# Ordering similarity
def calc_ordering_sim(ordering1, ordering2, matrix):
return similarity(ordering1, ordering2, matrix)
# Cuisine similarity
def calc_cuisine_sim(cuisine1, cuisine2, matrix):
return similarity(cuisine1, cuisine2, matrix)
if __name__ == '__main__':
# 直接執行r2R.py的時候，輸出餐廳相似矩陣的csv
from db import DBConn
conn = DBConn()
conn.open()
restaurants_num = conn.getRestaurantsNum()
order_sim＿matrix = conn.getOrderingSimMatrix()
cuisine_sim_matrix = conn.getCuisineSimMatrix()
num = len(restaurants_num)
similarities = []
for i, restaurant1 in enumerate(restaurants_num):
for j in range(i, num):
restaurant2 = restaurants_num[j]
sim = calc_r2r(restaurant1, restaurant2, conn.getRestaurantDistance(restaurant1['rid'], restaurant2['rid']),
order_sim＿matrix, cuisine_sim_matrix) if restaurant1['rid'] != restaurant2['rid'] else 1
similarities.append((restaurant1['rid'], restaurant2['rid'], sim))
conn.close()
output_filepath = output_dir + '/wp_posts/wp_links.json'
with codecs.open(output_filepath, 'w+', 'utf-8') as f:
f.write(json.dumps(wp_links, default=helpers.json_serial, indent=4))
output_filepath = output_dir + '/all_image_urls.json'
with codecs.open(output_filepath, 'w+', 'utf-8') as f:
f.write(
json.dumps(all_image_urls, default=helpers.json_serial, indent=4))
## dump s3_images links
"""
d.info('dump s3 image links ...')
s3_image_links = get_all_s3_images_url(cursor)
d.info('total count: {}'.format(len(s3_image_links)))
s3_image_results = {}
for s3_image_link in s3_image_links:
cdn_wp_online_links = phpserialize.loads(s3_image_link['meta_value'])['cdn-wp-online']
post_id = s3_image_link['post_id']
s3_image_results[post_id] = cdn_wp_online_links
output_filepath = output_dir + '/s3_images/s3_images.json'
with codecs.open(output_filepath, 'w+', 'utf-8') as f:
f.write(json.dumps(s3_image_results, default=helpers.json_serial, indent=4))
d.info('[OK] => {}\n'.format(output_filepath))
"""
db_conn.close()
