PDF file detected as non-binary

[stefan6419846]

I am feeding a PDF file to typecode.contenttype.is_binary. As PDF files are usually considered as binary files, I would have expected the file to be detected as binary, but apparently the first bytes used for detection are looking like plain-text, leading to a wrong classification.

Example file: antartica-3427135_640_1_libtiff.pdf