Is there any way to recognize whether it the file is a pdf file directly?

[pks9862728888]

No description provided.

[h2non]

You can import and instantiate this class, then call the match() method passing the bytes buffer:
https://github.com/h2non/filetype.py/blob/master/filetype/types/archive.py#L173

[pks9862728888]

@h2non How to find the bytes buffer? I only have the request object containing the file. I am using Django Rest Framework.

[h2non]

I guess you would have to read the file contents first, whether from the local file system or remotely (HTTP, SFTP...).

[oz123]

Please try this:

>>> from filetype import match
>>> from filetype.types.archive import Pdf, Deb
>>> pdf = open("/home/oznt/1501006622.pdf", "rb")
>>> match(pdf, matchers=(Pdf(), Deb()))
<filetype.types.archive.Pdf at 0x7f570ca796d8>

Now, notice if you call the function again it will return None

>>> match(pdf, matchers=(Pdf(), Deb()))

If you seek(0) it works again:

>>> pdf.seek(0)
0
>>> match(pdf, matchers=(Pdf(), Deb()))
<filetype.types.archive.Pdf at 0x7f570ca32a90>

You see a new object is created.

@h2non IMHO, this is a bug: filetype should not pass silently on EOF and throw an exception here.