Skip to content

C
crawler
Commit e2fe4476 authored by litaolemo's avatar litaolemo
Browse files
Options

update

parent 5e952134
Show whitespace changes
Inline Side-by-side
Showing with 50 additions and 49 deletions
crawler_sys/scheduler/xiaohongshu_to_rpc.py
View file @ e2fe4476
...@@ -129,54 +129,54 @@ for pid in pid_list: ...@@ -129,54 +129,54 @@ for pid in pid_list:
res_json["NoteView"]["content"]["desc_fix"] = desc_fix res_json["NoteView"]["content"]["desc_fix"] = desc_fix
# print(desc_fix) # print(desc_fix)
res = rds.hset("xiaohongshu_with_img", key=pid, value=json.dumps(res_json)) res = rds.hset("xiaohongshu_with_img", key=pid, value=json.dumps(res_json))
video_dic["platform"] = "9" # video_dic["platform"] = "9"
video_dic["platform_id"] = pid # video_dic["platform_id"] = pid
video_dic["platform_answer_id"] = pid # video_dic["platform_answer_id"] = pid
video_dic["title"] = res_json["NoteView"]["content"]["title"] # video_dic["title"] = res_json["NoteView"]["content"]["title"]
#
user_id_list_copy = copy.deepcopy(user_id_list) # user_id_list_copy = copy.deepcopy(user_id_list)
qustion_id = random.choice(user_id_list_copy) # qustion_id = random.choice(user_id_list_copy)
user_id_list_copy.remove(qustion_id) # user_id_list_copy.remove(qustion_id)
video_dic["user_id"] = qustion_id # video_dic["user_id"] = qustion_id
create_time = datetime.datetime.strptime(res_json["NoteView"]["content"]["time"], # create_time = datetime.datetime.strptime(res_json["NoteView"]["content"]["time"],
'%Y-%m-%d %H:%M') # '%Y-%m-%d %H:%M')
video_dic["create_time"] = create_time.timestamp() # video_dic["create_time"] = create_time.timestamp()
rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/question/batch_create") # rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/question/batch_create")
# print(rpc_res) # # print(rpc_res)
video_dic["platform_question_id"] = pid # video_dic["platform_question_id"] = pid
video_dic["content"] = desc_fix # video_dic["content"] = desc_fix
video_dic["user_id"] = random.choice(user_id_list_copy) # video_dic["user_id"] = random.choice(user_id_list_copy)
rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/answer/batch_create") # rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/answer/batch_create")
comment_list = [] # comment_list = []
try: # try:
if res_json["NoteView"].get("comments"): # if res_json["NoteView"].get("comments"):
# print(res_json["NoteView"].get("data")) # # print(res_json["NoteView"].get("data"))
for comment in res_json["NoteView"]["comments"]["data"]: # for comment in res_json["NoteView"]["comments"]["data"]:
video_dic["content"] = comment['content'] # video_dic["content"] = comment['content']
comment_id_list_copy = copy.deepcopy(majiayonghu_list) # comment_id_list_copy = copy.deepcopy(majiayonghu_list)
comment_id = random.choice(comment_id_list_copy) # comment_id = random.choice(comment_id_list_copy)
video_dic["user_id"] = comment_id # video_dic["user_id"] = comment_id
comment_id_list_copy.remove(comment_id) # comment_id_list_copy.remove(comment_id)
video_dic["create_time"] = (create_time + datetime.timedelta(hours=random.randint(0, 24), # video_dic["create_time"] = (create_time + datetime.timedelta(hours=random.randint(0, 24),
minutes=random.randint(0, 60))).timestamp() # minutes=random.randint(0, 60))).timestamp()
comment_list.append(copy.deepcopy(video_dic)) # comment_list.append(copy.deepcopy(video_dic))
# rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/reply/batch_create") # # rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/reply/batch_create")
elif res_json["NoteView"].get("commentInfo"): # elif res_json["NoteView"].get("commentInfo"):
for comment in res_json["NoteView"]["commentInfo"]["comments"]: # for comment in res_json["NoteView"]["commentInfo"]["comments"]:
video_dic["content"] = comment['content'] # video_dic["content"] = comment['content']
comment_id_list_copy = copy.deepcopy(majiayonghu_list) # comment_id_list_copy = copy.deepcopy(majiayonghu_list)
comment_id = random.choice(comment_id_list_copy) # comment_id = random.choice(comment_id_list_copy)
video_dic["user_id"] = comment_id # video_dic["user_id"] = comment_id
comment_id_list_copy.remove(comment_id) # comment_id_list_copy.remove(comment_id)
video_dic["create_time"] = (create_time + datetime.timedelta(hours=random.randint(0,24),minutes=random.randint(0,60))).timestamp() # video_dic["create_time"] = (create_time + datetime.timedelta(hours=random.randint(0,24),minutes=random.randint(0,60))).timestamp()
comment_list.append(copy.deepcopy(video_dic)) # comment_list.append(copy.deepcopy(video_dic))
# rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/reply/batch_create") # # rpc_res = post_single_data(copy.deepcopy(video_dic), "cims/reply/batch_create")
rpc_res = post_muilty_data(comment_list, "cims/reply/batch_create") # rpc_res = post_muilty_data(comment_list, "cims/reply/batch_create")
except Exception as e: # except Exception as e:
print("comment error") # print("comment error")
print(e) # print(e)
except Exception as e: # except Exception as e:
print(e) # print(e)
continue # continue
# break # break
\ No newline at end of file
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment