Il 09/03/2012 14:33, Warren Weckesser ha scritto:
On Fri, Mar 9, 2012 at 7:26 AM, Nicola Creati <ncre...@inogs.it
<mailto:ncre...@inogs.it>> wrote:
Hello,
I'm writing a library able to read LAS lidar files. I generally
use it under Linux without any problems. I'm now testing my
library on a Windows 7 64 bit computer and I meet some problems
reading the file. I generally use fromfile function to read the
file. I noted that the fromfile returns an arry with some zeros at
the end if I read the whole file or if i read only a finite number
of data i get only a fraction of them. By the way I have created a
sample script that reproduces the problem I noted:
""" Sample code """
import numpy as np
sz = 1000000
data = np.random.random(sz)
output = open('test.bin', 'w')
data.tofile(output)
output.close()
del output
print 'Original data:', data
fid = open('test.bin', 'r')
new_data = np.fromfile(fid)
fid.close()
del fid
print 'Data read:', new_data
fid = open('test.bin', 'r')
slice = np.fromfile(fid, count=100)
fid.close()
print '100 read data:', slice
The "new_data" array has several zeros at the end. The "slice"
array has not 100 items. At every code run the "slice" array size
changes.
I'm using numpy 1.6.1 64bit (taken from Christoph Gohlke website)
, with python 2.7.2 64 bit.
Under Linux Ubuntu 10.10 64bit and numpy 1.6.1 the code works as
expected.
Thanks.
Nicola
Use the binary mode ('wb' and 'rb') when you open the files.
Warren
_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion@scipy.org <mailto:NumPy-Discussion@scipy.org>
http://mail.scipy.org/mailman/listinfo/numpy-discussion
Hello,
thank you, I did not know that on Window you must explicitly use 'rb' or
'wb'.
Cheers.
Nicola
_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion@scipy.org
http://mail.scipy.org/mailman/listinfo/numpy-discussion
_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion@scipy.org
http://mail.scipy.org/mailman/listinfo/numpy-discussion
