Open sal-uva opened 5 months ago
Interesting. When I built the download_videos.py
, I added a special DatasetVideoLibrary
class that would collect all the previously downloaded videos. It was to avoid redownloading videos. You could use that to find the videos by their URL and then link to them. Something similar would work for images.
If @stijn-uva is ok with merging the Cartographer PR, the @app.route('/result/<path:query_file>')
was modified to serve archived files (assuming they are part of a file in the datasets folder). This would allow you to serve either videos or images from their archives pretty easily.
Linking to social media sites is pretty troublesome for a variety of reasons (expiring URLs, XSS, many requests) causing media files to often be omitted in the Explorer.
We could connect the
Download images
andDownload videos
processors to the Explorer so that media files can be retrieved from the 4CAT server instead of somewhere else, as long as they've been downloaded.