photo.py

import sys

args = sys.argv
input_file = open(args[1], 'r')
output_file = open(args[2], 'w')

lines = int(input_file.readline())

horizontal_slides = {}
vertical_slides = {}
verticals = {}
transitions_slides = {}
transitions = []
best_transitions = []
tags = []

for i in range(lines):
    photo_data = input_file.readline().rstrip()
    photo_data_list = photo_data.split(' ')

    # separate horizontals
    if photo_data_list[0] == 'H':
        for j in range(int(photo_data_list[1])):
            tags.append(photo_data_list[j + 2])
        horizontal_slides.update({(i,): set(tags)})

    # separate verticals
    if photo_data_list[0] == 'V':
        for j in range(int(photo_data_list[1])):
            tags.append(photo_data_list[j + 2])
        verticals[i] = tags
    tags = []


# create slides out of verticals
v_len = len(verticals)

if v_len == 2:
    v_ids = tuple(list(verticals.keys()))
    v_tags = set(verticals[1] + verticals[2])
    vertical_slides.update({v_ids: v_tags})
elif v_len > 2:
    verticals1 = list(verticals.items())[:v_len // 2]
    verticals2 = list(verticals.items())[v_len // 2:]
    for i, j in zip(verticals1, verticals2):
        v_ids = (i[0], j[0])
        v_tags = set(i[1] + j[1])
        vertical_slides.update({v_ids: v_tags})

horizontal_transitions = sorted(
    horizontal_slides.items(), key=lambda kv: kv[1])
vertical_transitions = sorted(vertical_slides.items(), key=lambda kv: kv[1])

len_hor = len(horizontal_slides)
len_ver = len(vertical_slides)
if len_hor > 0 and len_ver > 0:
    transitions_slides = {**horizontal_slides, **vertical_slides}
elif len_hor == 0:
    transitions_slides = vertical_slides
else:
    transitions_slides = horizontal_slides
transitions = sorted(transitions_slides.items(), key=lambda kv: kv[1])

# data for output
S = len(transitions)  # no. of slides
output_file.write("{}\n".format(S))


# re-arrange slides based off the interest factor
def interest_factor(tags_1, tags_2):
    tags_intersection = tags_1.intersection(tags_2)
    tags_in_1_not_2 = tags_1.difference(tags_2)
    tags_in_2_not_1 = tags_2.difference(tags_1)
    return len(min([tags_intersection, tags_in_1_not_2, tags_in_2_not_1]))


# start from thr first slide available
slide = transitions[0]
transitions.pop(transitions.index(slide))
best_transitions.append(slide)

# repeatedly select the next interesting slide
while len(transitions) > 0:
    next_slide = max([(interest_factor(slide[1], transitions[i][1]), transitions[i])
                      for i in range(min(100, len(transitions)))])
    best_transitions.append(next_slide[1])
    transitions.pop(transitions.index(next_slide[1]))
    slide = next_slide[1]

for i in best_transitions:
    if len(i[0]) == 2:
        output_file.write("{} {}\n".format(i[0][0], i[0][1]))
    else:
        output_file.write("{}\n".format(i[0][0]))

output_file.close()
input_file.close()