Advanced search

On other websites (4928)

• How to run a ffmpeg command in python with file paths containing spaces?

  21 July 2023, by tomatochan

  I'm developing a program in Python that requires the use of FFMPEG. In one step, I need to create a video from a text file containing the list of images (path and duration) that need to be assembled to make a video. To do this, I use the following code:

  


  cmd = [ffmpeg, "-y", "-f", "concat", "-safe", "0", "-i", tmp, "-c:v", "libx264", "-r", "25", "-pix_fmt", "yuv420p", out]
subprocess.run(cmd)
# os.system(" ".join(cmd)) # returns same error


  


  What returns the error: The syntax of the file, directory or volume name is incorrect.

  


  I know that paths can be problematic when they contain spaces, and as my username has one, I've taken care to quote the paths in such a way that :

  


  ffmpeg = "C:\Users\John Doe\Documents\ffmpeg\bin\ffmpeg.exe"
tmp = "C:\Users\John Doe\Desktop\path\to\file.txt"
out = "C:\Users\John Doe\Desktop\path\to\video.mp4"


  


  When I print(" ".join(cmd)), this is what I get in my terminal:

  


  "C:\Users\John Doe\Documentsffmpeg\binffmpeg.exe" -y -f concat -safe 0 -i "C:\Users\John Doe\Desktop\path\to\file.txt" -c:v libx264 -r 25 -pix_fmt yuv420p "C:\Users\John Doe\Desktop\path\to\video.mp4"


  


  However, the problem persists.
Has anyone ever had this problem and managed to solve it?

  


  I've also tried the method where you have to escape the spaces (taking care to replace the \ with / and then the with \ ) but nothing works... The error persists.

  


  For your information, here are the contents of one of my .txt files

  


  file C:\Users\John Doe\Desktop\path\to\image_1.png
duration 0.04
file C:\Users\John Doe\Desktop\path\to\image_2.png
duration 0.04
file C:\Users\John Doe\Desktop\path\toimage_3.png
duration 0.04


  


  When I quote the paths in the .txt files

  


  

  • subprocess gives me the error:
PermissionError: [WinError 5] Access denied

  


  • os.system gives me the error:

  


  


  C:\Users\John' is not recognized as an internal or external command, an executable program or a command file.
The syntax of the file, directory or volume name is incorrect.


  


• How to set the value of an arbitrary pixel in ffmpeg?

  14 May 2017, by Neb

  I want to modify the rgb value of each pixel inside ffmpeg.

  I need to implement the following function:

  A[x + expr_1][y] = expr_2

  where x and y are the current sampled pixel, A is the input frame whose pixel need to be modified and expr_1, expr_2 are two independent expressions.

  I tried using the geq filter but it seems there no way to set the value of a pixel other than from that currently sampled. For istance, the function p(x,y) only returns the value of the pixel at the specified location, but doesn’t allow to set a value for that pixel. In other word, ffmpeg seems allowing only something like:

  A[x][y] = expr

  Is there a way to tell ffmpeg to set the value of a specific pixel?

  Thanks for your time.

• How to specify a non default type when calling a FFmpeg library using ctypes

  1 August 2013, by Jim Ramberg

  I am calling an FFMPEG library called avformat_alloc_context() that returns a pointer to type AVFormatContext. The structure AVFormatContext is defined in the avformat library.

  Obviously; this is not one ctypes default types and results in an error when I try to pass a reference to it to another ffmpeg library further down in my code.

  Is there a way to add in the the classes defined inside of the library ? I read through what documentation I could find online and was not able to find a good answer to this question.