determining bytes read from a file.

[vineeth]

Hello all,
I have come across a weird problem, I need to determine the amount
of bytes read from a file, but couldn't figure it out ,
My program does this :
__
file = open("somefile")
data = file.read()
print "bytes read ", len(data)
---

But the bytes read is not being printed correctly, I think bytes are
being counted only till the first occurance of '\0' is encountered.
Even though the file is of a very large size, the bytes till the first
'\0' are counted.

Can someone pls advise me regarding this.
Thanks.

Best Regards,
Vineeth.

 

[Diez B. Roggisch]

Hello all,
I have come across a weird problem, I need to determine the amount
of bytes read from a file, but couldn't figure it out ,
My program does this :
__
file = open("somefile")
data = file.read()
print "bytes read ", len(data)
---

But the bytes read is not being printed correctly, I think bytes are
being counted only till the first occurance of '\0' is encountered.
Even though the file is of a very large size, the bytes till the first
'\0' are counted.

I doubt that. Python doesn't interpret data when reading, and byte-strings
don't have a implicit 0-based length.

So I think you must be doing something different - clearly the above is not
actual code, but something made up for this post. Show us your actual code,
please.

diez

 

[Marc 'BlackJack' Rintsch]

I have come across a weird problem, I need to determine the amount
of bytes read from a file, but couldn't figure it out ,
My program does this :
__
file = open("somefile")
data = file.read()
print "bytes read ", len(data)
---

But the bytes read is not being printed correctly, I think bytes are
being counted only till the first occurance of '\0' is encountered.
Even though the file is of a very large size, the bytes till the first
'\0' are counted.

If you want to deal with bytes better open the file in binary mode.
Windows alters line endings and stops at a specific byte (forgot the
value) otherwise.

Ciao,
Marc 'BlackJack' Rintsch

 

[John Machin]

Hello all,
I have come across a weird problem, I need to determine the amount
of bytes read from a file, but couldn't figure it out ,
My program does this :
__
file = open("somefile")
data = file.read()
print "bytes read ", len(data)
---

But the bytes read is not being printed correctly, I think bytes are
being counted only till the first occurance of '\0' is encountered.
Even though the file is of a very large size, the bytes till the first
'\0' are counted.

Can someone pls advise me regarding this.
Thanks.

Best Regards,
Vineeth.

 

[vineeth]

I doubt that. Python doesn't interpret data when reading, and byte-strings
don't have a implicit 0-based length.

So I think you must be doing something different - clearly the above is not
actual code, but something made up for this post. Show us your actual code,
please.

diez

Hi,
The program tries to create a C Byte array of HEX data from a binary
input file (for ex : to embed a .png image with the application as an
array),
Here is the program :

"""
python script to create a bit stream of a input binary file.
Usage : bit_stream_creator.py -i input_file -b bytes_to_dump
"""

import sys
from binascii import hexlify
from optparse import OptionParser

if len(sys.argv) != 5:
print "incorrect args, usage : %s -i input_file -b bytes_to_dump" %
(sys.argv[0])
sys.exit(0)

parser = OptionParser()
parser.add_option("-i", "--input", dest="inputfilename")
parser.add_option("-b", "--bytes", dest="bytes")

(options, args) = parser.parse_args()

print "-i",options.inputfilename
print "-b",options.bytes

# open input file
infile = open(options.inputfilename)

# create the member variable name.
mem_var_name = options.inputfilename
mem_var_name = mem_var_name.replace(' ','_')
mem_var_name = mem_var_name.replace('.','_')

outfile_c = open(mem_var_name + ".c","w")
outfile_h = open(mem_var_name + ".h","w")

# read the data.
print " Reading %d bytes..... " % (int(options.bytes))
bytes_reqd = int(options.bytes)
data = infile.read(bytes_reqd)
print "Bytes Read ", len(data)

# convert to hex decimal representation
hex_data = hexlify(data)
i = 0

# Write the c file with the memory dump.
outfile_c.write ( "unsigned char %s[%d] = {\n" %
(mem_var_name,bytes_reqd) )
while i < len(hex_data):
outfile_c.write( "0x%c%c" % ( hex_data,hex_data[i+1] ) )
i += 2
if i != len(hex_data):
outfile_c.write(",")
if i % 32 == 0:
outfile_c.write("\n")
outfile_c.write ( "\n};\n" )

# Write the .h file with forward declaration.
cpp_macro = "__"+mem_var_name.upper()+"_H__"
outfile_h.write("#ifndef "+cpp_macro + "\n")
outfile_h.write("#define "+cpp_macro + "\n")
outfile_h.write( "//%s, size %d \n" % (mem_var_name,len(data)) )
outfile_h.write( "extern unsigned char %s[%d];\n" %
(mem_var_name,bytes_reqd) )
outfile_h.write("#endif //"+cpp_macro + "\n")

#close the files.
outfile_c.close()
outfile_h.close()
infile.close()


________

But len(data) never proceeds beyond the NULL character.

Any help and tips is appreciated.

Thanks and Regards,
Vineeth.

 

[John Machin]

Hello all,
I have come across a weird problem, I need to determine the amount
of bytes read from a file, but couldn't figure it out ,
My program does this :
__
file = open("somefile")
data = file.read()
print "bytes read ", len(data)
---

But the bytes read is not being printed correctly, I think bytes are
being counted only till the first occurance of '\0' is encountered.
Even though the file is of a very large size, the bytes till the first
'\0' are counted.

Python will not stop on reading '\0'. On Windows in text mode (the
default), '\r\n' will be converted to '\n', and it will stop on
reading ^Z aka chr(26) aka '\x1A'. If you don't want that to happen,
use open('somefile', 'rb') to get binary mode.

 

[vineeth]

If you want to deal with bytes better open the file in binary mode.
Windows alters line endings and stops at a specific byte (forgot the
value) otherwise.

Ciao,
Marc 'BlackJack' Rintsch

Thanks, opening the file in binary mode helped, thanks a lot.

 

[Matt Nordhoff]

parser.add_option("-b", "--bytes", dest="bytes")

This is an aside, but if you pass 'type="int"' to add_option, optparse
will automatically convert it to an int, and (I think), give a more
useful error message on failure.
--

 

[Chris]

A couple potential optimizations:

# create the member variable name.
mem_var_name = options.inputfilename
mem_var_name = mem_var_name.replace(' ','_')
mem_var_name = mem_var_name.replace('.','_')

mem_var_name = options.inputfilename.replace(' ','_').replace('.','_')
No need to assign it 3 times.

while i < len(hex_data):
outfile_c.write( "0x%c%c" % ( hex_data,hex_data[i+1] ) )
i += 2
if i != len(hex_data):
outfile_c.write(",")
if i % 32 == 0:
outfile_c.write("\n")

This could be re-written as such:

for x in xrange(0, len(hex_data), 32):
output_c.write('%s\n' % ','.join(['0x%s'%''.join(['%c'%a for a in
hex_data[x:x+32][i:i+2]]) for i in xrange(0, 32, 2)]

 

[vineeth]

A couple potential optimizations:

# create the member variable name.
mem_var_name = options.inputfilename
mem_var_name = mem_var_name.replace(' ','_')
mem_var_name = mem_var_name.replace('.','_')

mem_var_name = options.inputfilename.replace(' ','_').replace('.','_')
No need to assign it 3 times.

while i < len(hex_data):
outfile_c.write( "0x%c%c" % ( hex_data,hex_data[i+1] ) )
i += 2
if i != len(hex_data):
outfile_c.write(",")
if i % 32 == 0:
outfile_c.write("\n")

This could be re-written as such:

for x in xrange(0, len(hex_data), 32):
output_c.write('%s\n' % ','.join(['0x%s'%''.join(['%c'%a for a in
hex_data[x:x+32][i:i+2]]) for i in xrange(0, 32, 2)]

Thanks everyone for all the suggestions, and the optimizations, I am
still a newbie and not aware of these ,
thanks!