read for upload.",
"path_attempted": video_path_to_process
}
}
else:
# This handles cases where input_string is neither a URL nor an existing file path
print(f"Input '{input_string}' is not a valid URL or an existing file path.")
return { # Return immediately
"status": "error",
"error_details": {
"message": f"Input '{input_string}' is not a valid URL or an existing file path.",
"input_received": input_string
}
}
# Common response handling
response.raise_for_status() # Raise an exception for HTTP errors (4xx or 5xx)
analysis_results = response.json()
print(f"Received results from Modal endpoint: {str(analysis_results)[:200]}...")
response_json = {
"status": "success",
"data": analysis_results
}
except requests.exceptions.Timeout:
print(f"Request to Modal endpoint {modal_endpoint_url} timed out.")
response_json = {
"status": "error",
"error_details": {
"message": "Request to video analysis service timed out.",
"endpoint_url": modal_endpoint_url
}
}
except requests.exceptions.HTTPError as e:
print(f"HTTP error calling Modal endpoint {modal_endpoint_url}: {e.response.status_code} - {e.response.text}")
response_json = {
"status": "error",
"error_details": {
"message": f"Video analysis service returned an error: {e.response.status_code}",
"details": e.response.text,
"endpoint_url": modal_endpoint_url
}
}
except requests.exceptions.RequestException as e: # General request exception
print(f"Error calling Modal endpoint {modal_endpoint_url}: {e}") # Corrected MODAL_ENDPOINT_URL to modal_endpoint_url
response_json = {
"status": "error",
"error_details": {
"message": "Failed to connect to video analysis service.",
"details": str(e),
"endpoint_url": modal_endpoint_url # Corrected MODAL_ENDPOINT_URL to modal_endpoint_url
}
}
except Exception as e: # Catch-all for other unexpected errors
print(f"An unexpected error occurred in process_video_input: {e}")
import traceback
traceback.print_exc()
response_json = {
"status": "error",
"error_details": {
"message": f"An unexpected error occurred: {str(e)}",
"exception_type": type(e).__name__
}
}
return response_json
def process_video_input_new(input_string: str) -> Dict[str, Any]:
"""
Processes the video (from URL or local file path) and returns its transcription status as a JSON object.
"""
if not input_string:
return {
"status": "error",
"error_details": {
"message": "No video URL or file path provided.",
"input_received": input_string
}
}
video_path_to_process = None
# Get base_modal_url and construct modal_endpoint_url
base_modal_url = os.getenv("MODAL_APP_BASE_URL")
if not base_modal_url:
print("ERROR: MODAL_APP_BASE_URL environment variable not set.")
return {
"status": "error",
"error_details": {
"message": "Modal application base URL is not configured. Please set the MODAL_APP_BASE_URL environment variable.",
"input_received": input_string
}
}
modal_endpoint_url = base_modal_url.rstrip('/')
print(f"Using Modal endpoint URL: {modal_endpoint_url}")
try:
if input_string.startswith("http://") or input_string.startswith("https://"):
# Send URL as JSON payload to the Modal backend
payload = {"video_url": input_string}
print(f"Sending video URL as JSON payload: {payload}")
response = requests.post(modal_endpoint_url, json=payload, timeout=1860)
else:
# Local file path - still need to send as JSON for now (until we support file uploads)
return {"status": "error", "error_details": {"message": "Local file upload not yet supported. Please provide a video URL."}}
response.raise_for_status()
result = response.json()
print(f"Modal backend response: {result}")
return result
except requests.exceptions.HTTPError as e:
error_msg = f"HTTP {e.response.status_code}: {e.response.text[:200] if e.response else 'Unknown error'}"
print(f"HTTP error: {error_msg}")
return {"status": "error", "error_details": {"message": f"Video analysis service returned an error: {e.response.status_code}", "details": error_msg, "endpoint_url": modal_endpoint_url}}
except requests.exceptions.RequestException as e:
print(f"Request error: {e}")
return {"status": "error", "error_details": {"message": "Failed to connect to video analysis service", "details": str(e), "endpoint_url": modal_endpoint_url}}
except Exception as e:
print(f"Unexpected error: {e}")
return {"status": "error", "error_details": {"message": "Unexpected error during video analysis", "details": str(e), "endpoint_url": modal_endpoint_url}}
# Gradio Interface for the API endpoint
api_interface = gr.Interface(
fn=process_video_input_new,
inputs=gr.Textbox(lines=1, label="Video URL or Local File Path for Interpretation",
placeholder="Enter YouTube URL, direct video URL (.mp4, .mov, etc.), or local file path..."),
outputs=gr.JSON(label="API Response"),
title="Video Interpretation Input",
description="Provide a video URL or local file path to get its interpretation status as JSON.",
flagging_options=None,
examples=[
["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],
["https://commondatastorage.googleapis.com/gtv-videos-bucket/sample/BigBuckBunny.mp4"]
]
)
# Gradio Interface for a simple user-facing demo
def demo_process_video(input_string: str) -> tuple[str, Dict[str, Any]]:
"""
A simple demo function for the Gradio UI.
It calls process_video_input and unpacks its result for separate display.
"""
result = process_video_input(input_string)
status_str = result.get("status", "Unknown Status")
# The second part of the tuple should be the 'data' if successful,
# or the 'error_details' (or the whole result) if there was an error.
if status_str == "success" and "data" in result:
details_json = result["data"]
elif "error_details" in result:
details_json = result["error_details"]
else: # Fallback, show the whole result
details_json = result
return status_str, details_json
def call_topic_analysis_endpoint(topic_str: str, max_vids: int) -> Dict[str, Any]:
"""Calls the Modal FastAPI endpoint for topic-based video analysis."""
if not topic_str:
return {"status": "error", "error_details": {"message": "Topic cannot be empty."}}
if not (1 <= max_vids <= 10): # Max 10 as defined in FastAPI endpoint, can adjust
return {"status": "error", "error_details": {"message": "Max videos must be between 1 and 10."}}
base_modal_url = os.getenv("MODAL_APP_BASE_URL")
if not base_modal_url:
print("ERROR: MODAL_APP_BASE_URL environment variable not set.")
return {
"status": "error",
"error_details": {
"message": "Modal application base URL is not configured. Please set the MODAL_APP_BASE_URL environment variable."
}
}
topic_endpoint_url = f"{base_modal_url.rstrip('/')}/analyze_topic"
params = {"topic": topic_str, "max_videos": max_vids}
print(f"Calling Topic Analysis endpoint: {topic_endpoint_url} with params: {params}")
try:
# Using POST as defined in modal_whisper_app.py for /analyze_topic
response = requests.post(topic_endpoint_url, params=params, timeout=3660) # Long timeout for multiple videos
response.raise_for_status()
results = response.json()
print(f"Received results from Topic Analysis endpoint: {str(results)[:200]}...")
return results # The endpoint should return the aggregated JSON directly
except requests.exceptions.Timeout:
print(f"Request to Topic Analysis endpoint {topic_endpoint_url} timed out.")
return {"status": "error", "error_details": {"message": "Request to topic analysis service timed out."}}
except requests.exceptions.HTTPError as e:
print(f"HTTP error calling Topic Analysis endpoint {topic_endpoint_url}: {e.response.status_code} - {e.response.text}")
return {"status": "error", "error_details": {"message": f"Topic analysis service returned an error: {e.response.status_code}", "details": e.response.text}}
except requests.exceptions.RequestException as e:
print(f"Error calling Topic Analysis endpoint {topic_endpoint_url}: {e}")
return {"status": "error", "error_details": {"message": "Failed to connect to topic analysis service.", "details": str(e)}}
except Exception as e:
print(f"An unexpected error occurred: {e}")
return {"status": "error", "error_details": {"message": "An unexpected error occurred during topic analysis call.", "details": str(e)}}
demo_interface = gr.Interface(
fn=demo_process_video,
inputs=gr.Textbox(lines=1, label="Video URL or Local File Path", placeholder="Enter YouTube URL, direct video URL, or local file path...", scale=3),
outputs=[gr.Textbox(label="Status"), gr.JSON(label="Comprehensive Analysis Output", scale=2)],
title="Video Interpretation Demo",
description="Provide a video URL or local file path to see its transcription status.",
flagging_options=None
)
js_code_for_head = """
console.log('[MCP Script] Initializing script to change API link text...');
let foundAndChangedGlobal = false; // Declare here to be accessible in setInterval
function attemptChangeApiLinkText() {
const links = document.querySelectorAll('a');
// console.log('[MCP Script] Found ' + links.length + ' anchor tags on this attempt.');
for (let i = 0; i < links.length; i++) {
const linkText = links[i].textContent ? links[i].textContent.trim() : '';
if (linkText === 'Use via API' || linkText === 'Share via Link') { // Target both possible texts
links[i].textContent = 'Use as an MCP or via API';
console.log('[MCP Script] Successfully changed link text from: ' + linkText);
foundAndChangedGlobal = true;
return true; // Indicate success
}
}
return false; // Indicate not found/changed in this attempt
}
let attempts = 0;
const maxAttempts = 50; // Try for up to 5 seconds (50 * 100ms)
let initialScanDone = false;
const intervalId = setInterval(() => {
if (!initialScanDone && attempts === 0) {
console.log('[MCP Script] Performing initial scan for API link text.');
initialScanDone = true;
}
if (attemptChangeApiLinkText() || attempts >= maxAttempts) {
clearInterval(intervalId);
if (attempts >= maxAttempts && !foundAndChangedGlobal) {
console.log('[MCP Script] Max attempts reached. Target link was not found or changed. It might not be rendered or has a different initial text.');
}
}
attempts++;
}, 100);
"""
# Combine interfaces into a Blocks app
with gr.Blocks(head=f"<script>{js_code_for_head}</script>") as app:
gr.Markdown("# LLM Video interpretation MCP")
gr.Markdown("This Hugging Face Space acts as a backend for processing video context for AI models.")
with gr.Tab("API Endpoint (for AI Models)"):
gr.Markdown("### Use this endpoint from another application (e.g., another Hugging Face Space).")
gr.Markdown("The `process_video_input` function (for video interpretation) is exposed here.")
api_interface.render()
gr.Markdown("**Note:** Some YouTube videos may fail to download if they require login or cookie authentication due to YouTube's restrictions. Direct video links are generally more reliable for automated processing.")
with gr.Tab("Interactive Demo"):
gr.Markdown("### Test the Full Video Analysis Pipeline")
gr.Markdown("Enter a video URL or local file path to get a comprehensive JSON output including transcription, caption, actions, and objects.")
input_text = gr.Textbox(lines=1, label="Video URL or Local File Path", placeholder="Enter YouTube URL, direct video URL, or local file path...", scale=3)
output_json = gr.JSON(label="Comprehensive Analysis Output", scale=2)
with gr.Column(scale=1):
submit_btn = gr.Button("Submit", variant="primary")
clear_btn = gr.Button("Clear")
# Define functions for button actions
def handle_submit(input_text):
if not input_text.strip():
return "Please enter a video URL or file path."
return process_video_input_new(input_text.strip())
def handle_clear():
return "", ""
# Connect button events
submit_btn.click(fn=handle_submit, inputs=input_text, outputs=output_json)
clear_btn.click(fn=handle_clear, outputs=[input_text, output_json])
# Example inputs
gr.Examples(
examples=[
"https://www.youtube.com/watch?v=dQw4w9WgXcQ",
"https://sample-videos.com/zip/10/mp4/SampleVideo_1280x720_1mb.mp4"
],
inputs=input_text
)
gr.Markdown("**Processing can take several minutes** depending on video length and model inference times. The cache on the Modal backend will speed up repeated requests for the same video.")
with gr.Tab("Demo (for Manual Testing)"):
gr.Markdown("### Manually test video URLs or paths for interpretation and observe the JSON response.")
demo_interface.render()
with gr.Tab("Topic Video Analysis"):
gr.Markdown("### Analyze Multiple Videos Based on a Topic")
gr.Markdown("Enter a topic, and the system will search for relevant videos, analyze them, and provide an aggregated JSON output.")
with gr.Row():
topic_input = gr.Textbox(label="Enter Topic", placeholder="e.g., 'best cat videos', 'Python programming tutorials'", scale=3)
max_videos_input = gr.Number(label="Max Videos to Analyze", value=3, minimum=1, maximum=5, step=1, scale=1) # Max 5 for UI, backend might support more
topic_analysis_output = gr.JSON(label="Topic Analysis Results")
with gr.Row():
topic_submit_button = gr.Button("Analyze Topic Videos", variant="primary")
topic_clear_button = gr.Button("Clear")
topic_submit_button.click(
fn=call_topic_analysis_endpoint,
inputs=[topic_input, max_videos_input],
outputs=[topic_analysis_output]
)
def clear_topic_outputs():
return [None, 3, None] # topic_input, max_videos_input (reset to default), topic_analysis_output
topic_clear_button.click(fn=clear_topic_outputs, inputs=[], outputs=[topic_input, max_videos_input, topic_analysis_output])
gr.Examples(
examples=[
["AI in healthcare", 2],
["sustainable energy solutions", 3],
["how to make sourdough bread", 1]
],
inputs=[topic_input, max_videos_input],
outputs=topic_analysis_output,
fn=call_topic_analysis_endpoint,
cache_examples=False
)
gr.Markdown("**Note:** This process involves searching for videos and then analyzing each one. It can take a significant amount of time, especially for multiple videos. The backend has a long timeout, but please be patient.")
# Launch the Gradio application
if __name__ == "__main__":
app.launch(debug=True, server_name="0.0.0.0")