getting at individual bits inside byte field: struct module : bitwiseoperator

[harijay]

In my last post I had asked about reading data from a binary file
using the struct module.
Thanks to some excellent help , I have managed to read in
successfully
most of the header of this binary format that I want to parse. These
are some time-voltage traces from a digital
to analog converter for my experiments. The output is according to a
format
mentioned here : ( http://www.dataq.com/support/techinfo/ff.htm)

I have a question about how to bitmask a bunch of bytes read in from
such a binary formatted file .

For eg the spec says the first two bytes have different parameters in
different bits .
Byte 1 Byte 0
SN16 SD9 SD8 SD7 SD6 SD5 SD4 SD3
SD2 SD1 SD0 T4 T3 T2 T1 T0

I am reading in the two bytes using the following code

import struct
f.seek(0)
element1_format = struct.Struct("<H")
(element1,) = element1_format.unpack(f.read(2))

Now element1 has type "str" . How do I apply a bitmask to this to get
at information in the component bits .
Since the entire file format has many such bitmasked fields and since
this is my first venture into binary formats and c-type structs , I
wanted to know how to read values inside a byte using python.
My few tries at using bitwise operators ( element1 & 0x001f) are
frustrated by messages that say " unsupported operand type(s) for &:
'str' and 'int' " .
How can I keep my string objects as bits and apply bitmasks to them
Any help in this will be greatly appreciated.
Thanks
hari

 

[MRAB]

In my last post I had asked about reading data from a binary file
using the struct module.
Thanks to some excellent help , I have managed to read in
successfully
most of the header of this binary format that I want to parse. These
are some time-voltage traces from a digital
to analog converter for my experiments. The output is according to a
format
mentioned here : ( http://www.dataq.com/support/techinfo/ff.htm)

I have a question about how to bitmask a bunch of bytes read in from
such a binary formatted file .

For eg the spec says the first two bytes have different parameters in
different bits .
Byte 1 Byte 0
SN16 SD9 SD8 SD7 SD6 SD5 SD4 SD3
SD2 SD1 SD0 T4 T3 T2 T1 T0

I am reading in the two bytes using the following code

import struct
f.seek(0)
element1_format = struct.Struct("<H")
(element1,) = element1_format.unpack(f.read(2))

Now element1 has type "str" .

That's not what I get. For me it's an int.

How do I apply a bitmask to this to get
at information in the component bits .
Since the entire file format has many such bitmasked fields and since
this is my first venture into binary formats and c-type structs , I
wanted to know how to read values inside a byte using python.
My few tries at using bitwise operators ( element1 & 0x001f) are
frustrated by messages that say " unsupported operand type(s) for &:
'str' and 'int' " .
How can I keep my string objects as bits and apply bitmasks to them
Any help in this will be greatly appreciated.
>

The T bits are (element1 & 0x1F), the SD bits are ((element1 >> 5) &
0x2FF) and the SN16 bit is (element1 >> 15).

 

[Gabriel Genellina]

mentioned here : ( http://www.dataq.com/support/techinfo/ff.htm)

I have a question about how to bitmask a bunch of bytes read in from
such a binary formatted file .

For eg the spec says the first two bytes have different parameters in
different bits .
Byte 1 Byte 0
SN16 SD9 SD8 SD7 SD6 SD5 SD4 SD3
SD2 SD1 SD0 T4 T3 T2 T1 T0

I am reading in the two bytes using the following code

import struct
f.seek(0)
element1_format = struct.Struct("<H")
(element1,) = element1_format.unpack(f.read(2))

Now element1 has type "str" .

No, should be type 'int', that's what the H format means.

How do I apply a bitmask to this to get
at information in the component bits .
Since the entire file format has many such bitmasked fields and since
this is my first venture into binary formats and c-type structs , I
wanted to know how to read values inside a byte using python.
My few tries at using bitwise operators ( element1 & 0x001f) are
frustrated by messages that say " unsupported operand type(s) for &:
'str' and 'int' " .

Try again - it should work. Perhaps you used the wrong variable name?
instead of the resultant from struct.unpack

py> import struct
py> element1_format = struct.Struct("<H")
py> (element1,) = element1_format.unpack('AB')
py> element1
16961
py> element1 & 0x007F
65
py> chr(65)
'A'

 

[Mark Tolonen]

In my last post I had asked about reading data from a binary file
using the struct module.
Thanks to some excellent help , I have managed to read in
successfully
most of the header of this binary format that I want to parse. These
are some time-voltage traces from a digital
to analog converter for my experiments. The output is according to a
format
mentioned here : ( http://www.dataq.com/support/techinfo/ff.htm)

I have a question about how to bitmask a bunch of bytes read in from
such a binary formatted file .

For eg the spec says the first two bytes have different parameters in
different bits .
Byte 1 Byte 0
SN16 SD9 SD8 SD7 SD6 SD5 SD4 SD3
SD2 SD1 SD0 T4 T3 T2 T1 T0

I am reading in the two bytes using the following code

import struct
f.seek(0)
element1_format = struct.Struct("<H")
(element1,) = element1_format.unpack(f.read(2))

Now element1 has type "str" . How do I apply a bitmask to this to get
at information in the component bits .
Since the entire file format has many such bitmasked fields and since
this is my first venture into binary formats and c-type structs , I
wanted to know how to read values inside a byte using python.
My few tries at using bitwise operators ( element1 & 0x001f) are
frustrated by messages that say " unsupported operand type(s) for &:
'str' and 'int' " .
How can I keep my string objects as bits and apply bitmasks to them
Any help in this will be greatly appreciated.
Thanks
hari

Please post a small example that fails as you describe. From your example,
element1 should be an int (unpacked from two bytes), and there is no f
defined. Without the actual code we can't figure out what you are doing
wrong.

This works as expected:
1

-Mark

 

[harijay]

Thanks Gabriel , MRAB and Mark

Instead of using the bitwise operator on the unpacked data I was
trying to do a bitwise operator directly after read. This was giving
the unsupported operand error

i.e

# version in my code - wrong
file.read(2) & 0x001f

Instead of

# version in my example and now in my working code
(element1,) = struct.unpack("<H",file.read(2) )
Tonethrufour = element1 & 0x1F

I only realized it after I read your replies and went back to my
actual code which had the mistake .
Thanks a lot again.

Hari



Thanks for your help and hope I dont commit any more such stupid
mistakes