Site request: Steam

[GiovanH]

Would be good to have a extractor for steam shared files like screenshots and artwork, e.g. https://steamcommunity.com/sharedfiles/filedetails/?id=####
I would be interested in doing the development work on this myself, just putting this in as a note.

[mikf]

I guess this is a different kind of link than requested in the other Steam-related issues?

• [Request] Steam screenshots/artwork #1079
• [Request] steam community workshops videos (wallpaper engine) #6546
• Site Request: Steam Community Galleries #6978

[hrxn]

It's basically #6546 I think..

[EuropaYou]

I did some research and found out that /sharedfiles includes screenshots, artworks, videos, workshop mods and guides pages. Media in these pages are stored like this:

• Screenshots, artworks, images, GIFs are stored on images.steamusercontent.com. This url is also being used to store images that are sent through chat.
• Videos are stored on cdn.steamusercontent.com.
• Steam Awards and profile avatars are stored at store.fastly.steamstatic.com.
• shared.fastly.steamstatic.com stores game banners and logos.
• clan.fastly.steamstatic.com stores images in the News section.

[ImVantexHD]

for workshop-shared files from applications like Wallpaper Engine/Desktop Engine you're also required to own the app in order to download the files, but even so the downloaded files are packaged in a proprietary .pkg format so you'd have to use a tool like repkg to extract the images or videos.

[GiovanH]

Unfortunately I'm not seeing any kind of API exposed for this, so my approach has been based on extracting values from text.

With that being the case, there will need to be some sort of logic to determine which extractor to use beyond matching the url (e.g. the screenshot extractor can't handle workshop pages). @mikf , can I get a pointer on how to do this? From what I'm seeing extractor.find only considers the url pattern.

[mikf]

Looking at an example user profile, I'd just use regular URL patterns for

• Screenshots (SteamScreenshotsExtractor or SteamUserScreenshotsExtractor)
  https://steamcommunity.com/id/BrewsterTKoopa/screenshots/
• Workshop Items (SteamWorkshopExtractor or SteamUserWorkshopExtractor)
  https://steamcommunity.com/id/BrewsterTKoopa/myworkshopfiles/
• general Shared Files (SteamFileExtractor)
  https://steamcommunity.com/sharedfiles/filedetails/?id=3380503485

I can't see anything at a first glance that doesn't have a matchable URL, but you can always just make up some "virtual" URLs like steam:... and match those in pattern.

[GiovanH]

A specific example of a non-matchable URL is the one I gave in my original post, https://steamcommunity.com/sharedfiles/filedetails/?id=####. gallery-dl should be able to handle those as part of this feature without the user having to hand-craft virtual inputs.

[mikf]

But these are real, copy-able URLs available on the Steam website.

You "Right Click" -> "Copy URL" and paste it in a terminal. No need to hand-craft anything.

edit: or I just don't understand what the problem is supposed to be

edit2: OK, I guess downloading the actual files from https://steamcommunity.com/sharedfiles/filedetails/?id=3380503485 links is impossible when just using the web interface, but matching them to an extractor shouldn't be a problem. And at least getting screenshots should be doable.