0

I have a compressed HDF file (HDF.Z) and would like to open it like that:

from subprocess import Popen, PIPE
f = Popen(['zcat', 'myfile.HDF.Z'], stdout=PIPE).stdout

In order to get the data I need to use pyhdf:

from pyhdf.SD import SD, SDC
mydata = SD(f, SDC.READ)

However, this results in an error message:

*** TypeError: coercing to Unicode: need string or buffer, file found

Is there a way to open this file as a buffer to read it in? Btw: what is a buffer?

oz123
  • 27,559
  • 27
  • 125
  • 187
HyperCube
  • 3,870
  • 9
  • 41
  • 53

1 Answers1

1

At a short glance, I found no way to make it access an open stream.

You can do the following:

  • Create a temporary file where you unpack the file.
  • Give the name of this temporary file to SD().

Another option is very system dependent: you could take the file handle and do

SD('/dev/fd/%d' % f.fileno(), SDC.READ)

bit this is very platform-dependent (Linux only) and, if SD() does mmap() by any chance, it will fail.

glglgl
  • 89,107
  • 13
  • 149
  • 217
  • The temporary unpacking of the file works, although this takes more time (considering the fact that I perform this action on hundreds of files). – HyperCube Feb 05 '14 at 11:30