Recherche avancée

Sur d’autres sites (10282)

• undefined reference to pl_log_create_341 whenever I try to compile ffmpeg and mpv under Arch Linux [closed]

  3 février 2024, par Mike Nguyen

  Ever since yesterday, I have been struggling to compile ffmpeg, and to a further extent, mpv, on Arch Linux due to the following error that is probably unique to my install :

  


  /usr/bin/ld: libavfilter/libavfilter.so: undefined reference to pl_log_create_341 collect2: error: ld returned 1 exit status make: *** [Makefile:133: ffplay_g] Error 1 make: *** Waiting for unfinished jobs.... /usr/bin/ld: libavfilter/libavfilter.so: undefined reference to pl_log_create_341'
collect2 : error : ld returned 1 exit status
make : *** [Makefile:133 : ffprobe_g] Error 1
/usr/bin/ld : libavfilter/libavfilter.so : undefined reference to pl_log_create_341' collect2: error: ld returned 1 exit status make: *** [Makefile:133: ffmpeg_g] Error 1

  


  I was trying to install mpd on my system, but pacman forced me to remove a ton of essential packages relating to multimedia, Qt, etc. I have since been able to reinstall most of these packages.

  


  However, my ffmpeg has been corrupted (as well as mpv failing to start) due to another error that apparently no one else has been getting (I have searched Google for this) :

  


  symbol lookup error: /usr/lib/libavfilter.so.9: undefined symbol: pl_tone_map_auto

  


  I have both libavfilter.so.9.12.100 and libavfilter.so.9.13.100 in my /usr/lib directory, and I am forced to run sudo ln -s /usr/lib/libavfilter.so.9.12.100 /usr/lib/libavfilter.so.9 every time I run pacman.

  


  I have spent countless hours browsing the web about what solutions are available, but as I said, I seem to be the only one getting these errors.

  


  Whether it's installing ffmpeg-git via the AUR, or removing and reinstalling mpv, nothing seems to fix the problem. And no one else is having this problem.

  


  I might be forced to do a clean install of Arch Linux, but I don't have the time for it.

  


• Leading Google Analytics alternative, Matomo, parodies Christopher Nolan blockbuster ahead of the UA sunset

  4 juillet 2023, par Erin — Press Releases

  Wellington, New Zealand, 4 July 2023 : In the world of online data, Google Analytics has long reigned supreme. Its dominance has been unquestioned, leaving website owners with little choice but to rely on the tech giant for their data insights. However, a new dawn in web analytics is upon us, and Matomo, the leading alternative to Google Analytics, is seizing a unique opportunity to position itself as the go-to provider. In a bold move, Matomo has launched a parody trailer, “Googleheimer,” humorously taking a satirical swipe at Google in the style of the upcoming Oppenheimer biopic by Christopher Nolan.

  Capitalising on a time-bound decision

  With an important decision looming for marketers and web specialists who need to switch analytics providers by July 1st, Matomo has found the perfect window to capture their attention.

  The urgency of the situation, combined with the high intent to switch providers, sets the stage for Matomo to establish itself as the leading alternative analytics platform of choice.

  Matomo’s parody trailer addresses the frustrations of GA4 head-on by highlighting the issues and the uncertainties caused by the sunset of Universal Analytics in humorous satire with lines such as :

  “But we’re keeping everyone’s data, right ? Right ?? …RIGHT ?!”

  Riding on the coat tails of this summer’s anticipated blockbuster from Christopher Nolan, Matomo openly points at the downsides of GA4, and reflects many frustrated marketers pain points in an entertaining way. Beneath the comedic and satirical tone lies the message that users have choices, and no longer need to surrender to the behemoth incumbent.

  Matomo was founded to challenge the status quo and provide a solution for those who believe in privacy and in ethical analytics, and who prefer that their customer data not be concentrated in the hands of just a few corporations.

  Watch the full trailer here. 


  About Matomo
  

  Matomo is a world-leading open-source privacy-friendly ethical web analytics platform, trusted by over 1.4 million websites in 190 countries and translated into over 50 languages. Matomo helps businesses and organisations track and optimise their online presence allowing users to easily collect, analyse, and act on their website and marketing data to gain a deeper understanding of their visitors and drive conversions and revenue. Matomo’s vision is to create, as a community, the leading open digital analytics platform that gives every user complete control of their data.

  Visit matomo.org for more information.




  More on Google Analytics changes



  A new dawn in web analytics is upon us, and Matomo – the leading alternative to Google Analytics – is here for it. After 20 years, Google is blowing up Universal Analytics (or GA3) – and taking your data with it. Inspired by Christopher Nolan’s upcoming biopic about physicist J. Robert Oppenheimer and the making of his atomic bomb (also known as “The Manhattan Project”), this parody trailer openly points to Google and draws the comparison in humorous satire. GA4 comes with a new set of metrics, setups and reports that change how you analyse your data.

• Serving video stream in Node with ffmpeg

  30 mars 2023, par Spedwards

  I have a local-only utility where the backend is Adonis.js and the frontend is Vue.js. I'm trying to get a readable stream and have it play in my frontend. I have got the very basics down, the video plays, but I can't skip to anywhere else in the video, it'll just jump back to where it left off and continue playing.

  


  I've been told that it requires a bi-directional data flow. What I was planning on doing was updating the frontend stream URL to add a query string to the end with the timestamp of where the user (me) skips to. This would go back to the backend and I'd use ffmpeg to create a new stream from the video starting at that timestamp.

  


  The problem is that I've never really messed around with streams before and I'm finding all of this very confusing. I'm able to get a ReadStream of my video and serve it, but I can't write to it. I can create a WriteStream and have it start at my timestamp (I think) but I can't serve it because I can only return ReadStream, ReadWriteStream, or ReadableStream. The ReadWriteStream sounds perfect but I have no idea how to create one and I couldn't find anything fruitful after a few hours of searching, nor could I find anyway of converting a WriteStream to a ReadStream.

  


  There's also the problem I alluded to ; I have no idea if my ffmpeg method is actually working since I can't serve it to test.

  


  My working controller method without any of the timestamp stuff is as follows :

  


  public async stream({ params, response }: HttpContextContract) {
    const file = await File.find(params.id)
    if (!file) {
        return response.badRequest()
    }
    const stream = await Drive.getStream(file.path) // this creates a ReadableStream
    return response.stream(stream)
}


  


  For all the ffmpeg stuff, I'm using fluent-ffmpeg as it was the best wrapper I could find.

  


  This was my first attempt.

  


  public async stream({ params, request, response }: HttpContextContract) {
    const file = await File.find(params.id)
    if (!file) {
        return response.badRequest()
    }
    const stream = await Drive.getStream(file.path) // this creates a ReadableStream
    if (request.input('t')) {
        const timestamp = request.input('t')
        ffmpeg()
            .input(stream)
            .seekInput(timestamp)
            .output(stream)
    }
    return response.stream(stream)
}


  


  How can I achieve what I want ? Am I going about this the wrong way and/or is there a better way ?