Recherche avancée

Sur d’autres sites (8402)

• How to Correctly Implement ffmpeg Complex Filters in Node.js for Image Processing ?

  24 janvier 2024, par Luke

  Problem :

  


  I am trying to add filters and transitions between my image slideshow array, and am struggling to apply the proper filters. For example, I get errors like this :

  


  {&#xA;  "errorType": "Error",&#xA;  "errorMessage": "ffmpeg exited with code 234: Failed to set value &#x27;fade=type=in:start_time=0:duration=1,zoompan=z=zoom&#x2B;0.002:d=120:x=if(gte(zoom,1.2),x,x&#x2B;1):y=if(gte(zoom,1.2),y,y&#x2B;1)&#x27; for option &#x27;filter_complex&#x27;: Invalid argument\nError parsing global options: Invalid argument\n",&#xA;  "trace": [&#xA;    "Error: ffmpeg exited with code 234: Failed to set value &#x27;fade=type=in:start_time=0:duration=1,zoompan=z=zoom&#x2B;0.002:d=120:x=if(gte(zoom,1.2),x,x&#x2B;1):y=if(gte(zoom,1.2),y,y&#x2B;1)&#x27; for option &#x27;filter_complex&#x27;: Invalid argument",&#xA;    "Error parsing global options: Invalid argument",&#xA;    "",&#xA;    "    at ChildProcess.<anonymous> (/opt/nodejs/node_modules/fluent-ffmpeg/lib/processor.js:182:22)",&#xA;    "    at ChildProcess.emit (node:events:517:28)",&#xA;    "    at ChildProcess._handle.onexit (node:internal/child_process:292:12)"&#xA;  ]&#xA;}&#xA;</anonymous>

  


  Lambda Function Code :

  


   async function concat(bucketName, imageKeys) {
      const imageStreams = await Promise.all(
        imageKeys.map(async (key, i) => {
          const command = new GetObjectCommand({ Bucket: bucketName, Key: key });
          const response = await s3.send(command);
          // Define the temporary file path based on the index
          const tempFilePath = `/tmp/${i}.png`;
    
          // Write the image data to the temporary file
          await fs.writeFile(tempFilePath, response.Body);
    
          // Return the file path to be used later
          return tempFilePath;
        })
      );
    
      // Create a file list content with durations
      let fileContent = "";
      for (let i = 0; i < imageStreams.length; i++) {
        fileContent += `file '${imageStreams[i]}'\nduration 1\n`;
    
        // Check if it's the last image, and if so, add it again
        if (i === imageStreams.length - 1) {
          fileContent += `file '${imageStreams[i]}'\nduration 1\n`;
        }
      }
    
      // Define the file path for the file list
      const fileListPath = "/tmp/file_list.txt";
    
      // Write the file list content to the file
      await fs.writeFile(fileListPath, fileContent);
    
      try {
        await fs.writeFile(fileListPath, fileContent);
      } catch (error) {
        console.error("Error writing file list:", error);
        throw error;
      }
    
      // Create a complex filter to add zooms and pans
      // Simplified filter example
  let complexFilter = [
    // Example of a fade transition
    {
      filter: 'fade',
      options: { type: 'in', start_time: 0, duration: 1 },
      inputs: '0:v', // first video stream
      outputs: 'fade0'
    },
    // Example of dynamic zoompan
    {
      filter: 'zoompan',
      options: {
        z: 'zoom+0.002',
        d: 120, // duration for this image
        x: 'if(gte(zoom,1.2),x,x+1)', // dynamic x position
        y: 'if(gte(zoom,1.2),y,y+1)' // dynamic y position
      },
      inputs: 'fade0',
      outputs: 'zoom0'
    }
    // Continue adding filters for each image
  ];

  let filterString = complexFilter
    .map(
      (f) =>
        `${f.filter}=${Object.entries(f.options)
          .map(([key, value]) => `${key}=${value}`)
          .join(":")}`
    )
    .join(",");
    
      let filterString = complexFilter
        .map(
          (f) =>
            `${f.filter}=${Object.entries(f.options)
              .map(([key, value]) => `${key}=${value}`)
              .join(":")}`
        )
        .join(",");
    
      console.log("Filter String:", filterString);
    
      return new Promise((resolve, reject) => {
        ffmpeg()
          .input(fileListPath)
          .complexFilter(filterString)
          .inputOptions(["-f concat", "-safe 0"])
          .outputOptions("-c copy")
          .outputOptions("-c:v libx264")
          .outputOptions("-pix_fmt yuv420p")
          .outputOptions("-r 30")
          .on("end", () => {
            resolve();
          })
          .on("error", (err) => {
            console.error("Error during video concatenation:", err);
            reject(err);
          })
          .saveToFile("/tmp/output.mp4");
      });
    }


  


  Filter String Console Log :

  


  Filter String: fade=type=in:start_time=0:duration=1,zoompan=z=zoom+0.002:d=120:x=if(gte(zoom,1.2),x,x+1):y=if(gte(zoom,1.2),y,y+1)


  


  Questions :

  


  

  1. What is the correct syntax for implementing complex filters like zoompan and fade in ffmpeg when used in a Node.js environment ?

  


  2. How do I ensure the filters are applied correctly to each image in the sequence ?

  


  3. Is there a better way to dynamically generate these filters based on the number of images or their content ?

  


  


  Any insights or examples of correctly implementing this would be greatly appreciated !

  


• Error opening input files : Invalid data found when processing input

  27 mars 2024, par Master's Time

  I am creating disnake music bot. The error is :

  


  [in#0 @ 00000233f8d71500] Error opening input: Invalid data found when processing input
Error opening input file https://www.youtube.com/watch?v=duDUqBtxwXk.
Error opening input files: Invalid data found when processing input


  


  Here is part of my code :

  


  import disnake
import asyncio
from yt_dlp import YoutubeDL
import ffmpeg


YTDL_OPTIONS = {'format': 'bestaudio', 'noplaylist': 'False', 'simulate':'True', 'key':"FFmpegExtractAudio"}
FFMPEG_OPTIONS = {'before_options': '-reconnect 1 -reconnect_streamed 1 -reconnect_delay_max 5', 'options': '-vn'}


async def play_music(inter):
        global YDTL_OPTIONS, FFMPEG_OPTIONS
    print(tm.now().strftime("%H:%M:%S"),"play_music begin")
    id = int(inter.guild.id)
    with YoutubeDL(YTDL_OPTIONS) as ydl:
        info = ydl.extract_info(url, download=False)

    song = {
            'link': 'https://www.youtube.com/watch?v=' + url,
            'thumbnail': 'https://i.ytimg.com/vi/' + url + '/hqdefault.jpg?sqp=-oaymwEcCOADEI4CSFXyq4qpAw4IARUAAIhCGAFwAcABBg==&rs=AOn4CLD5uL4xKN-IUfez6KIW_j5y70mlig',
            'source': info['formats'][0]['url'],
            'title': info['title']
        }   self.vc[id].play(disnake.FFmpegPCMAudio(executable=r"C:\\ffmpeg\\ffmpeg\\bin\\ffmpeg.exe",source=song["source"], **FFMPEG_OPTIONS))
    print(tm.now().strftime("%H:%M:%S"),"play_music end")


  


  I tried to write source = song['source'] instead of source = song['link'], but it didn't seem helpful.

  


• fftools/ffmpeg_filter : change processing order in fg_finalise_bindings()

  5 avril 2024, par Anton Khirnov

  fftools/ffmpeg_filter : change processing order in fg_finalise_bindings()
  First bind all inputs in all filtergraphs, only then check that all
  
  outputs are bound.
  Needed by the following commit.
  

  • [DH] fftools/ffmpeg.h
  • [DH] fftools/ffmpeg_filter.c
  • [DH] fftools/ffmpeg_opt.c