Recherche avancée

Sur d’autres sites (8483)

• uploading video to ftp while transcoding with Xuggler

  27 novembre 2013, par Holly

  I'm trying to transcode a video on a client PC (Win, Linux, IOS - x32 or x64 - all 6) and write the output directly to a ftp server using Xuggler. With pure ffmpeg it would look like this :

  ffmpeg -i "local.mp4" -ftp-write-seekable 0 -c:v libx264 -crf 25 -f flv ftp://user:pass@server.net:1234/uploaded.flv

  I'm assuming that it must be possible since ffmpeg can do it and Xuggler is supposed to be a wrapper for ffmpeg. I got it working using exec("ffmeg") but it needs to work on all 6 OS mentioned above.

  I tried to addapt this example : https://groups.google.com/forum/# !msg/xuggler-users/QeFTxqgc8Bg/0j1ntsl3tI0J by just using a ftp://user:pass@server.net:1234/uploaded.flvas the url but that does not work. Xuggler is unable to create a container based on such an url.

  I guess I should be able to write into an OutputStream and have ftp4j read from that stream and write to ftp. What would I need to consider for that ?

  Failing all else I could write into a temp file and upload that, but I really don't like that as a solution.

  1. Can someone please provide me with a code example of how to make Xuggler write to ftp directly ?

  2. As a side question. I'm still looking into how to best capture data for displaying the progress in the GUI. So far I found : IMediaListener onWritePacket(IWritePacketEvent event) which I would register to the writer. However I'm not sure how to get the details ffmped reports (current compression, output size so far). The only useful info I found so far is the TimeStamp. Any help on that would also be most welcome.

• Merge commit '0c0b87f12d48d4e7f0d3d13f9345e828a3a5ea32'

  28 septembre 2017, par James Almer

  Merge commit '0c0b87f12d48d4e7f0d3d13f9345e828a3a5ea32'
  * commit '0c0b87f12d48d4e7f0d3d13f9345e828a3a5ea32' :
  
    aarch64 : vp9itxfm : Fix incorrect vertical alignment
  
    aarch64 : vp9itxfm : Update a comment to refer to a register with a different name
  
    aarch64 : vp9itxfm : Use the right lane sizes in 8x8 for improved readability
  
    aarch64 : vp9itxfm : Use a single lane ld1 instead of ld1r where possible
  
    aarch64 : vp9itxfm : Share instructions for loading idct coeffs in the 8x8 function
  
    arm : vp9itxfm : Share instructions for loading idct coeffs in the 8x8 function
  
    aarch64 : vp9itxfm : Do separate functions for half/quarter idct16 and idct32
  
    arm : vp9itxfm : Do a simpler half/quarter idct16/idct32 when possible
  
    aarch64 : vp9itxfm : Move the load_add_store macro out from the itxfm16 pass2 function
  
    arm : vp9itxfm : Move the load_add_store macro out from the itxfm16 pass2 function
  
    aarch64 : vp9itxfm : Make the larger core transforms standalone functions
  
    arm : vp9itxfm : Make the larger core transforms standalone functions
  This commit is a noop, see
  
  f8fcee0daf42d16581ec581bd5ff6373c3a76103
  
  dc47bf3872fde18f6aea33f015a95b70e24dff19
  
  3bd9b39108076e1fca8cd26970cb946fce66523a
  
  a681c793a30386d01d273ce86b3368311cffb511
  
  824589556cb7c4bfafb8a0190e71a10c628f5339
  
  9532a7d4d060f2a2741225a76945daed52dbc478
  
  1d8ab576a7e40ea6209dd753929f26dcb37e1444
  
  3006e5253afc6a0ee001916806aae80f157d7043
  
  19a0f9529ccdb48696f0caa251fe36b1d30df739
  
  6752318c737663f0ac019de3acd63e3cea706864
  
  d0fbf7f34e7243f6a4cbde9925f34d924e6d93f7
  
  16ef000799b227d0226b7a678d28c34ff1d09410
  Merged-by : James Almer <jamrial@gmail.com>
  

• avcodec/x86 : cleanup simple_idct10

  1er juin 2017, par James Darnley

  avcodec/x86 : cleanup simple_idct10
  Use named arguments for the functions so we can remove a define.  The
  
  stride/linesize argument is now ptrdiff_t type so we no longer need to
  
  sign extend the register.
  

  • [DH] libavcodec/x86/proresdsp.asm
  • [DH] libavcodec/x86/simple_idct10.asm
  • [DH] libavcodec/x86/simple_idct10_template.asm