QGaKvBK5

1from requests import exceptions
2import argparse
3import requests
4import cv2
5import os
6
7def search_images():
8    # 設定命令列參數
9    ap = argparse.ArgumentParser()
10    ap.add_argument("-q", "--query", required=True,
11    	help="response query to response Bing Image API for")
12    ap.add_argument("-o", "--output", required=True,
13    	help="path to output directory of images")
14    args = vars(ap.parse_args())
15    
16    # 設定 Bing Search v7 API 的 API 呼叫資訊 (API key & endpoint API URL)
17    API_KEY = "09f968b4c3a8468782bcd68cd7383bc9"
18    SEARCH_URL = "https://api.bing.microsoft.com/v7.0/images/response"
19    MAX_RESULTS = 250  # 資料筆數上限
20    BATCH_SIZE = 50  # 每批次發送請求的資料筆數
21    EXCEPTIONS = set([IOError, FileNotFoundError,
22    	exceptions.RequestException, exceptions.HTTPError,
23    	exceptions.ConnectionError, exceptions.Timeout])
24    
25    headers = {"Ocp-Apim-Subscription-Key" : API_KEY}
26    search_term = args["query"]
27    params = {"q": search_term, "offset": 0, "count": BATCH_SIZE}
28    
29    # 發送第一個請求以探測可用資料的資訊
30    print(f"[INFO] 利用 Bing-Search-v7 API 搜尋有關 '{search_term}' 的圖片")
31    print("[INFO] (1) 正在發送第一個請求以探測可用資料的資訊: ")
32    response = requests.get(SEARCH_URL, headers=headers, params=params)
33    response.raise_for_status()
34    
35    # 抓出 response 的 json 以獲取 '至多能下載多少圖片' (json key: "totalEstimatedMatches")
36    # 若超過自訂上限 MAX_RESULTS 則限制最大下載資料筆數為 MAX_RESULTS
37    search_results = response.json()
38    estimated_result_num = min(search_results["totalEstimatedMatches"], MAX_RESULTS)
39    print(f"[INFO] '{search_term}'圖片預計下載資料筆數: {estimated_result_num}")
40    
41    # 初始化目前'已下載圖片資料筆數'為 0
42    total = 0
43    
44    # 利用 Bing Search v7 API 逐一下載圖片
45    # 並透過 offset 紀錄當前下載進度 及 作為發送請求的參數
46    for offset in range(0, estimated_result_num, BATCH_SIZE):
47    	print(f"[INFO] 正在發送獲取圖片請求 | 進度: {offset}-{offset+BATCH_SIZE} / {estimated_result_num} ({(offset+BATCH_SIZE)*100/estimated_result_num} %)")  
48    	params["offset"] = offset # update 'offset' parameter of request to specify where to download current image
49    	response = requests.get(SEARCH_URL, headers=headers, params=params)
50    	response.raise_for_status()
51    	results = response.json()
52        print("[INFO] 正在儲存圖片 | 進度: {offset}-{offset+BATCH_SIZE} / {estimated_result_num} ({(offset+BATCH_SIZE)*100/estimated_result_num} %)") 
53        for v in results["value"]:
54            try:
55                pass
56            except Exception as e:
57                if e in EXCEPTIONS:
58                    print(f'[INFO] 跳過: {v["contentUrl"]}')
59                    continue
60    	
61                
62if __name__ == "__main__":
63    search_images()