from __future__ import division
import sys
import thread
import time
import numpy as np

def naive_convolve(f, g):
    # f is an image and is indexed by (v, w)
    # g is a filter kernel and is indexed by (s, t),
    #   it needs odd dimensions
    # h is the output image and is indexed by (x, y),
    #   it is not cropped
    if g.shape[0] % 2 != 1 or g.shape[1] % 2 != 1:
        raise ValueError("Only odd dimensions on filter supported")
    # smid and tmid are number of pixels between the center pixel
    # and the edge, ie for a 5x5 filter they will be 2.
    #
    # The output size is calculated by adding smid, tmid to each
    # side of the dimensions of the input image.
    vmax = f.shape[0]
    wmax = f.shape[1]
    smax = g.shape[0]
    tmax = g.shape[1]
    smid = smax // 2
    tmid = tmax // 2
    xmax = vmax + 2*smid
    ymax = wmax + 2*tmid
    # Allocate result image.
    h = np.zeros([xmax, ymax], dtype=f.dtype)
    # Do convolution
    for x in range(xmax):
        for y in range(ymax):
            # Calculate pixel value for h at (x,y). Sum one component
            # for each pixel (s, t) of the filter g.
            s_from = max(smid - x, -smid)
            s_to = min((xmax - x) - smid, smid + 1)
            t_from = max(tmid - y, -tmid)
            t_to = min((ymax - y) - tmid, tmid + 1)
            value = 0
            for s in range(s_from, s_to):
                for t in range(t_from, t_to):
                    v = x - smid + s
                    w = y - tmid + t
                    value += g[smid - s, tmid - t] * f[v, w]
            h[x, y] = value
    return h

N = 200
f = np.arange(N*N, dtype=np.int).reshape((N,N))
g = np.arange(81, dtype=np.int).reshape((9, 9))

def callback(*args):
    pass

if '--with-gil' in sys.argv:
    # initialize the GIL
    thread.start_new_thread(callback, ())

t = time.time()
naive_convolve(f, g)
print time.time() - t