Recherche avancée

Sur d’autres sites (4682)

• Anomalie #4325 : Fatal error "undefined spip_versions_sqlite2" pendant restoration

  2 mai 2019, par jluc -

  phpinfo indique :

  • SQLite 3.x driver for PDO Wez Furlong
  • SQLite3 Scott MacVicar, Ilia Alshanetsky, Brad Dewar
  • API Extensions mysqli,pdo_mysql,mysql
  • PDO drivers mysql, pgsql
  • PDO Driver for MySQL enabled
  • Client API version mysqlnd 5.0.11-dev - 20120503 - $Id : 76b08b24596e12d4553bd41fc93cccd5bac2fe7a $

  Donc, dans dump_type_serveur, spip_versions_sqlite3() appelle _sqlite_charger_version(3).
  Ça teste les extensions, et là ;

  • extension_loaded(’pdo’)=true
  • extension_loaded(’pdo_sqlite’)=false
    Donc ça ne renvoie rien et dump_type_serveur enchaîne sur sqlite2 dont le fichier de définitions n’est pas chargé dans req.

• A Docker Ubuntu image with ARM Mali G610 GPU support [closed]

  7 novembre 2023, par Тимур Тимаев

  I have a docker-compose project which automates creation of GFX-video files used in TV production. The program takes an order from a user, takes screenshots of various web-pages on the internet, composes these screenshots into a html-based animation and renders this animation in a video file. User gets broadcast ready .MP4 file.

  


  I currently export animation from html by extracting every single frame as a PNG sequence and then stitch them together using FFMPEG. However there's a much faster way to do this - record the html animation as it is playing in a browser.

  


  The issue is that the animation is choppy due to the lack of GPU support in a Docker container. I'm using Seleniarm/Chrome and it's not hardware accelerated.

  


  I'm trying to run this project on a OrangePI 5 Plus SBC which features Rockchip RK3588 SoC with ARM Mali G610 GPU. I have found a Ubuntu image by Joshua Riek which implements GPU acceleration the best in my opinion. It's very smooth. Joshua adds SoC GPU drivers to the Ubuntu image thus adding the gpu support.

  


  I want to build a Docker image from this repo to be able to spin it up, install Chrome, Selenium to the container, play the html animation and record it via x11grab device of FFMPEG.

  


  My problem is that I have little knowledge about building containers myself. Let alone tinkering with container drivers and etc. I'm not even sure if it's possible at all.

  


  Can anybody help me or point me in the right direction ?

  


  I have found some links, but I have absolutely no clue for where to start from.
I've looked into Seleniarm but some discussions mentioned that it uses XVFB which can not be hardware accelerated by definition.

  


• NodeJS server that converts m4a audio blobs into mp3 files works locally, but not on server

  14 juillet 2023, par Joris508

  I'm experiencing issues when running my NodeJS server on a remote host. The server is supposed to convert webm and m4a audio files to mp3 files using the ffmpeg-installer and fluent-ffmpeg (npm) packages.

  


  The original files are created using the MediaRecorder Web API. Chrome supports the 'audio/webm' (webm) mimeType, but Safari only supports 'audio/mp4' (m4a). When I run my node server locally, it converts both the m4a files and the webm files just fine. When I host my node server, the webm files are converted without problems, but the m4a files specifically return as 1 second audio files. They're all 8398 bytes in size. You can hear the start of the recording in the 1 second file as well. I'm completely lost on what the issue could be, especially since it works fine when the node server runs locally (through the terminal).

  


  Does anyone have any ideas ? I checked both the node version as well as all the packages ; they are identical. The server doesn't give any errors or feedback. It behaves as if everything is going fine.

  


  I simplified the code and selected just the part that converts the file. Here's the code that converts the m4a file to an mp3. It's triggered 2 seconds after the server starts up (the file is already there on startup in this testing scenario) :

  


  const convertFunc = () => {
    ffmpeg('./public/testfile.m4a')
        .output('./public/testfile.mp3')
        .on('error', (err) => {
            console.error('Error: ' + err.message);
        })
        .on('end', () => {
            console.log('Conversion completed successfully');
        })
        .run()
}