I believe there is a probem with how file contents are hashed to determine whether content is restricted for API downloads.
Here is an exmaple file which is 8 octets long and contains the string "target/" followed by a newline charecter, i.e.:
$ od -c TestRestrictedContent
0000000 t a r g e t / \n
Clearly, this file cannot be restricted, but I cannot download via the API. The response I get is:
< HTTP/1.1 409 path/restricted_content/
< X-Dropbox-Request-Id: 10d9c8fff32950e657a05901aa77bcd8
{
"error_summary": "path/restricted_content/",
"error": {
".tag": "path",
"path": {
".tag": "restricted_content"
}
}
}