Recherche avancée

Sur d’autres sites (12949)

• swscale/arm : re-enable neon rgbx to nv12 routines

  22 février 2016, par Xiaolei Yu

  swscale/arm : re-enable neon rgbx to nv12 routines
  Commit ’842b8f4ba2e79b9c004a67f6fdb3d5c5d05805d3’ fixed clang/iphone
  
  build but failed on some versions of cygwin. It has now been verified
  
  to work on both platforms.
  Signed-off-by : Michael Niedermayer <michael@niedermayer.cc>
  

  • [DH] libswscale/arm/Makefile
  • [DH] libswscale/arm/swscale_unscaled.c

• Respecter les rotations des vidéos

  2 mai 2011

  Les vidéos prises avec des dispositifs mobiles (Apple Iphone notamment) disposent d’une métadata de Rotation que Mediainfo peut récupérer.

  Il serait donc mieux d’inverser les hauteurs / largeurs à ce moment là et d’encoder avec rotation ... Peut être

  cf :

  • http://superuser.com/questions/1622...
  • http://www.ffmpeg.org/ffmpeg-doc.ht...
  • https://github.com/danielgtaylor/qt...
  • http://mediaspip.ptitdeb.infini.fr/... (pour la video source)

• How to make an automated video of changing webpage

  21 mars 2023, par jonas wouters

  I'm currently working on a project where I need to make a recording of a webpage without opening this browser. (headless browser)
The file I'm working on is stored locally on my machine and is generated with a Python script. It's generated because it will be different for every user and will be deleted after the recording is made.

  


  I'm currently stuck with trying to make a recording of a webpage.
Does somebody know how I can record a webpage.

  


  Currently I'm doing this

  


  # Make a video
def create_video(duration, name):
    # Path of the HTML file you want to record
    html_file_path = os.path.join(py_dir_path, 'templates/video/', f'{name}.html')
    width = 1080
    height = 1920
    framerate = 30
    
    options = Options()
    options.headless = True
    options.add_experimental_option('mobileEmulation', {'deviceName': 'iPhone SE'})
    driver = webdriver.Chrome(options=options)
    driver.get(f'file://{html_file_path}')
    print(driver)
    outputName = f'{name}.mp4'

    cmd= f'ffmpeg -y -f xcbgrab -s {width}x{height} -framerate {framerate} -i :0.0+0,0 -f alsa -i default -vcodec libx264 -pix_fmt yuv420p -preset ultrafast {outputName}'
    p = subprocess.Popen(cmd, shell=True)

    import time
    time.sleep(duration)

    p.kill()


  


  This code makes a headless browser and plays the website, but the recording process is not working for me

  


  I've already have had working code, but this was based on making screenshots of the webpage and then pasting these screenshots after each other, this code was difficult to read and worst of all very slow.

  


  Working bad code

  


  # Make a video
def create_video(duration, name, timesFaster):
    # Path of the HTML file you want to record
    html_file_path = os.path.join(py_dir_path, 'templates/video/', f'{name}.html')
    # Use function create_driver to create a driver and use this driver
    try:
        # Make a chrome window with size --> width: 390px, height: 850px
        options = webdriver.ChromeOptions()
        options.add_argument("--headless")
        options.add_experimental_option('mobileEmulation', {'deviceName': 'iPhone SE'})
        driver = webdriver.Chrome(options=options)
        driver.get(f'file://{html_file_path}')
        
        # Use function capture_screenshots to take screenshots for 
        capture_screenshots(driver, int(duration), name, timesFaster)
    finally:
        driver.quit

# Make as many screens as possible in ... amount of time (... = animation_duration)
def capture_screenshots(driver, animation_duration, name, timesFaster):
    screenshots = []
    # Calculate the ending time
    end_time = time.time() + animation_duration
    # Keeps track of amount of screenshots
    index = 1

    try:
        # Take as many screenshots as possible until the current time exceeds the end_time
        while time.time() < end_time:
            # Each time a new filename (so it does not overwrite)
            screenshot_file_name = f'capture{index}.png'
            # Save the screenshot on device
            driver.save_screenshot(screenshot_file_name)
            # Append the screenshot in screenshots ([])
            screenshots.append(screenshot_file_name)
            index += 1
        
        
        # Calculate the FPS
        fps = (len(screenshots)/animation_duration) * timesFaster
        print("sec: ", animation_duration/timesFaster)
        print("fps: ", fps)
        # Make the video with the FPS calculated above
        clip = ImageSequenceClip(screenshots, fps=fps)
        # File name of the result (video)
        output_file_path = os.path.join(mp4_dir_path, f"part/{name}.mp4")
        # Write the videoFile to the system
        clip.write_videofile(output_file_path, codec='libx264', bitrate="2M")
    finally:
        # Delete all screenshots
        for screenshot in screenshots:
            try:
                os.remove(screenshot)
            except:
                pass


  


  At the moment it's not that important for me that it's a local file, if I would be able to record a webpage (for example https://jonaswouters.sinners.be/3d-animatie/) this will be equally helpfull

  


  Thanks in advance
Jonas