No DMs are imported. The functionality for that seems to be broken in their servers. Even then, they are to be manually approved for uploading (when or if they get it to work)
Breaking TOS is a sure way to get your account deleted, I don't know what to tell you here. If OF has any sort of decent logging in place (which they do), your account glows in the dark when you do this, there's no way around it. If you're rapid-firing GET requests for all files shared by the creators you're subbed to you're 100% scraping and they know this. Is a risk the end user/leaker assumes.
It's up to you tbh, do you want the devs to be able to debug any errors you might encounter, basically if your account encounters an unknown error/bug and you didn't check this, they can't debug it easily.
Same for allowing the importer to save session keys, sometimes the importer is so clogged that your job won't run before 24 hours or will get stuck and not finish, this automatically launches a new import job every 24 hours fixing most of those annoyances
Regarding the values, just get , you're looking for:
"username": "uXXXXXXXXX", (remove the u)
"cookie": "...; sess=xxxxxxxxxxxxxxxxxxxx; ...",
"user_agent": "If you're using different browsers even if in the same device for logging in to OF and scraping, paste this from the cookie helper too",
"x_bc": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
new downloaders are always cool but the issue is that most of them are pet projects that do one or two very specific things and get abandoned... Hell, even I have a couple of abandoned downloader repositories of my own because I saw the futility in building such a tool on my own when there are robust, thriving projects with many collaborators providing a 100x better experience/feature set.
Taking a look at it, paints itself as being faster than CDL for the few sites it benchmarks (Coomer mainly), which I don't doubt as even the downloaders I wrote had faster download speeds than CDL, it could be good but being a flutter app is a huge turndown for me, CDL's terminal-first approach is one of its biggest strengths, you can plug and play it into any automation system, bot, NAS you want, even on a headless system, by contrast, this one serves just a subset of users (GUI users/casual downloaders) while CDL tackles the entire spectrum, from casual one-off downloads to archivists/automation freaks and given the huge list of supported sites/forums/hosting services, I'd say it even fits data hoarders.
I don't know what CDL is using in the background for handling their downloads (haven't looked at their code but I assume both selenium and beautiful soup for scraping, probably requests for downloads) that might cause it to get slower down speeds than using CURL directly from the terminal, but the devs are very proficient and active, the thing is packed with options to configure, operation modes, filters, loggers and many, many more tools and functionalities. CDL is imho one of, if not the very best downloader for OF, Coomer, this very forum (and many other), reddit and easily 20+ filehosting/leaking services.
I wonder if CDL has some sort of bandwidth limit functionality for coomer because it is the only site where the download speeds in CDL are truly, TRULY bad and it is a well known fact that their servers are constantly hit with massive amounts of inbound and outbund traffic from importers, user browser requests and scrapers, so a download/request speed throttle being implemented for the service whose servers are always super busy is not as outlandish as it might sound.
I mean, it's not a surprise that the fastest download speeds and fastest importing processes I've seen on coomer have always been when the site had just came back online after being down for a while
Last edited:
