mirror of
https://github.com/zhenxun-org/zhenxun_bot.git
synced 2025-12-15 14:22:55 +08:00
62 lines
2.1 KiB
Python
62 lines
2.1 KiB
Python
import os
|
|
import imagehash
|
|
from PIL import Image
|
|
try:
|
|
import ujson as json
|
|
except ModuleNotFoundError:
|
|
import json
|
|
|
|
IMAGE_PATH = r"/home/hibiki/hibikibot/resources/img/"
|
|
TXT_PATH = r"/home/hibiki/hibikibot/resources/txt/"
|
|
|
|
|
|
def get_img_hash(image_file):
|
|
with open(image_file, 'rb') as fp:
|
|
hash_value = imagehash.average_hash(Image.open(fp))
|
|
return hash_value
|
|
|
|
|
|
def check_file_index():
|
|
for dir_name in ['setu/', 'r18/']:
|
|
lens = len(os.listdir(IMAGE_PATH + dir_name))
|
|
for i in range(lens):
|
|
if i > lens:
|
|
return
|
|
if not os.path.exists(f"{IMAGE_PATH}{dir_name}{i}.jpg"):
|
|
os.rename(f"{IMAGE_PATH}{dir_name}{lens}.jpg", f"{IMAGE_PATH}{dir_name}{i}.jpg")
|
|
print(f'{lens}.jpg --> {i}.jpg')
|
|
lens -= 1
|
|
if not i % 100:
|
|
print(f'已检测 {i} 份数据')
|
|
print(f'{dir_name} 检测完毕')
|
|
|
|
|
|
def check_setu_hash():
|
|
check_file_index()
|
|
for dir_name in ['setu/', 'r18/']:
|
|
img_data = {}
|
|
if dir_name == 'setu/':
|
|
fn = 'setu_img_hash.json'
|
|
else:
|
|
fn = 'r18_setu_img_hash.json'
|
|
file_list_len = len(os.listdir(IMAGE_PATH + dir_name)) - 1
|
|
print(file_list_len)
|
|
for i in range(file_list_len):
|
|
file = f"{i}.jpg"
|
|
index = file.split(".")[0]
|
|
img_hash = str(get_img_hash(IMAGE_PATH + dir_name + file))
|
|
print(f'{index}.jpg --> {img_hash}')
|
|
if img_hash in img_data.values():
|
|
k = [k for k, v in img_data.items() if v == img_hash]
|
|
print(f'文件 {index}.jpg 与 {k}.jpg 重复，使用 {file_list_len}.jpg 进行替换')
|
|
os.remove(IMAGE_PATH + dir_name + file)
|
|
os.rename(IMAGE_PATH + f'{dir_name}{file_list_len}.jpg', IMAGE_PATH + dir_name + file)
|
|
file_list_len -= 1
|
|
continue
|
|
img_data[index] = img_hash
|
|
# print(f'{index}.jpg --> {img_data}')
|
|
with open(TXT_PATH + fn, 'w') as f:
|
|
json.dump(img_data, f, indent=4)
|
|
|
|
|
|
check_setu_hash() |