zipping indexed sparse lists in python

zip_sparse_lists.py

arra = [(1.0, 0.5), (3.4, 1), (9.1, 5)]
arrb = [(0.3, 1), (2.0, 2), (4.7, 6)]
arrc = [(0.1, 2), (0.3, 3), (5.7, 7)]
arrd = [(1.1, 3), (2.1, 4), (3.3, 8)]

def index(x):
	return x[0]

def data(x):
	return x[1]

def set_data(lis, idx, data):
	lis[idx] = (index(lis[idx]), data)

def zip(arrays):
	result = [arrays[0]]
	for arr in arrays[1:]:
		# find the missing datapoints from the other arrays to extend ours
		# we only need to check the first one, as it should be always extended
		# when we add a new array
		newarr = zip_arrs(arr, result[0])
		# now add all our datapoints to all the other arrays
		for i in range(0, len(result)):
			result[i] = zip_arrs(result[i], arr)
		# and add our new array
		result.append(newarr)
	return result


def zip_arrs(target, source):
	# build a dict from the target
	t_dict = dict(target)
	# iterate over the source, and add to the dict
	for elem in source:
		if not index(elem) in t_dict:
			t_dict[index(elem)] = "undefined"
	# serialise the t_dict back to a list
	t_list = sorted(t_dict.items(), key=index)
	# iterate over it, and where the entry is "undefined", average the points
	# on the other sides
	# print t_list
	# for i in range(0, len(t_list)):
	# 	print "ITERATION"
	# 	if data(t_list[i]) == "undefined":
	# 		if i == 0:
	# 			print "first element"
	# 			# find the next highest, not-undefined datapoint
	# 			j = i + 1 
	# 			while data(t_list[j]) == "undefined":
	# 				print "data undefined"
	# 				j = j + 1
	# 			print "setting: " + str(t_list[j])
	# 			set_data(t_list, i, data(t_list[j]))
	# 		elif i == len(t_list)-1:
	# 			set_data(t_list, i, data(t_list[i-1]))
	# 		else: 
	# 			# find the next highest, not-undefined datapoint
	# 			j = i + 1 
	# 			while j < len(t_list) and data(t_list[j]) == "undefined":
	# 				j = j + 1
	# 			if j < len(t_list):
	# 				# get current, previous and next indexs
	# 				current_index = index(t_list[i])
	# 				previous_index = index(t_list[i-1])
	# 				next_index = index(t_list[j])
	# 				print "current index: " + str(current_index)
	# 				print "previous index: " + str(previous_index)
	# 				print "next index: " + str(next_index)
	# 				# get previous and next datapoints
	# 				previous_data = data(t_list[i-1])
	# 				next_data = data(t_list[j])
	# 				print "previous data: " + str(previous_data)
	# 				print "next data: " + str(next_data)
	# 				# get "how far between" we are, as a fraction
	# 				fractional_dist = \
	# 					(current_index-previous_index) / (next_index-previous_index)
	# 				# get how far between the previous and next data we are
	# 				our_data = \
	# 					previous_data + ((next_data-previous_data) / fractional_dist)
					
	# 				set_data(t_list,i,our_data)
	# 			else:
	# 				# give up and just use the previous datapoint
	# 				set_data(t_list,i, data(t_list[i-1]))
	return t_list


def main():
	print zip_arrs(arra, arrb)
	zipped = zip([arra, arrb, arrc, arrd])
	for z in zipped:
		print z


if __name__ == '__main__':
	main()