View Single Post
Old 08-27-2012, 01:41 AM   #764
dailyalice
Geek of All Trades
dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!dailyalice rocks like Gibraltar!
 
dailyalice's Avatar
 
Posts: 55
Karma: 100878
Join Date: Aug 2012
Location: Minneapolis
Device: Kindle Oasis (3rd gen)
Quote:
Originally Posted by JimmXinu View Post
There are options in FFDL now for only using site data to fill tags(etc) on new books, not updates. You've already found the only mechanism FFDL offers for changing metadata: replace_metadata. You can use it to both change and remove incoming tags.
Oops. I'd somehow missed the feature where you can specify which columns should only be updated/filled the first time. I must not have poked around in preferences after the last few updates. I like it, it's excellent!

If I want to remove a tag entirely, do I just say EXAMPLETAG=> and leave the space after the bracket blank? Or is there something else to do?

Quote:
Originally Posted by JimmXinu View Post
Are you sure your regexps and escaping are correct?
I suspect I messed up something there, it's been ages since I mucked about with regular expressions. And I know some of the things I want to change involve punctuation that would probably mess things up. (Like changing "Harry P." to "Harry Potter" or "Dark Knight Rises (2012)" to "Batman" and other things of that ilk.)

Quote:
Originally Posted by JimmXinu View Post
AO3 continues to add new features, so it doesn't surprise me there's things FFDL doesn't support. I'll look into what we might be able to do about https URLs if you can provide an example URL.
Honestly it was just that sometimes when I went to copy paste a story URL, it was the usual story URL but with "https" instead of "http". My "fix" was to remove the "s" by hand once I figured that out.

I haven't yet sussed out if it was because I was grabbing the URLs in a different place than usual, but I don't think that's the case as I'm pretty consistent in getting the URL from a list of works in a particular fandom by a particular writer. I pretty much always grab URLs while logged in to AO3 that are both "can only see if logged in" and of the regular variety and never ran into the https issue except a few times so I don't think everyone's getting https by default while logged in or anything like that.

If I run into it again, I'll take notes on when/how it cropped up.

Quote:
Originally Posted by JimmXinu View Post
AO3's pairing is included in 'characters' (which is included in tags by default, see include_subject_tags) because the alternative was to throw it away entirely. AO3 is pretty much the site that has a 'pairing' metadatum and I'm not inclined to go to all the bother of separating it out for one site.
Sure, makes sense. Better to have all the data you can get.

Okay, I'm looking at include_subject_tags. Let me see if I've got this straight: if I don't want character names, genre, status, categories included in "tags" (but just in their respective columns for genre, status, characters, etc.), I can comment that out (or remove some of the items on the default list). Is this correct? I still want to collect that data, but I don't want that stuff listed in the regular tag section.

Thanks!
dailyalice is offline