I have once again run into problems with the hashes not properly importing, though I expect to find a solution for this later today. This way we might actually accomplish this in a somewhat short amount of time. This will ease the burden of a single computer duing this alone. I have a server with a database, and a bunch of clients/workers who actually access furaffinity, and sends the hashes and tags to the database. The status right now: I got a networked way of doing this working. >255 >204 I haven't really been working on this in the last couple of days, because of school, I got a little done on it yesterday though. Same thing for species, though I may want to write a parser to break out "feline - tiger" to "species:feline" and "species:tiger", et cetera. "Multiple characters" is a shit convention, but getting out "gender:male" or whatever for all the solo pics would be great. OP, does this grab gender at all? If not it should. I'm not at all sure how I'm going to do this, but it'll be a fun challenge - I'm just waiting till I have the time and energy to really pound on it for a couple days straight. What I'm thinking about doing is figuring a way to take the output of this and remove all tags that aren't artists AND have a hash count of under whatever threshold gives the best signal to noise ratio (rough guess: <500 is probably trash, but no way to tell til we have the whole database ripped). That's still huge though - regexing artists out of filenames is a great hack but it's not 100% reliable and it only works if you saved from FA in the first place. My thought is that this will be 95% only useful for artist, rating, species, and maybe a few hundred very common tags.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |