Recherche avancée
Autres articles (73)
-
Participer à sa documentation
10 avril 2011La documentation est un des travaux les plus importants et les plus contraignants lors de la réalisation d’un outil technique.
Tout apport extérieur à ce sujet est primordial : la critique de l’existant ; la participation à la rédaction d’articles orientés : utilisateur (administrateur de MediaSPIP ou simplement producteur de contenu) ; développeur ; la création de screencasts d’explication ; la traduction de la documentation dans une nouvelle langue ;
Pour ce faire, vous pouvez vous inscrire sur (...) -
Supporting all media types
13 avril 2011, parUnlike most software and media-sharing platforms, MediaSPIP aims to manage as many different media types as possible. The following are just a few examples from an ever-expanding list of supported formats : images : png, gif, jpg, bmp and more audio : MP3, Ogg, Wav and more video : AVI, MP4, OGV, mpg, mov, wmv and more text, code and other data : OpenOffice, Microsoft Office (Word, PowerPoint, Excel), web (html, CSS), LaTeX, Google Earth and (...)
-
Keeping control of your media in your hands
13 avril 2011, parThe vocabulary used on this site and around MediaSPIP in general, aims to avoid reference to Web 2.0 and the companies that profit from media-sharing.
While using MediaSPIP, you are invited to avoid using words like "Brand", "Cloud" and "Market".
MediaSPIP is designed to facilitate the sharing of creative media online, while allowing authors to retain complete control of their work.
MediaSPIP aims to be accessible to as many people as possible and development is based on expanding the (...)
Sur d’autres sites (11178)
-
Installing FFMPEG on my EC2 instance takes too long ; what am I doing wrong ? [closed]
24 août 2023, par Shaban KhawarI found a good article on how to download onto my EC2 instance : link
Here's the issue. It takes forever ! It takes around 15 minutes to install.


I'm running my EC2 instance on Amazon Linux 2023, so no apt-get for me.
Also my EC2 instance is created by my EB environment, so if auto-scaling occurs, my instance will be terminated and a new one will be created. I can set .ebextensions to run all the commands to install FFMPEG again but as mentioned above, it takes forever ! That means for 15-20 minutes, my server is down because of one dependancy. I feel like I'm going about this the wrong way, so any advice is appreciated.


-
xdotool to tab to a button on a web page and use the mouse to disable a drop down menu option
25 juin 2023, par MashI have a Bash script that open a Amazon chime meeting URL in firefox, uses XDOtool to enter a meetig participant name, and tab and mouse click functions. and next uses ffmpeg to stream the video and audio output of the Amazon chime meeting to an RTMP destination.


When this is streamed, the Amazon chime web app has "More" drop down Menu. Within the Menu it has a option to disable the self view. I want to add xdotool commands to disable this self view option from the more drop down menu on the amazon chime web app page.


the Amazon chime meeting URL is - https://app.chime.aws/meetings/


Here is the Bash Script


#!/bin/bash
BROWSER_URL=${MEETING_URL}
SCREEN_WIDTH=1920
SCREEN_HEIGHT=1080
SCREEN_RESOLUTION=${SCREEN_WIDTH}x${SCREEN_HEIGHT}
CAPTURE_SCREEN_RESOLUTION=1920x1080
COLOR_DEPTH=24
X_SERVER_NUM=2
VIDEO_BITRATE=6000
VIDEO_FRAMERATE=30
VIDEO_GOP=$((VIDEO_FRAMERATE * 2))
AUDIO_BITRATE=160k
AUDIO_SAMPLERATE=44100
AUDIO_CHANNELS=2

# Start PulseAudio server so Firefox will have somewhere to which to send audio
pulseaudio -D --exit-idle-time=-1
pacmd load-module module-virtual-sink sink_name=v1 # Load a virtual sink as `v1`
pacmd set-default-sink v1 # Set the `v1` as the default sink device
pacmd set-default-source v1.monitor # Set the monitor of the v1 sink to be the default source

# Start X11 virtual framebuffer so Firefox will have somewhere to draw
Xvfb :${X_SERVER_NUM} -ac -screen 0 ${SCREEN_RESOLUTION}x${COLOR_DEPTH} > /dev/null 2>&1 &
export DISPLAY=:${X_SERVER_NUM}.0
sleep 0.5 # Ensure this has started before moving on

# Create a new Firefox profile for capturing preferences for this
firefox --no-remote --new-instance --createprofile "foo4 /tmp/foo4"

# Install the OpenH264 plugin for Firefox
mkdir -p /tmp/foo4/gmp-gmpopenh264/1.8.1.1/
pushd /tmp/foo4/gmp-gmpopenh264/1.8.1.1 >& /dev/null
curl -s -O http://ciscobinary.openh264.org/openh264-linux64-2e1774ab6dc6c43debb0b5b628bdf122a391d521.zip
unzip openh264-linux64-2e1774ab6dc6c43debb0b5b628bdf122a391d521.zip
rm -f openh264-linux64-2e1774ab6dc6c43debb0b5b628bdf122a391d521.zip
popd >& /dev/null

# Set the Firefox preferences to enable automatic media playing with no user
# interaction and the use of the OpenH264 plugin.
cat <<eof>> /tmp/foo4/prefs.js
user_pref("media.autoplay.default", 0);
user_pref("media.autoplay.enabled.user-gestures-needed", false);
user_pref("media.navigator.permission.disabled", true);
user_pref("media.gmp-gmpopenh264.abi", "x86_64-gcc3");
user_pref("media.gmp-gmpopenh264.lastUpdate", 1571534329);
user_pref("media.gmp-gmpopenh264.version", "1.8.1.1");
user_pref("doh-rollout.doorhanger-shown", true);
EOF

# Start Firefox browser and point it at the URL we want to capture
#
# NB: The `--width` and `--height` arguments have to be very early in the
# argument list or else only a white screen will result in the capture for some
# reason.
firefox \
 -P foo4 \
 --width ${SCREEN_WIDTH} \
 --height ${SCREEN_HEIGHT} \
 --new-instance \
 --first-startup \
 --foreground \
 --kiosk \
 --ssb \
 "${BROWSER_URL}" \
 &
sleep 10 # Ensure this has started before moving on, waiting for loading the Chime web app
xdotool key Return #Select yes for the pop-up window of "Would you like to open this link with Chime app?"
sleep 3
xdotool key Escape #Close the pop-up window
sleep 3
xdotool type Livestream #Type "Livestream" on the name input field
sleep 3
xdotool key Tab #Move to "join the meeting" button
sleep 3
xdotool key Return #Click "join the meeting" button
sleep 3
xdotool key Return #Close the pop-up window once again
sleep 3
xdotool key Escape #Close the pop-up window once again
sleep 3
xdotool key Return #Click "Use system audio" setting
sleep 3
xdotool key Escape #Close warning message
sleep 3
xdotool mousemove 1 1 click 1 # Move mouse out of the way so it doesn't trigger the "pause" overlay on the video tile 

# Start ffmpeg to transcode the capture from the X11 framebuffer and the
# PulseAudio virtual sound device we created earlier and send that to the RTMP
# endpoint in H.264/AAC format using a FLV container format.
#
# NB: These arguments have a very specific order. Seemingly inocuous changes in
# argument order can have pretty drastic effects, so be careful when
# adding/removing/reordering arguments here.
ffmpeg \
 -hide_banner -loglevel error \
 -nostdin \
 -s ${CAPTURE_SCREEN_RESOLUTION} \
 -r ${VIDEO_FRAMERATE} \
 -draw_mouse 0 \
 -f x11grab \
 -i ${DISPLAY} \
 -f pulse \
 -ac 2 \
 -i default \
 -vf "crop=1600:980:0:1080" \
 -c:v libx264 \
 -pix_fmt yuv420p \
 -profile:v main \
 -preset slow \
 -x264opts "nal-hrd=cbr:no-scenecut" \
 -minrate ${VIDEO_BITRATE} \
 -maxrate ${VIDEO_BITRATE} \
 -g ${VIDEO_GOP} \
 -filter_complex "aresample=async=1000:min_hard_comp=0.100000:first_pts=1" \
 -async 1 \
 -c:a aac \
 -b:a ${AUDIO_BITRATE} \
 -ac ${AUDIO_CHANNELS} \
 -ar ${AUDIO_SAMPLERATE} \
 -f flv ${RTMP_URL}``

</eof>

what i have tried so far in in the bash script


-
Extracting Thumbnail URL from Video URL
29 août 2023, par Deepak SangleSo, I have a video URL in an Amazon S3 bucket. I must extract the thumbnail image (preview image) from the video URL. The catch is that I do not have the luxury to download or upload the complete video since it takes time. My approach is something like this.
I first created a blob object containing the first few seconds of that video using something like this.


const chunkSize = 1024*1024; // 1MB
 const start = 0;
 const end = chunkSize - 1;
 const rangeHeader = `bytes=${start}-${end}`;
 
 const VideoResponse = await axios.get(url, { 
 responseType: 'arraybuffer',
 headers: { Range: rangeHeader }
 });

 const chunkData = VideoResponse.data; 
 
 const videoBlob = new Blob([chunkData], { type: 'video/mp4' });
 const videoUrl = URL.createObjectURL(videoBlob);
 console.log({chunkData, videoBlob, videoUrl});


Console gives me something like this


{
 chunkData: <buffer 00="00" 1c="1c" 66="66" 74="74" 79="79" 70="70" 6d="6d" 34="34" 32="32" 69="69" 73="73" 6f="6f" 61="61" 76="76" 63="63" 31="31" 02="02" bd="bd" ab="ab" 6c="6c" 68="68" 64="64" c7="c7" 1048526="1048526" more="more" bytes="bytes">,
 videoBlob: Blob { size: 1048576, type: 'video/mp4' },
 videoUrl: 'blob:nodedata:764fce87-792f-47e8-bc3e-15921ee5787f'
}
</buffer>

Now, there are many options after this. I am trying to download this blob object by converting it into a file. After I successfully convert it into an mp4 file, I can easily take a screenshot of it using
fluent-ffmpegpackage. However, I couldn't convert it into a file. If I try to do something like this

const BlobResponse = await axios({
 method: 'GET',
 url: videoUrl,
 responseType: 'stream'
 });

 const file = fs.createWriteStream(`video.mp4`);
 BlobResponse.data.pipe(file);
 
 file.on("error", (error) => {
 console.log(`There was an error writing the file. Details: $ {error}`);
 });

 file.on('finish', () => {
 file.close();
 console.log("File downloaded successfully");
 });


But it throws me the error
AxiosError: Unsupported protocol blob:

I also tried to directly convert the blob object to a jpeg object but to no avail.


I have searched complete StackOverflow, and I have not found a single solution working so please let me know why those are not working or if there is any new solution to do this.


Syndiquer tout le site
