I've got picture uploads working with s3 on heroku no problem.
I also have a method that enables users to upload from a web-address.
Unfortunately, it seems as though when pictures are uploaded using thi开发者_运维问答s method they are saved without their file extension.
So I get this sort of link for image urls...
http://s3.amazonaws.com/mysite/images/23/original.?1311799466
If the same image had been uploaded locally it would look like this:
http://s3.amazonaws.com/mysite/images/23/original.JPG?1311799466
In my pic
model this is the code I'm using to allow uploads via web addresses:
def download_remote_image
begin
self.image = open(URI.parse(self.pic_url))
rescue
errors.add_to_base("- something is wrong with the image url.")
return false
else
return true
end
end
Any ideas?
Change with this:
def download_remote_image
begin
io = open(URI.parse(pic_url))
def io.original_filename; base_uri.path.split('/').last; end
io.original_filename.blank? ? nil : io
rescue
end
end
Careful, recent versions of Paperclip throw errors when encountering a io object instead of File, not sure if they fixed that.
since you are using paperclip the picture url should be generated with paperclip such as:
self.image(:thumb)
精彩评论