Recherche avancée

Sur d’autres sites (6438)

• ffmpeg video streaming issue

  20 avril, par Personboiii

  I am trying to embed an adb video stream into an html site with flask, and the code I have keeps on returning this same error :

  


  FFmpeg: [mjpeg @ 0x156631c10] Could not find codec parameters for stream 0 (Video: mjpeg, none(bt470bg/unknown/unknown)): unspecified size
FFmpeg: Consider increasing the value for the 'analyzeduration' (1000000) and 'probesize' (5000000) options
FFmpeg: Input #0, mjpeg, from 'pipe:0':
FFmpeg:   Duration: N/A, bitrate: N/A
FFmpeg:   Stream #0:0: Video: mjpeg, none(bt470bg/unknown/unknown), 25 fps, 1200k tbr, 1200k tbn
FFmpeg: Output #0, mpegts, to 'pipe:1':
FFmpeg: [out#0/mpegts @ 0x156632110] Output file does not contain any stream
FFmpeg: Error opening output file pipe:1.
FFmpeg: Error opening output files: Invalid argument


  


  this is my code :

  


  
from flask import Flask, Response
import subprocess
import json
import threading

app = Flask(__name__)

with open("data_file.json", "r") as f:
    config_data = json.load(f)

user = config_data["Users"]["Test User 1"]


def log_ffmpeg_errors(proc):
    for line in iter(proc.stderr.readline, b''):
        if line:
            print("FFmpeg:", line.decode(), end='')


def connect_device(ip, port):
    try:
        # Reconnect if device is offline
        subprocess.run(["adb", "tcpip", str(port)])
        subprocess.run(["adb", "connect", ip])
        # Check if the device is online
        devices = subprocess.check_output(["adb", "devices"]).decode()
        if "offline" in devices:
            raise Exception("Device is offline")
    except Exception as e:
        print(f"Error connecting device: {e}")


def generate_video_stream():
    adb_cmd = ["adb", "exec-out", "screenrecord", "--output-format=mjpeg"]  # Use MJPEG output
    ffmpeg_cmd = [
        "ffmpeg",
        "-f", "mjpeg",
        "-analyzeduration", "1000000", 
        "-probesize", "5000000",  
        "-i", "pipe:0", 
        "-q:v", "5",
        "-r", "10",
        "-vcodec", "mjpeg",  
        "-s", "1280x720", 
        "-f", "mpegts", 
        "pipe:1" 
    ]

    adb_proc = subprocess.Popen(adb_cmd, stdout=subprocess.PIPE)
    ffmpeg_proc = subprocess.Popen(ffmpeg_cmd, stdin=adb_proc.stdout, stdout=subprocess.PIPE, stderr=subprocess.PIPE)

    threading.Thread(target=log_ffmpeg_errors, args=(ffmpeg_proc,), daemon=True).start()

    try:
        while True:
            frame = ffmpeg_proc.stdout.read(4096)
            if not frame:
                break
            yield (b'--frame\r\n'
                   b'Content-Type: image/jpeg\r\n\r\n' + frame + b'\r\n')

    finally:
        adb_proc.terminate()
        ffmpeg_proc.terminate()

@app.route('/video_feed')
def video_feed():
    return Response(generate_video_stream(), mimetype='multipart/x-mixed-replace; boundary=frame')

if __name__ == "__main__":
    connect_device(user["IP"], user["port"])
    app.run(debug=True, host='0.0.0.0', port=8080)



  


  I also changed it so that it is adb_cmd = ["adb","exec-out","screenrecord", "-output-format=h264","-"] and the error left but now the site header just keeps on loading and the embed in the html shows nothing. (ngrok for the site page says 200 ok)

  


• How to Use SVG Image Files Directly in FFmpeg ? [closed]

  10 mars, par Pubg Mobile

  I generated a bar chart race using the Flourish Studio website and captured the frames as a PDF sequence using a Python Playwright script. Then, I converted all the PDF files into SVG format using the following Python script, because SVG is the only image format that maintains quality without loss when zoomed in :

  


  import os
import subprocess
import multiprocessing

# Define paths
pdf2svg_path = r"E:\Desktop\dist-64bits\pdf2svg.exe"  # Full path to pdf2svg.exe
input_dir = r"E:\Desktop\New folder (4)\New folder"
output_dir = r"E:\Desktop\New folder (4)\New folder (2)"

# Ensure output directory exists
os.makedirs(output_dir, exist_ok=True)

def convert_pdf_to_svg(pdf_file):
    """ Convert a single PDF file to SVG. """
    input_pdf = os.path.join(input_dir, pdf_file)
    output_svg = os.path.join(output_dir, os.path.splitext(pdf_file)[0] + ".svg")

    try:
        subprocess.run([pdf2svg_path, input_pdf, output_svg], check=True)
        print(f"Converted: {pdf_file} -> {output_svg}")
    except FileNotFoundError:
        print(f"Error: Could not find {pdf2svg_path}. Make sure the path is correct.")
    except subprocess.CalledProcessError:
        print(f"Error: Conversion failed for {pdf_file}")

if __name__ == "__main__":
    # Get list of PDF files in input directory
    pdf_files = [f for f in os.listdir(input_dir) if f.lower().endswith(".pdf")]

    # Use multiprocessing to speed up conversion
    with multiprocessing.Pool(processes=multiprocessing.cpu_count()) as pool:
        pool.map(convert_pdf_to_svg, pdf_files)

    print("All conversions completed!")


  


  Problem :

  


  Now, I want to use these SVG images in FFmpeg to create a high-quality video file. However, FFmpeg does not support SVG files directly, and I have read that I must convert SVG files into PNG before using them in FFmpeg. The problem is that PNG images reduce quality, especially when zooming in, which is why I want to avoid converting to PNG.

  


  Is there any way to use SVG files directly in FFmpeg or another method to convert them into a high-quality video while maintaining full resolution ? Any ideas or suggestions would be greatly appreciated !

  


• How to Capture a Sequence of High-Quality PDF Frames from a Website (Without Screen Recording) ?

  9 mars, par Pubg Mobile

  In Firefox, I can take very high-quality screenshots of a webpage by using Ctrl + P and saving the page as a PDF. This method preserves the text, images, and code in excellent resolution.

  


  Now, I have created a movable bar chart race in Flourish Studio and want to convert it into a high-quality video. However, I do not want to use screen recording tools.

  


  My Goal :
  
I want to capture 30 high-resolution PDF frames from the website at different points in time (like a video sequence). Ideally, I need a tool or script that can automate the process of saving multiple PDFs from the website as it plays the animation.

  


  What I Tried :
  
I attempted to write a Python script that :

  


  Opens the local HTML file of my Flourish chart in Firefox using Selenium.
  
Waits for the page to load.
  
Listens for the F1 key and triggers Ctrl + P to print the page as a PDF.
  
However, the script does not save the PDF file in the output folder. I'm not sure why.

  


  Here is my code :

  


  import time
import keyboard
from selenium import webdriver
from selenium.webdriver.firefox.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.firefox.options import Options

# Define paths
html_file_path = r"E:\Desktop\New folder (4)\20250309101616805.html"
geckodriver_path = r"E:\Desktop\New folder (4)\geckodriver.exe"
save_path = r"E:\Desktop\New folder (4)\New folder\output.pdf"  # Save PDF location

# Set up Firefox options
options = Options()
options.set_preference("print.always_print_silent", True)  # Silent printing
options.set_preference("print.show_print_progress", False)  # Hide progress
options.set_preference("print.print_to_file", True)  # Print to file
options.set_preference("print.save_print_settings", True)  # Save settings
options.set_preference("print.printer_PDF", "Save as PDF")  # Set printer
options.set_preference("print.print_to_file", True)  # Enable saving print output to file
options.set_preference("print.print_file_name", save_path)  # Define the save location for PDF

# Start WebDriver
service = Service(executable_path=geckodriver_path)
driver = webdriver.Firefox(service=service, options=options)

# Open the HTML file
driver.get("file:///" + html_file_path)

# Wait for the page to load
time.sleep(2)

print("Press F1 to save as PDF.")

# Listen for F1 key press
while True:
    if keyboard.is_pressed('F1'):
        print("F1 pressed, saving as PDF...")
        
        # Trigger print command (Ctrl + P)
        body = driver.find_element(By.TAG_NAME, 'body')
        body.send_keys(Keys.CONTROL + 'p')
        
        # Wait for the print dialog to process
        time.sleep(2)

        print("PDF should be saved to:", save_path)
        break

# Close browser
driver.quit()


  


  My Questions :

  


  Why is my script not saving the PDF in the specified output folder ?

  


  Is there a better way to automate capturing 30 sequential PDFs from the website at different animation frames ?

  


  Is there any tool or script that can generate a sequence of PDFs (like 30 frames per second) from a webpage ?

  


  Important :

  


  I do NOT want to use screen recording tools.

  


  I only need high-quality PDF frames that can later be converted into a video.

  


  Any help would be greatly appreciated !