Goodreads Developers discussion
bugs
>
Clarify results of book/review_counts/json ?
date
newest »


Correct.
All the ones prefixed with "work_" mean stats on all editions (not just the isbn passed in).
id is the unique goodreads identifier for that isbn.
Are their any other fields that don't make sense?
Is there any way these can be excluded from the API results?
Eventually we can add this.

Just for completeness, my solution to the empty review problem (for now) is as follows:
I am just pulling the maximum number of per_page for one API call (100), manually removing the empty reviews, then caching the results and doing my own pagination. That seems to be working pretty well.
Best,
Justin

http://www.goodreads.com/book/isbn?is...
There are actually only 97 records here (according to grep). The number varies. Another call I was using last night returned 93.
Best,
Justin

This can happen for a variety of reasons. The most common reason is that we give users the option of not having their reviews show up in the api, in that case we filter those reviews after we do the database query (for performance reasons). Sorry about the discrepency!
http://www.goodreads.com/book/review_...
returns:
[average_rating] => 3.94
[isbn13] => 9780340839935
[isbn] => 0340839937
[work_ratings_count] => 105144
[reviews_count] => 129965
[ratings_count] => 97526
[work_reviews_count] => 140233
[text_reviews_count] => 2339
[id] => 234225
[work_text_reviews_count] => 3191
From what I can tell using the /book/isbn, the text_reviews_count value is probably the number of "real" reviews.
A related problem I'm having is that /book/isbn returns a bunch of results where the body of the review is empty. It appears that it returns an empty review for every user that has the book on their bookshelf. For books like the one above (Dune), you have to go far into the results to get to them, but for books that have only a few reviews, you hit the empty reviews quickly. Is there any way these can be excluded from the API results?
Best,
Justin