Is there a way to rip graphics from a screensaver made with Axialis Screensaver Producer? Resource Hacker hasn't been helpful.
if resource hacker didn't do it for you then likely not. once they're compiled into exe/scr it's pretty hard to extract anything useful. might be more worth your time to look for the original source of the screensaver
>>1564208What I'm particularly interested in LEGO screensavers from 2011 or newer. What are the odds I'd even find the "source" of those, really?
>>1564208Maybe decompress the EXE/SCR files first?
Rename it to a .zip and open it up.
>>1564219>>1564232No luck in that regard. I took a peek in hexed.it and the first part had the signs of being an EXE. That said, I did stumble upon this curiosity. Think it means anything?
>>1564311I'm not that guy, but this looks very promising.JPEG files start with the bytes FF D8 FF. This sequence is literally there in the middle of line 2F80. This is where the file starts.There's also a marker sequence for the end of a jpeg file: FF D9.So if you start from FF D8 FF and take everything up to (including) the next occurrence of FF D9, that's your JPEG image.Copy that into a new file with your hex editor and try it.
>>1564316With a bit of help from AI, I followed that line of thinking. It cooked up a Python script that managed to extract every image there!Now if only it could "see" the filenames associated with each image and tag them accordingly, though that might take a bit more work with a bit more understanding of the screensaver's overall format.
>>1564318Oh yeah, the script did somehow extract what seemed to be two instances of "junk". I guess it was something that kinda looked like it fit the criteria of those byte sequences but still weren't actually images
>>1564318Cool, AI is very useful for tasks like this.If it always includes the original path, you could try searching for that alongside the jpeg file signature (like capturing (b"C:\" ... b".jpg") ... 0xFF 0xD8 0xFF) and try to parse that. The string is May be more trouble than it's worth unless you want to extract lots of these files.>>1564319That could happen. It's worth checking manually to confirm the context is not like that of actual images.
>>1564311>>1564316>>1564318fucking cool
>>1564321Honestly, I think it would be better figuring out how to parse the overall collection "properly", starting with figuring out where the beginning of that chunk is. I'd wager it also contains either a reference to how many images are embedded (in this case 12) or the overall size (don't know how to even estimate that one)
>>1564410Yes, there's a high likelihood it at least includes the size of the image chunk somewhere and there's a way to find the file names.
>>1564202I managed to extract the images with filename from the slideshow example from their website: https://pastebin.com/jPGCHkzt. Maybe you can adapt it to your screensaver. Or upload it so I can try.Before each image are its full path and its filename. There are some unidentified bytes in between. The image is preceded by its size as a little-endian uint32. Same but uint8 for path and name.
