Recherche avancée

Sur d’autres sites (4134)

• avformat/au : Avoid allocation for metadata string

  13 juillet 2020, par Andreas Rheinhardt

  avformat/au : Avoid allocation for metadata string
  When there are potentially annotation (i.e. metadata) fields to write,
  
  au_get_annotations() is called to produce a string with them. To do so,
  
  it uses an AVBPrint which is finalized to create the string. This is
  
  wasteful, because it always leads to an allocation even if the string
  
  actually fits into the internal buffer of the AVBPrint. This commit
  
  changes this by making au_get_annotations() modify an AVBPrint that
  
  resides on the stack of the caller (i.e. of au_write_header()).
  Furthermore, the AVBPrint is now checked for truncation ; limiting
  
  the allocations implicit in the AVBPrint allowed to offload the overflow
  
  checks. Notice that these were not correct before : The size parameter of
  
  avio_write() is an int, yet the string in the AVBPrint was allowed to
  
  grow bigger than INT_MAX. And if the length of the string was so near
  
  UINT_MAX that the length + 32 overflowed, the old code would write the
  
  first eight bytes of the string and nothing more, leading to an invalid
  
  file.
  Finally, the special case in which the metadata dictionary of the
  
  AVFormatContext is empty (in which case one still has to write eight
  
  binary zeroes) is now no longer treated specially, because this case
  
  no longer incurs any allocation.
  Signed-off-by : Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
  

  • [DH] libavformat/au.c

• ffmpeg - filter_complex list too long

  2 mars 2016, par Baumi

  Let’s say I want to overlay a clock in the video using special font, color, etc to video that is aprox 30 min long. I end up with command :

  ffmpeg -y -i in.mp4 -filter_complex "
  
  [0:v]drawtext=fontfile=/var/www/sites/manage/elements/digital-7.ttf:text='00\:00':fontcolor=white@1.0:fontsize=26:x=100:y=65:enable='between(t,0,7)'[tmp];
  
  [tmp]drawtext=fontfile=/var/www/sites/manage/elements/digital-7.ttf:text='00\:01':fontcolor=white@1.0:fontsize=26:x=100:y=65:enable='between(t,7,8)'[tmp]; 
  
  [tmp]drawtext=fontfile=/var/www/sites/manage/elements/digital-7.ttf:text='00\:02':fontcolor=white@1.0:fontsize=26:x=100:y=65:enable='between(t,8,9)'[tmp]; 
  
  [tmp]drawtext=fontfile=/var/www/sites/manage/elements/digital-7.ttf:text='00\:03':fontcolor=white@1.0:fontsize=26:x=100:y=65:enable='between(t,9,10)'[tmp]; 
  
  [tmp]drawtext=fontfile=/var/www/sites/manage/elements/digital-7.ttf:text='00\:04':fontcolor=white@1.0:fontsize=26:x=100:y=65:enable='between(t,10,11)'[tmp];
  
  ......." 
  
  -map "[tmp]" -map 0:a -acodec copy -c:v h264 out.mp4

  This clock is not the only overlay I have so finally I have end up with command 216kB long but this I cannot even run in bash because of argument list being too long.

  I wanted to re-encode the video only once. Is there any other way I can do that ?

  Thanks !

• Use FFMPEG to put images separately inside a "box" and keeping their original positions of the X and Y boundaries and maybe modify their offsets

  14 septembre 2020, par karl-police

  I have a collection of images, for this example, I created 3 images that have different sizes, but the image itsself is the same, except it is moved more down or left or right.

  


  1 : 2 : 3 :

  


  These are the images. The total sizes of all images together is 35x39, so that means that they need to go inside an 35x39 image so it can later be used to craft them into a GIF as example. Since "crop" does not really work as it crops it smaller and can't make them bigger and I can't imagine it being the best solution for that anyway, perhaps.

  


  So this is the invisible 35x39 sized box.

  


  image :

  


  So what I'm trying to do is to figure out how I can put each of these images separately in the 35x39 sized box, but maintaining the original X position or the Y position or both, from the boundaries of the images. I'm trying to figure out how I can do this for other transparent images for similar things, mostly used to craft animations. Here into a GIF out of image collections, but the images need to be fixed first.

  


  I tried to look in the FFMPEG documentation, but there's many many filters and etc. that I had a few issues finding the right thing. I'm also not sure if it is also then possible to change the offset of the X and Y, because I think if there's something to make it keep the original position of X and Y, then I also think that there's probably something to change the offset of it aswell.

  


   

  


  End result of the images could basically be :

  


  1 : 2 : 3 :

  


  This end result example of the images, basically have their X aligned on the top and Y aligned on the left. I'm not sure if you can call it "original X position", because if I compare it to Photoshop's special paste and keep at original position, it puts the first image a bit more down as example, for some reason. So I just moved the X all the way to the top and the Y all the way to the left.