I use API for downloading images. I also always try to download least sized of current image available so i make requests to the file_url and large_file_url adresses, read their content-length header and then actually download image.
Since it not very obvious as to in what parameter sample image stored (I encountered situations in which both file_url and large_file_url used for storing path to sample image) - i had to check it myself, just to find out that when requesting url with image it not returns content-length header at all. Which leads to accidential original image download.
Can this be fixed? content-length absense I mean
Also as offtopic - why store *sample* images in LARGE_file_url field?