For some reason, the onebox doesn’t actually fill anything out
http://www.amazon.com/gp/product/B00H4SSB02
Doesn’t appear limited to just .com either:
http://www.amazon.co.uk/gp/product/B00H4SSB02
Edit: It seems the Amazon side is working OK:
For some reason, the onebox doesn’t actually fill anything out
http://www.amazon.com/gp/product/B00H4SSB02
Doesn’t appear limited to just .com either:
http://www.amazon.co.uk/gp/product/B00H4SSB02
Edit: It seems the Amazon side is working OK:
Hmm. Testing full link (for a different product):
Stripping parameters:
Stripping ref:
huh. it seems to be something about the onebox information for that product.
I think I fixed it for most of the products.
https://github.com/discourse/onebox/pull/304

http://www.amazon.com/gp/product/B00H4SSB02 seems to be a special case.
Just a thought: It would be nice if this code also matched the smile.amazon.com domain (the rest of the URL is always the same) which allows non-profits to get a cut of the revenue.
Submit a PR to expand the URL matching:
Is there a reason we try to scrape using mobile pages from Amazon? The desktop pages contains a description meta tag that we can use to improve reliability and avoid breaking Amazon onebox everytime they change their HTML.
The idea was that mobile would have less cruft to sort through, if desktop is easier, go for it.
嗯,描述 meta 标签写得不错,但我不确定“在 Amazon.com 上。符合条件的商品免运费”这部分是否过于分散注意力。想听听您的看法 ![]()
Have we pulled in the updated onebox @eviltrout?
I just bumped it. I can’t say for sure if we were on the latest because our gemfile referenced a version that was pushed to Rubygems that wasn’t also set in our repo. I can confirm that now we are on the very latest onebox code.
Test again:
Looks like this bug isn’t fixed, or we still aren’t deploying the latest version of onebox.
The PR hasn’t been merged yet ![]()
Ok can we please merge that @neil? And merge in onebox version update too.
I think it is working now at last, thank you very much @tgxworld!
This is done now.