Pythonic way to iterate over sequence, 4 items at a time [duplicate]

Question

Possible Duplicate:
What is the most “pythonic” way to iterate over a list in chunks?

I am reading in some PNG data, which has 4 channels per pixel. I would like to iterate over the data 1 pixel at a time (meaning every 4 elements = 1 pixel, rgba).

red_channel = 0
while red_channel < len(raw_png_data):
    green_channel, blue_channel, alpha_channel = red_channel +1, red_channel +2, red_channel +3
    # do something with my 4 channels of pixel data ... raw_png_data[red_channel] etc
    red_channel += 4

This way doesnt really seem "right". Is there a more Pythonic way to iterate over a sequence, 4 items at a time, and have those 4 items unpacked?

Answer 1

(Python's itertools should really make all recipes as standard functions...)

You could use the grouper function:

from itertools import zip_longest
def grouper(n, iterable, fillvalue=None):
    "grouper(3, 'ABCDEFG', 'x') --> ABC DEF Gxx"
    args = [iter(iterable)] * n
    return izip_longest(fillvalue=fillvalue, *args)

Then you can iterate the pixels by

for r,g,b,a in grouper(4, raw_png_data):
  ....

Alternatively, you could use

irpd = iter(raw_png_data)
for r,g,b,a in zip(irpd, irpd, irpd, irpd):  # use itertools.izip in Python 2.x
  ....

Note that this will chop the last few bytes if the iterable's length is not a multiple of 4. OTOH, the grouper function uses izip_longest, so the extra bytes will be padded with None for that.

Answer 2

vars = [1, 2, 3, 4, 5, 6, 7, 8]
for a, b, c, d in zip(*[iter(vars)]*4):
    print a, b, c, d

Answer 3

from itertools import izip
for r,g,b,a in izip(*[iter(data)]*4):
    ...

Answer 4

for r, g, b, t in (data[i:i+4] for i in xrange(0, len(data)/4*4, 4)):
    print r, g, b, t