增加图片存储目录

zhihu_login.py

@@ -35,6 +35,7 @@ import traceback
 import pymysql
 import jieba
 import jieba.analyse
+import os
 
 
 
@@ -293,20 +294,24 @@ class ZhihuAccount(object):
             print(50*"*")
 
 
-    def _dispose_content_url(self,content,img_url_list,cur_image_index):
+    def _dispose_content_url(self,content,img_url_list,cur_image_index,cur_word_index):
         try:
+            img_dir = "./img_" + str(cur_word_index)
+            if not os.path.exists(img_dir):
+                os.makedirs(img_dir)
+
             for ori_img_url in img_url_list:
                 if ori_img_url.find(".jpg") >= 0 or ori_img_url.find(".png") >= 0:
                     cur_image_index += 1
 
-                    local_img_url_path = "./image/img_" + str(cur_image_index) + ".png"
+                    local_img_url_path = img_dir + "/img_" + str(cur_image_index) + ".png"
                     print(ori_img_url, local_img_url_path)
 
                     urlretrieve(ori_img_url, local_img_url_path)
                     local_cv2_img = cv2.imread(local_img_url_path)
                     height, weidth, channel = local_cv2_img.shape
                     local_cropped_img = local_cv2_img[0:(height - 100), 0:weidth]
-                    local_cropped_img_url_path = "./image/cropped_image_" + str(cur_image_index) + ".png"
+                    local_cropped_img_url_path = img_dir + "/cropped_image_" + str(cur_image_index) + ".png"
                     cv2.imwrite(local_cropped_img_url_path, local_cropped_img)
                     qiniu_url = upload_file(local_cropped_img_url_path)
                     content = content.replace(ori_img_url, qiniu_url)
@@ -317,8 +322,9 @@ class ZhihuAccount(object):
             return content,cur_image_index
 
     # 知乎搜索词搜索
-    def zhihu_query_by_word(self,query_word,zhihu_spider_fd,zhihu_spider_question_fd,cur_image_index,question_answer_dict):
+    def zhihu_query_by_word(self,query_word,zhihu_spider_fd,zhihu_spider_question_fd,question_answer_dict,cur_word_index):
 
+        cur_image_index = 0
         for begin_index in range(0,200,10):
             query_by_word_url = "https://www.zhihu.com/api/v4/search_v3?t=general&correction=1&lc_idx=62&" \
                                 "show_all_topics=0&search_hash_id=1dbb1e923a17f147356177932d1236e1&" \
@@ -344,10 +350,10 @@ class ZhihuAccount(object):
                             have_saved_this_answer = False
 
                             img_url_list = re.findall('src="(.*?)"', content)
-                            content,cur_image_index = self._dispose_content_url(content=content,img_url_list=img_url_list,cur_image_index=cur_image_index)
+                            content,cur_image_index = self._dispose_content_url(content=content,img_url_list=img_url_list,cur_image_index=cur_image_index,cur_word_index=cur_word_index)
 
                             img_url_list = re.findall('data-original="(.*?)"', content)
-                            content,cur_image_index = self._dispose_content_url(content=content,img_url_list=img_url_list,cur_image_index=cur_image_index)
+                            content,cur_image_index = self._dispose_content_url(content=content,img_url_list=img_url_list,cur_image_index=cur_image_index,cur_word_index=cur_word_index)
 
 
                             if data_type == "article":
@@ -363,7 +369,7 @@ class ZhihuAccount(object):
                                 if question_id not in question_answer_dict:
                                     question_answer_dict[question_id] = set()
                                     question_answer_dict[question_id].add(platform_id)
-                                    self.zhihu_answers_list_by_question(question_id,question_answer_dict,zhihu_spider_fd,cur_image_index)
+                                    self.zhihu_answers_list_by_question(question_id,question_answer_dict,zhihu_spider_fd,cur_image_index,cur_word_index)
 
                                     question_item_dict = {
                                         "user_id": user_id,
@@ -403,7 +409,7 @@ class ZhihuAccount(object):
 
 
     # 知乎问题对应的回答列表
-    def zhihu_answers_list_by_question(self,question_id,question_answer_dict,zhihu_spider_fd,cur_image_index):
+    def zhihu_answers_list_by_question(self,question_id,question_answer_dict,zhihu_spider_fd,cur_image_index,cur_word_index):
 
         for begin_index in range(0,200,10):
             answers_list_by_question_url = "https://www.zhihu.com/api/v4/questions/" + str(question_id) + \
@@ -430,11 +436,11 @@ class ZhihuAccount(object):
 
                         img_url_list = re.findall('src="(.*?)"', data_content)
                         data_content, cur_image_index = self._dispose_content_url(content=data_content, img_url_list=img_url_list,
-                                                                             cur_image_index=cur_image_index)
+                                                                             cur_image_index=cur_image_index,cur_word_index=cur_word_index)
 
                         img_url_list = re.findall('data-original="(.*?)"', data_content)
                         data_content, cur_image_index = self._dispose_content_url(content=data_content, img_url_list=img_url_list,
-                                                                             cur_image_index=cur_image_index)
+                                                                             cur_image_index=cur_image_index,cur_word_index=cur_word_index)
 
                         if data_type == "answer" and "question" in data_item:
                             question_id = str(data_item["question"]["id"])
@@ -483,11 +489,12 @@ if __name__ == '__main__':
 
     # 问题回答映射词典
     question_answer_dict = dict()
-    cur_image_index = 0
 
+    cur_word_index = 0
     for query_word in top_query_list:
-        print("query_word:%s" % query_word)
-        account.zhihu_query_by_word(query_word,zhihu_spider_fd,zhihu_spider_question_fd,cur_image_index,question_answer_dict)
+        cur_word_index += 1
+        print("query_word:%s" % query_word,flush=True)
+        account.zhihu_query_by_word(query_word=query_word,zhihu_spider_fd=zhihu_spider_fd,zhihu_spider_question_fd=zhihu_spider_question_data_file,question_answer_dict=question_answer_dict,cur_word_index=cur_word_index)