TalkAPI Information

Recommend Site Improvements

Join LibraryThing to post.

API Information

This topic is currently marked as "dormant"—the last message is more than 90 days old. You can revive it by posting a reply.

1WorldMaker First Message
Dec 21, 2006, 3:32pm

First of all, I was looking for precisely this sort of website and stumbled upon it thanks to Neil Gaiman's blog. I needed an excuse to recatalog my collection (which turned up some interesting information such as loans I had almost forgotten about), and wanted something that I could integrate back into my blog.

Expect a LibraryThing blog post from me shortly, and the CSV export to be probably be put into good use in enriching some of my blog posts.

That brings me to my wishlist: I'd love for a method to quickly poll data changes in an automated script. Even if it's just the CSV data, it would be nice to have some way to pull that data without having to be logged in, perhaps using a API Key methodology like so many other websites these days.

Why are there some discrepancies between the fields in the "tab delimited" and CSV formats? Particularly missing in the CSV data is the Book ID. Why does the tab-delimited appear to be a binary file when I attempt to open it in a text editor? It seems that it may indeed be rudimentary Excel and not tab-delimited as the labelling suggests.

You have posted the link syntax for an ISBN, but what if the book doesn't have an ISBN?

Anyway, thanks for such an interesting site.

2timspalding
Dec 22, 2006, 12:47am

>That brings me to my wishlist: I'd love for a method to quickly poll data changes in an automated script. Even if it's just the CSV data, it would be nice to have some way to pull that data without having to be logged in, perhaps using a API Key methodology like so many other websites these days.

I might be able to do it with some sort of "key" system, but generalized pulling of book data from LibraryThing runs into the Amazon TOS. A real API to LibraryThing would be effectively an API to Amazon, and that's forbidden.

>Why are there some discrepancies between the fields in the "tab delimited" and CSV formats? Particularly missing in the CSV data is the Book ID. Why does the tab-delimited appear to be a binary file when I attempt to open it in a text editor? It seems that it may indeed be rudimentary Excel and not tab-delimited as the labelling suggests.

No real reason for the former. I can look at them again and figure out what the difference is.

The format is UTF-16LE, which Excel wants its tab-delimited data in. I humored it because 95% of people will be using it.

>You have posted the link syntax for an ISBN, but what if the book doesn't have an ISBN?

Well, you can go there with a book id (/catalog.php?book=XXX), and I posted a /title/ syntax, which isn't guaranteed to work, but usually does.

3WorldMaker
Dec 22, 2006, 3:24am

Thanks for the response, Tim.

> I might be able to do it with some sort of "key" system, but generalized pulling of book data from LibraryThing runs into the Amazon TOS. A real API to LibraryThing would be effectively an API to Amazon, and that's forbidden.

I understand. It can't be helped right now that Amazon is the current most "open" database of this information. I certainly realize that if I need that sort of data I can get an Amazon key directly. I'm mostly interested in the more social data of the site and the bits and pieces to richly link from my site to here (and back): the LT Book ID, the book title and author, my tags on the book (it would be really cool to provide an address like my http://tags.worldmaker.net/{TAGNAME}/ pages so that people browsing my tags here in my catalog might have the opportunity to push out into my larger tag cloud of blog posts, pictures and whatnot), the ability to auto-publish copies of reviews (with a link to my blog post) I might do on my blog (so, some sort of review "ping" might be cool). The only remaining "cool" thing to have might be the cover image and I understand those are tough to use because they are mostly Amazon.

> The format is UTF-16LE, which Excel wants its tab-delimited data in.

I see. In hindsight it should have been more obvious. Good to know.

> ...you can go there with a book id (/catalog.php?book=XXX), ...

Great. Maybe you should add that to the list on the API page?

4WorldMaker
Dec 22, 2006, 7:05pm

Yet another suggestion: In addition to the Book ID, could you add the Work ID to the exported data? You might also want to add Work IDs to the thingISBN output, too.

5samwilson2007au
Dec 5, 2007, 6:07pm

Even if a 'full' API for everything in LibraryThing is not possible because of the Amazon TOS, surely an API giving up access to things like reviews, common knowledge, etc. would be okay, given that it's just using LibraryThing data and not Amazon's? But maybe I'm missing something.

Anyway, I'd love to have a way of adding (well, CRUDing I guess) books to my LibraryThing library from within my own code. I like LibraryThing! :-)

Thanks.

6conceptDawg
Dec 5, 2007, 6:17pm

We are certainly planning on releasing a full-featured API that is just what you are describing, and I'm pumped about it. We just have to find the time to do it. :)

7samwilson2007au
Dec 5, 2007, 6:37pm

Huzza! This is great news, thankyou. :-) I'll stick with LibraryThing...

8Noisy
Dec 23, 2007, 7:59am

Lucky I looked around first, because I was just about to request an API feature to return the title and author given an LT work number. This would be so helpful for use on the profile page or on WikiThing.

9r.orrison
Dec 23, 2007, 5:39pm

Would the data available depend on the original source? I commonly use Amazon, but then always edit to match my data. I'd be happy to use other sources first if that would result in making the data within LibraryThing more available to other users.