Commit 18fd035c authored by 张彦钊's avatar 张彦钊

change test file

parent 881c37b1
from itertools import chain, islice, cycle
import datetime
from collections import Counter
from gm_types.gaia import DIARY_ORDER_TYPE
from gm_types.doris import ANSWER_SORT_TYPE
from gm_types.doris import ARTICLE_SORT_TYPE
from gm_types.mimas import CONTENT_CLASS
from gm_types.doris import CARD_TYPE
from gm_types.gaia import CITY_LEVEL
from gm_rpcd.all import bind
import traceback
from search.utils.diary import recall_diary
from search.utils.answer import recall_answers
from search.utils.article import recall_articles
from gm_rpcd.all import context
from libs.algorithms import drop_dup
from libs.cache import redis_client
from libs.error import logging_exception
from extend.models.gaia import City, CityScale
from extend.models.gold import (
QAQueue,
WikiQueue,
IconQueue,
UserTopicQueue,
DoctorTopicQueue,
DiaryQueue,
ArticleQueue,
AnswerQueue,
DeviceQAQueue,
DeviceIconQueue,
DeviceUserTopicQueue,
DeviceDoctorTopicQueue,
DeviceAnswerQueue,
DeviceArticleQueue,
DeviceDiaryQueue,
QuestionQueue,
DeviceQuestionQueue
)
import logging
import redis
import json
from django.conf import settings
import traceback
MAX_LOAD = 200
logger = logging.getLogger(__name__)
@bind("doris/recommend/get_diaries")
def get_diaries(tags, city, offset=0, size=10, city_tag_id=None):
# NOTE: city as city id
sort_params = {}
if city_tag_id:
sort_params["user_city_tag_id"] = city_tag_id
elif city:
try:
x = City.objects.get(id=city)
sort_params["user_city_tag_id"] = x.tag_id
except City.DoesNotExist:
pass
filters = {
"is_sink": False,
"has_before_cover": True,
"has_after_cover": True,
"content_level_is_good": True
}
if tags:
filters["closure_tag_ids"] = tags
tail = offset + size
diaries_ids = []
if tail < MAX_LOAD:
diaries = recall_diary(None, 0, 200, filters, DIARY_ORDER_TYPE.RECOMMEND, sort_params, fields=["id", "user.id"])
diaries_items = [(diary['id'], diary['user']['id']) for diary in diaries]
drop_dup_diaries = drop_dup(diaries_items)
drop_dup_size = len(drop_dup_diaries)
if tail <= drop_dup_size:
diaries_ids = [item[0] for item in drop_dup_diaries[offset:tail]]
if len(diaries_ids) == 0: # 如果头200条去重结束 后面的排序不去重
diaries = recall_diary(None, offset, size, filters, DIARY_ORDER_TYPE.RECOMMEND, sort_params, fields=["id"])
diaries_ids = [diary['id'] for diary in diaries]
return {"diaries_ids": diaries_ids}
@bind("doris/recommend/get_articles")
def get_articles(tags, offset=0, size=10):
filters = {
"content_level": [CONTENT_CLASS.EXCELLENT, CONTENT_CLASS.FINE]
}
if tags:
filters["tag_ids"] = tags
articles = recall_articles(None, offset, size, filters, ARTICLE_SORT_TYPE.RECOMMEND, {})
article_ids = [article['id'] for article in articles]
return {"article_ids": article_ids}
@bind("doris/recommend/get_answers")
def get_answers(tags, offset=0, size=10):
filters = {
"content_level": [CONTENT_CLASS.EXCELLENT, CONTENT_CLASS.FINE]
}
if tags:
filters["tag_ids"] = tags
tail = offset + size
answer_ids = []
if tail < MAX_LOAD:
answers = recall_answers(None, 0, MAX_LOAD, filters, ANSWER_SORT_TYPE.RECOMMEND, {}, fields=["id", "user_id"])
answers = filter(lambda answer: "id" in answer and "user_id" in answer, answers)
answer_items = [(answer["id"], answer["user_id"]) for answer in answers]
drop_dup_answers = drop_dup(answer_items)
if tail <= len(drop_dup_answers):
answer_ids = [item[0] for item in drop_dup_answers[offset:tail]]
if len(answer_ids) == 0:
answers = recall_answers(None, offset, size, filters, ANSWER_SORT_TYPE.RECOMMEND, {})
answer_ids = [answer['id'] for answer in answers]
return {"answer_ids": answer_ids}
@bind('doris/recommend/icon')
def fetch_icon(device_id, size):
try:
card_type = "icon"
try:
que = DeviceIconQueue.objects.get(device_id=device_id)
except DeviceIconQueue.DoesNotExist:
que = IconQueue.objects.last()
if not que:
return {"icon": []}
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
data = list(islice(cycle(que), cursor, cursor + size))
return {card_type: list(map(int, data))}
except:
logging_exception()
return {"icon": []}
@bind('doris/recommend/homepage_polymer')
def fetch_polymer_ids(device_id, size):
try:
card_type = "polymer_ids"
try:
que = DeviceIconQueue.objects.get(device_id=device_id)
except DeviceIconQueue.DoesNotExist:
que = IconQueue.objects.last()
if not que:
return {"polymer_ids": []}
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
data = list(islice(cycle(que), cursor, cursor + size))
return {card_type: list(map(int, data))}
except:
logging_exception()
return {"polymer_ids": []}
@bind('doris/recommend/feed')
def recommend_feed(device_id, card_type, city_id, size):
try:
return RecommendFeed.dispatch(device_id, card_type,
city_id, size)
except:
logging_exception()
return {card_type: []}
class RecommendFeed:
@classmethod
def dispatch(cls, device_id, card_type, city_id, size):
data = []
if card_type == CARD_TYPE.QA:
data = cls.fetch_qa(device_id, card_type, size)
elif card_type == CARD_TYPE.ANSWER:
data = cls.fetch_answer(device_id, card_type, size)
data = list(map(int, data))
elif card_type == CARD_TYPE.ARTICLE:
data = cls.fetch_article(device_id, card_type, size)
data = list(map(int, data))
elif card_type == CARD_TYPE.QUESTION:
data = cls.fetch_question(device_id, card_type, size)
data = list(map(int, data))
elif card_type == CARD_TYPE.DIARY:
data = cls.fetch_diary(device_id, card_type, city_id, size)
elif card_type == CARD_TYPE.USERTOPIC:
data = cls.fetch_user_topic(device_id,card_type,size)
elif card_type == CARD_TYPE.DOCTORTOPIC:
data = cls.fetch_doctor_topic(device_id,card_type,size)
data = list(map(int, data))
elif card_type == CARD_TYPE.ENCYCLOPEDIA:
data = cls.fetch_wiki(device_id,card_type,size)
return {card_type: data}
@staticmethod
def current_date():
return datetime.datetime.now().strftime('%Y-%m-%d')
@staticmethod
def fetch_question(device_id, card_type, size):
key = '{device_id}-{card_type}-{date}'.format(device_id=device_id,
card_type=card_type, date=RecommendFeed.current_date())
try:
que = DeviceQuestionQueue.objects.get(device_id=device_id)
except DeviceQuestionQueue.DoesNotExist:
que = QuestionQueue.objects.last()
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = redis_client.get(key) or 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
redis_client.set(key, cursor + size, ex=24 * 60 * 60)
return list(islice(cycle(que), cursor, cursor + size))
@staticmethod
def fetch_icon(device_id, card_type, size):
key = '{device_id}-{card_type}-{date}'.format(device_id=device_id,
card_type=card_type, date=RecommendFeed.current_date())
try:
que = DeviceIconQueue.objects.get(device_id=device_id)
except DeviceIconQueue.DoesNotExist:
que = IconQueue.objects.last()
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = redis_client.get(key) or 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
redis_client.set(key, cursor + size, ex=24 * 60 * 60)
return list(islice(cycle(que), cursor, cursor + size))
@staticmethod
def fetch_wiki(device_id, card_type, size):
try:
key = '{device_id}-{card_type}-{date}'.format(device_id=device_id,
card_type=card_type, date=RecommendFeed.current_date())
que = WikiQueue.objects.last()
if not que:
return []
# que = list(filter(None, que.queue.split(',')))
que = json.loads(que.queue)
# adjust args.
cursor = redis_client.get(key) or 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
redis_client.set(key, cursor + size, ex=24 * 60 * 60)
return list(islice(cycle(que), cursor, cursor + size))
except:
logging_exception()
return []
@staticmethod
def fetch_answer(device_id, card_type, size):
try:
key = '{device_id}-{card_type}-{date}'.format(device_id=device_id,
card_type=card_type, date=RecommendFeed.current_date())
try:
que = DeviceAnswerQueue.objects.get(device_id=device_id)
except DeviceAnswerQueue.DoesNotExist:
que = AnswerQueue.objects.last()
if not que:
return []
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = redis_client.get(key) or 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
redis_client.set(key, cursor + size, ex=24 * 60 * 60)
return list(islice(cycle(que), cursor, cursor + size))
except:
logging_exception()
return []
@staticmethod
def fetch_qa(device_id, card_type, size):
try:
def get_after_filter_qa():
try:
return json.loads(gmkv.get(after_filter_key))
except:
return []
def write_after_filter_qa(cid_list):
try:
if gmkv.exists(after_filter_key):
gmkv.set(after_filter_key, json.dumps(cid_list))
else:
gmkv.set(after_filter_key, json.dumps(cid_list),ex = 6*60*60)
except:
logging_exception()
logger.error("catch exception,err_log:%s" % traceback.format_exc())
def filter_qa(device_id,cid_list):
try:
key = str(device_id) + "_dislike_qa"
if gmkv.exists(key):
dislike = gmkv.smembers(key)
cid_list = [i for i in cid_list if str(i) not in dislike]
return cid_list
else:
return cid_list
except:
return cid_list
def read_history(cid_list):
if redis_client.exists(today_qa_key):
redis_client.sadd(today_qa_key, *cid_list)
else:
redis_client.sadd(today_qa_key, *cid_list)
redis_client.expire(today_qa_key, 15 * 24 * 60 * 60)
if redis_client.exists(read_qa_key) and redis_client.exists(old_qa_key):
redis_client.sdiffstore(read_qa_key, read_qa_key, old_qa_key)
redis_client.delete(old_qa_key)
redis_client.expire(read_qa_key, time=13 * 24 * 60 * 60)
redis_client.sadd(read_qa_key, *cid_list)
search_qa_recommend_list = list()
read_qa_key = "TS:recommend_answer_set:device_id:" + str(device_id)
old_qa_key = "TS:recommend_answer_set:device_id:{}:{}"\
.format(device_id,(datetime.date.today() - datetime.timedelta(days=14)).strftime("%Y-%m-%d"))
today_qa_key = "TS:recommend_answer_set:device_id:{}:{}"\
.format(device_id, datetime.date.today().strftime("%Y-%m-%d"))
answer_queue_key = "qa_is_tail:" + str(device_id)
after_filter_key = "device_qa_after_filter:device_id:" + str(device_id)
gmkv = redis.Redis(host="172.16.40.135", port=5379, db=2)
if device_id != '0':
search_qa_recommend_key = "TS:search_recommend_answer_queue:device_id:" + str(device_id)
if redis_client.exists(search_qa_recommend_key):
search_qa_recommend_dict = redis_client.hgetall(search_qa_recommend_key)
queue_list = json.loads(search_qa_recommend_dict[b'answer_queue'])
queue_list = filter_qa(device_id, queue_list)
if len(queue_list) == 0:
redis_client.delete(search_qa_recommend_key)
elif len(queue_list) == 1:
size = size - 1
search_qa_recommend_list = queue_list
redis_client.delete(search_qa_recommend_key)
else:
size = size - 1
search_qa_recommend_list.append(queue_list[0])
redis_client.hset(search_qa_recommend_key,"answer_queue",json.dumps(queue_list[1:]))
if gmkv.exists(answer_queue_key):
if len(search_qa_recommend_list) > 0:
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
read_history(search_qa_recommend_list)
return search_qa_recommend_list
elif gmkv.exists(after_filter_key):
que = get_after_filter_qa()
que = filter_qa(device_id,que)
if len(que) == 0:
gmkv.set(answer_queue_key,"tail",ex = 6*60*60)
if len(search_qa_recommend_list) > 0:
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
read_history(search_qa_recommend_list)
return search_qa_recommend_list
elif len(que) <= size:
search_qa_recommend_list.extend(que)
gmkv.set(answer_queue_key, "tail", ex=6 * 60 * 60)
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
read_history(search_qa_recommend_list)
return search_qa_recommend_list
else:
search_qa_recommend_list.extend(que[:size])
write_after_filter_qa(que[size:])
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
read_history(search_qa_recommend_list)
return search_qa_recommend_list
try:
que = DeviceQAQueue.objects.get(device_id=device_id)
except DeviceQAQueue.DoesNotExist:
que = AnswerQueue.objects.last()
if not que:
if len(search_qa_recommend_list) > 0:
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
read_history(search_qa_recommend_list)
return search_qa_recommend_list
qa = list(filter(None, que.queue.split(',')))
if device_id != "0":
qa = filter_qa(device_id,qa)
if len(qa) == 0:
if device_id != "0":
gmkv.set(answer_queue_key, "tail", ex=6 * 60 * 60)
if len(search_qa_recommend_list) > 0:
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
read_history(search_qa_recommend_list)
return search_qa_recommend_list
elif len(qa) <= size:
search_qa_recommend_list.extend(qa)
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
if device_id != "0":
gmkv.set(answer_queue_key, "tail", ex=6 * 60 * 60)
read_history(search_qa_recommend_list)
return search_qa_recommend_list
else:
search_qa_recommend_list.extend(qa[:size])
search_qa_recommend_list = list(map(int, search_qa_recommend_list))
if device_id != "0":
write_after_filter_qa(qa[size:])
read_history(search_qa_recommend_list)
return search_qa_recommend_list
except:
logging_exception()
return []
@staticmethod
def fetch_article(device_id, card_type, size):
key = '{device_id}-{card_type}-{date}'.format(device_id=device_id,
card_type=card_type, date=RecommendFeed.current_date())
try:
que = DeviceArticleQueue.objects.get(device_id=device_id)
except DeviceArticleQueue.DoesNotExist:
que = ArticleQueue.objects.last()
if not que:
return []
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = redis_client.get(key) or 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
redis_client.set(key, cursor + size, ex=24 * 60 * 60)
return list(islice(cycle(que), cursor, cursor + size))
@staticmethod
def fetch_user_topic(device_id, card_type, size):
try:
def filter_topic(cid_list):
try:
if gmkv.exists(dislike_key):
dislike = gmkv.smembers(dislike_key)
cid_list = [i for i in cid_list if str(i) not in dislike]
return cid_list
else:
return cid_list
except:
return cid_list
def write_after_filter_tractate(cid_list):
try:
if gmkv.exists(after_filter_key):
gmkv.set(after_filter_key, json.dumps(cid_list))
else:
gmkv.set(after_filter_key, json.dumps(cid_list), ex=6 * 60 * 60)
except:
logging_exception()
logger.error("catch exception,err_log:%s" % traceback.format_exc())
def get_filter_tractate():
try:
return json.loads(gmkv.get(after_filter_key))
except:
return []
def read_history(cid_list):
if redis_client.exists(today_key):
redis_client.sadd(today_key, *cid_list)
else:
redis_client.sadd(today_key, *cid_list)
redis_client.expire(today_key, 15 * 24 * 60 * 60)
if redis_client.exists(read_key) and redis_client.exists(old_key):
redis_client.sdiffstore(read_key, read_key, old_key)
redis_client.delete(old_key)
redis_client.expire(read_key, time=13 * 24 * 60 * 60)
redis_client.sadd(read_key, *cid_list)
dislike_key = str(device_id) + "_dislike_tractate"
search_topic_recommend_key = "TS:search_recommend_tractate_queue:device_id:" + str(device_id)
after_filter_key = "device_tractate_after_filter:device_id:" + str(device_id)
tractate_key = "tractate_is_tail" + str(device_id)
read_key = "TS:recommend_tractate_set:device_id:" + str(device_id)
old_key = "TS:recommend_tractate_set:device_id:{}:{}"\
.format(device_id,(datetime.date.today() - datetime.timedelta(days=14)).strftime("%Y-%m-%d"))
today_key = "TS:recommend_tractate_set:device_id:{}:{}"\
.format(device_id,datetime.date.today().strftime("%Y-%m-%d"))
search_list = list()
gmkv = redis.Redis(host="172.16.40.135", port=5379, db=2)
if (device_id != '0') and size >= 2:
if redis_client.exists(search_topic_recommend_key):
search_topic_recommend_dict = redis_client.hgetall(search_topic_recommend_key)
search_topic_recommend_list = json.loads(search_topic_recommend_dict[b'tractate_queue'])
search_topic_recommend_list = filter_topic(search_topic_recommend_list)
if len(search_topic_recommend_list) == 0:
redis_client.delete(search_topic_recommend_key)
elif len(search_topic_recommend_list) <= 2:
search_list = search_topic_recommend_list
size = size - len(search_list)
redis_client.delete(search_topic_recommend_key)
else:
search_list = search_topic_recommend_list[:2]
size = size - 2
redis_client.hset(search_topic_recommend_key, 'tractate_queue',
json.dumps(search_topic_recommend_list[2:]))
if gmkv.exists(tractate_key):
if len(search_list) > 0:
search_list = list(map(int, search_list))
read_history(search_list)
return search_list
elif gmkv.exists(after_filter_key):
que = get_filter_tractate()
que = filter_topic(que)
if len(que) == 0:
gmkv.set(tractate_key,"tail",ex = 2*60*60)
if len(search_list) > 0:
search_list = list(map(int, search_list))
read_history(search_list)
return search_list
elif len(que) <= size:
search_list.extend(que)
gmkv.set(tractate_key, "tail",ex = 2*60*60)
search_list = list(map(int, search_list))
read_history(search_list)
return search_list
else:
search_list.extend(que[:size])
write_after_filter_tractate(que[size:])
search_list = list(map(int, search_list))
read_history(search_list)
return search_list
try:
que = DeviceUserTopicQueue.objects.get(device_id=device_id)
except DeviceUserTopicQueue.DoesNotExist:
que = UserTopicQueue.objects.last()
if not que:
if len(search_list) > 0:
search_list = list(map(int, search_list))
read_history(search_list)
return search_list
qa = list(filter(None, que.queue.split(',')))
if device_id != "0":
qa = filter_topic(qa)
if len(qa) == 0:
if device_id != "0":
gmkv.set(tractate_key, "tail", ex=2 * 60 * 60)
if len(search_list) > 0:
search_list = list(map(int, search_list))
read_history(search_list)
return search_list
elif len(qa) <= size:
search_list.extend(qa)
search_list = list(map(int, search_list))
if device_id != "0":
gmkv.set(tractate_key, "tail", ex=2 * 60 * 60)
read_history(search_list)
return search_list
else:
search_list.extend(qa[:size])
search_list = list(map(int, search_list))
if device_id != "0":
write_after_filter_tractate(qa[size:])
read_history(search_list)
return search_list
except:
logging_exception()
return []
@staticmethod
def fetch_doctor_topic(device_id, card_type, size):
try:
key = '{device_id}-{card_type}-{date}'.format(device_id=device_id,
card_type=card_type, date=RecommendFeed.current_date())
try:
que = DeviceDoctorTopicQueue.objects.get(device_id=device_id)
except DeviceDoctorTopicQueue.DoesNotExist:
que = DoctorTopicQueue.objects.last()
if not que:
return []
que = list(filter(None, que.queue.split(',')))
# adjust args.
cursor = redis_client.get(key) or 0
cursor = int(cursor) % len(que)
size = min(size, len(que))
redis_client.set(key, cursor + size, ex=24 * 60 * 60)
return list(islice(cycle(que), cursor, cursor + size))
except:
logging_exception()
return []
@classmethod
def get_gm_kv_ins(cls,redis_ip, redis_port, redis_db, redis_password=""):
try:
if len(redis_password) == 0:
cli_ins = redis.Redis(host=redis_ip, port=redis_port, db=redis_db, socket_timeout=2)
else:
cli_ins = redis.Redis(host=redis_ip, port=redis_port, db=redis_db, password=redis_password,
socket_timeout=2)
cli_ins.ping()
return cli_ins
except:
return None
@classmethod
def fetch_diary_queue_data(cls, city_id, device_id=None):
local = list()
nearby = list()
nation = list()
megacity = list()
use_city_id = city_id
try:
gm_kv_ins = None
for gm_kv_host_item in settings.GM_KV_HOSTS:
gm_kv_ins = cls.get_gm_kv_ins(redis_ip=gm_kv_host_item["host"], redis_port=gm_kv_host_item["port"], redis_db=gm_kv_host_item["db"],redis_password=gm_kv_host_item["password"])
if gm_kv_ins:
break
specify_city_id_key = "diary_queue:city_id:" + use_city_id
world_city_id_key = "diary_queue:city_id:world"
if device_id is not None:
specify_city_id_key = "device_diary_queue:device_id:" + device_id + ":city_id:" + use_city_id
city_val_dict = gm_kv_ins.hgetall(specify_city_id_key)
if len(city_val_dict) == 0:
city_val_dict = gm_kv_ins.hgetall(world_city_id_key)
use_city_id = "world"
if b"native_queue" in city_val_dict and city_val_dict[b"native_queue"]:
local = list(filter(None, city_val_dict[b"native_queue"].split(b",")))
if b"nearby_queue" in city_val_dict and city_val_dict[b"nearby_queue"]:
nearby = list(filter(None, city_val_dict[b"nearby_queue"].split(b",")))
if b"nation_queue" in city_val_dict and city_val_dict[b"nation_queue"]:
nation = list(filter(None, city_val_dict[b"nation_queue"].split(b",")))
if b"megacity_queue" in city_val_dict and city_val_dict[b"megacity_queue"]:
megacity = list(filter(None, city_val_dict[b"megacity_queue"].split(b",")))
return (local, nearby, nation, megacity, use_city_id)
except:
logging_exception()
logger.error("catch exception,err_log:%s" % traceback.format_exc())
qs = DiaryQueue.objects.filter(city_id__in=[city_id, 'world'])
# Assume that world queue must exist.
if len(qs) == 1:
obj = qs[0]
else:
obj = qs[0] if qs[0].city_id == city_id else qs[1]
if obj.native_queue:
local = list(filter(None, obj.native_queue.split(',')))
if obj.nearby_queue:
nearby = list(filter(None, obj.nearby_queue.split(',')))
if obj.nation_queue:
nation = list(filter(None, obj.nation_queue.split(',')))
if obj.megacity_queue:
megacity = list(filter(None, obj.megacity_queue.split(',')))
use_city_id = obj.city_id if obj else use_city_id
return (local, nearby, nation, megacity, use_city_id)
@classmethod
def fetch_device_diary_queue_data(cls, city_id, device_id):
local = list()
nearby = list()
nation = list()
megacity = list()
use_city_id = city_id
try:
gm_kv_ins = None
for gm_kv_host_item in settings.GM_KV_HOSTS:
gm_kv_ins = cls.get_gm_kv_ins(redis_ip=gm_kv_host_item["host"], redis_port=gm_kv_host_item["port"], redis_db=gm_kv_host_item["db"],redis_password=gm_kv_host_item["password"])
if gm_kv_ins:
break
specify_city_id_key = "device_diary_queue:device_id:" + device_id + ":city_id:" + use_city_id
city_val_dict = gm_kv_ins.hgetall(specify_city_id_key)
if b"native_queue" in city_val_dict and city_val_dict[b"native_queue"]:
local = list(filter(None, city_val_dict[b"native_queue"].split(b",")))
if b"nearby_queue" in city_val_dict and city_val_dict[b"nearby_queue"]:
nearby = list(filter(None, city_val_dict[b"nearby_queue"].split(b",")))
if b"nation_queue" in city_val_dict and city_val_dict[b"nation_queue"]:
nation = list(filter(None, city_val_dict[b"nation_queue"].split(b",")))
if b"megacity_queue" in city_val_dict and city_val_dict[b"megacity_queue"]:
megacity = list(filter(None, city_val_dict[b"megacity_queue"].split(b",")))
return (local, nearby, nation, megacity, use_city_id)
except:
logging_exception()
logger.error("catch exception,err_log:%s" % traceback.format_exc())
obj = DeviceDiaryQueue.objects.filter(device_id=device_id, city_id=city_id).first()
if obj and obj.native_queue:
local = list(filter(None, obj.native_queue.split(',')))
if obj and obj.nearby_queue:
nearby = list(filter(None, obj.nearby_queue.split(',')))
if obj and obj.nation_queue:
nation = list(filter(None, obj.nation_queue.split(',')))
if obj and obj.megacity_queue:
megacity = list(filter(None, obj.megacity_queue.split(',')))
use_city_id = obj.city_id if obj else use_city_id
return (local, nearby, nation, megacity, use_city_id)
@classmethod
def fetch_diary(cls, device_id, card_type, city_id, size):
def read_history(cid_list):
if redis_client.exists(today_key):
redis_client.sadd(today_key, *cid_list)
else:
redis_client.sadd(today_key, *cid_list)
redis_client.expire(today_key, 15 * 24 * 60 * 60)
if redis_client.exists(read_key) and redis_client.exists(old_key):
redis_client.sdiffstore(read_key, read_key, old_key)
redis_client.delete(old_key)
redis_client.expire(read_key, time=13 * 24 * 60 * 60)
redis_client.sadd(read_key, *cid_list)
def dislike_cid_filter(device_id, cid_list):
try:
key = str(device_id) + "_dislike_diary"
if gmkv.exists(key):
value = gmkv.smembers(key)
cid_list = [i for i in cid_list if str(i) not in value]
return cid_list
except:
return cid_list
def fetch_after_filter_queue(device_id, city_id):
local = list()
nearby = list()
nation = list()
megacity = list()
use_city_id = city_id
try:
specify_city_id_key = "device_diary_queue_after_filter:device_id:" + device_id + ":city_id:" + use_city_id
if gmkv.exists(specify_city_id_key):
queue = gmkv.get(specify_city_id_key).split(b";")
local = list(filter(None, queue[0].split(b",")))
nearby = list(filter(None, queue[1].split(b",")))
nation = list(filter(None, queue[2].split(b",")))
megacity = list(filter(None, queue[3].split(b",")))
return (local, nearby, nation, megacity)
else:
return local, nearby, nation, megacity
except:
return local, nearby, nation, megacity
def write_after_filter_queue(device_id, city_id, local, nearby, megacity, nation):
try:
specify_city_id_key = "device_diary_queue_after_filter:device_id:" + device_id + ":city_id:" + city_id
queue = local + ";" + nearby + ";" + nation + ";" + megacity
if gmkv.exists(specify_city_id_key):
gmkv.set(specify_city_id_key, queue)
else:
gmkv.set(specify_city_id_key, queue, ex=6 * 60 * 60)
except:
logging_exception()
logger.error("catch exception,err_log:%s" % traceback.format_exc())
def get_data(local, nearby, nation, megacity, cx, cy, cm, cz, x, y, z, m, size):
nx = int(round(x * 1.0 / (x + y + z + m) * size))
ny = int(round(y * 1.0 / (x + y + z + m) * size))
nz = int(round(z * 1.0 / (x + y + z + m) * size))
nm = int(round(m * 1.0 / (x + y + z + m) * size))
nxyz = [nx, ny, nm, nz]
xyz = [x, y, m, z]
counter = Counter([nx, ny, nm, nz])
if counter[0] == 2:
nxyz[nxyz.index(0)] += size - sum(nxyz)
else:
nxyz[xyz.index(max(xyz))] += size - sum(nxyz)
nx, ny, nm, nz = nxyz
local_filter = dislike_cid_filter(device_id, cx)
if len(local_filter) == 0:
local_filter = dislike_cid_filter(device_id, local)
slocal = local_filter[:nx]
have_x = local_filter[nx:]
x_str = ",".join([str(i) for i in have_x])
ny += (nx - len(slocal))
nearby_filter = dislike_cid_filter(device_id, cy)
if len(nearby_filter) == 0:
nearby_filter = dislike_cid_filter(device_id, nearby)
snearby = nearby_filter[:ny]
have_y = nearby_filter[ny:]
y_str = ",".join([str(i) for i in have_y])
nm += (ny - len(snearby))
megacity_filter = dislike_cid_filter(device_id, cm)
if len(megacity_filter) == 0:
megacity_filter = dislike_cid_filter(device_id, megacity)
smegacity = megacity_filter[:nm]
have_m = megacity_filter[nm:]
m_str = ",".join([str(i) for i in have_m])
nz += (nm - len(smegacity))
nation_filter = dislike_cid_filter(device_id, cz)
if len(nation_filter) == 0:
nation_filter = dislike_cid_filter(device_id, nation)
snation = nation_filter[:nz]
have_z = snation[nz:]
z_str = ",".join([str(i) for i in have_z])
return chain(slocal, snearby, smegacity, snation), x_str, y_str, m_str, z_str
if device_id != '0':
portrait_list = list()
click_diary_size = 1
search_diary_size = 4
read_key = "TS:recommend_diary_set:device_id:" + str(device_id)
old_key = "TS:recommend_diary_set:device_id:{}:{}"\
.format(device_id, (datetime.date.today() - datetime.timedelta(days=14)).strftime("%Y-%m-%d"))
today_key = "TS:recommend_diary_set:device_id:{}:{}"\
.format(device_id, datetime.date.today().strftime("%Y-%m-%d"))
user_portrait_diary_key = 'user_portrait_recommend_diary_queue:device_id:%s:%s' % \
(device_id, datetime.datetime.now().strftime('%Y-%m-%d'))
gmkv = redis.Redis(host="172.16.40.135", port=5379, db=2)
if redis_client.exists(user_portrait_diary_key):
user_portrait_diary_dict = redis_client.hgetall(user_portrait_diary_key)
user_portrait_cursor = str(user_portrait_diary_dict[b'cursor'], encoding='utf-8')
if user_portrait_cursor == '0':
if b'len_cursor' in user_portrait_diary_dict.keys():
user_portrait_diary_list = json.loads(user_portrait_diary_dict[b'diary_queue'])
filter_user_portrait_diary_list = dislike_cid_filter(device_id, user_portrait_diary_list)
if len(filter_user_portrait_diary_list) > size:
portrait_list = filter_user_portrait_diary_list[:size]
redis_client.hset(user_portrait_diary_key, 'diary_queue',
json.dumps(filter_user_portrait_diary_list[size:]))
portrait_list = list(map(int, portrait_list))
read_history(portrait_list)
return portrait_list
else:
size = size - len(filter_user_portrait_diary_list)
portrait_list = filter_user_portrait_diary_list
redis_client.delete(user_portrait_diary_key)
search_diary_recommend_key = "TS:search_recommend_diary_queue:device_id:" + str(device_id)
search_list = list()
if redis_client.exists(search_diary_recommend_key) and size > 3:
search_diary_recommend_dict = redis_client.hgetall(search_diary_recommend_key)
search_diary_recommend_list = json.loads(search_diary_recommend_dict[b'diary_queue'])
search_diary_recommend_list = dislike_cid_filter(device_id, search_diary_recommend_list)
if len(search_diary_recommend_list) == 0:
redis_client.delete(search_diary_recommend_key)
elif len(search_diary_recommend_list) <= search_diary_size:
search_list = search_diary_recommend_list
size = size - len(search_diary_recommend_list)
redis_client.delete(search_diary_recommend_key)
else:
search_list = search_diary_recommend_list[:search_diary_size]
size = size - search_diary_size
redis_client.hset(search_diary_recommend_key, 'diary_queue',
json.dumps(search_diary_recommend_list[search_diary_size:]))
if size <= 0:
portrait_list.extend(search_list)
portrait_list = list(map(int, portrait_list))
read_history(portrait_list)
return portrait_list
diary_recommend_key = "TS:recommend_diary_queue:device_id:" + str(device_id)
ts_recommend_list = list()
if redis_client.exists(diary_recommend_key) and size > 0:
diary_recommend_dict = redis_client.hgetall(diary_recommend_key)
diary_recommend_list = json.loads(diary_recommend_dict[b'diary_queue'])
diary_recommend_list = dislike_cid_filter(device_id, diary_recommend_list)
if len(diary_recommend_list) == 0:
redis_client.delete(diary_recommend_key)
elif len(diary_recommend_list) <= click_diary_size:
ts_recommend_list = diary_recommend_list
redis_client.delete(diary_recommend_key)
size = size - len(ts_recommend_list)
else:
size = size - click_diary_size
ts_recommend_list = diary_recommend_list[:click_diary_size]
diary_recommend_list_json = json.dumps(diary_recommend_list[click_diary_size:])
redis_client.hset(diary_recommend_key, 'diary_queue', diary_recommend_list_json)
if size <= 0:
portrait_list.extend(search_list)
portrait_list.extend(ts_recommend_list)
portrait_list = list(map(int, portrait_list))
read_history(portrait_list)
return portrait_list
if size > 0:
try:
(local, nearby, nation, megacity, city_id) = cls.fetch_device_diary_queue_data(city_id,
device_id)
if len(local) == 0 and len(nearby) == 0 and len(nation) == 0 and len(megacity) == 0:
(local, nearby, nation, megacity, city_id) = cls.fetch_diary_queue_data(city_id)
except:
logging_exception()
(local, nearby, nation, megacity, city_id) = cls.fetch_diary_queue_data(city_id)
x, y, m, z = cls.get_city_scale(city_id)
cx, cy, cm, cz = fetch_after_filter_queue(device_id, city_id)
data, x_str, y_str, m_str, z_str = get_data(
local, nearby, nation, megacity,
cx, cy, cm, cz,
x, y, z, m, size)
write_after_filter_queue(device_id, city_id, x_str, y_str, m_str, z_str)
portrait_list.extend(search_list)
portrait_list.extend(ts_recommend_list)
portrait_list.extend(data)
if len(portrait_list) == 0:
(local, nearby, nation, megacity, city_id) = cls.fetch_diary_queue_data(city_id)
portrait_list = cls.get_queue(local, nearby, nation, megacity,
device_id, city_id, size, x, y, z, m)
portrait_list = list(map(int, portrait_list))
if len(portrait_list) != 0:
read_history(portrait_list)
return portrait_list
else:
try:
(local, nearby, nation, megacity, city_id) = cls.fetch_device_diary_queue_data(city_id, device_id)
if len(local) == 0 and len(nearby) == 0 and len(nation) == 0 and len(megacity) == 0:
(local, nearby, nation, megacity, city_id) = cls.fetch_diary_queue_data(city_id)
except:
logging_exception()
(local, nearby, nation, megacity, city_id) = cls.fetch_diary_queue_data(city_id)
x, y, m, z = cls.get_city_scale(city_id)
data = cls.get_queue(local, nearby, nation, megacity, device_id, city_id, size, x, y, z, m)
return data
@classmethod
def get_queue(cls,local, nearby, nation, megacity, device_id,city_id,size,x,y,z,m):
key = '{device_id}-{city_id}-{date}'.format(device_id=device_id,
city_id=city_id, date=RecommendFeed.current_date())
counter_key = key + '-counter_v1'
counter = redis_client.incr(counter_key)
if counter == 1:
redis_client.expire(counter_key, 24 * 60 * 60)
cursor_key = key + '-cursor_v1'
cursor = redis_client.get(cursor_key) or b'0-0-0-0'
cx, cy, cm, cz = map(int, cursor.split(b'-'))
def get_scale(local, nearby, nation, megacity, cx, cy, cm, cz, x, y, z, m, size):
nx = int(round(x * 1.0 / (x + y + z + m) * size))
ny = int(round(y * 1.0 / (x + y + z + m) * size))
nz = int(round(z * 1.0 / (x + y + z + m) * size))
nm = int(round(m * 1.0 / (x + y + z + m) * size))
nxyz = [nx, ny, nm, nz]
xyz = [x, y, m, z]
counter = Counter([nx, ny, nm, nz])
if counter[0] == 2:
nxyz[nxyz.index(0)] += size - sum(nxyz)
else:
nxyz[xyz.index(max(xyz))] += size - sum(nxyz)
nx, ny, nm, nz = nxyz
slocal = local[cx:cx + nx]
cx = min(cx + nx, len(local))
ny += (nx - len(slocal))
snearby = nearby[cy:cy + ny]
cy = min(cy + ny, len(nearby))
nm += (ny - len(snearby))
smegacity = megacity[cm: cm + nm]
cm = min(cm + nm, len(megacity))
nz += (nm - len(smegacity))
snation = nation[cz:cz + nz]
cz = min(cz + nz, len(nation))
return chain(slocal, snearby, smegacity, snation), cx, cy, cm, cz
data, ncx, ncy, ncm, ncz = get_scale(
local, nearby, nation, megacity,
cx, cy, cm, cz,
x, y, z, m, size)
if ncx == cx and ncy == cy: # native queue and nearby queue
logger.info("diary queue reach end,cx:%d,cy:%d,cm:%d,cz:%d", cx, cy, cm, cz)
ncx = ncy = ncm = ncz = 0
val = '-'.join(map(str, [ncx, ncy, ncm, ncz]))
redis_client.set(cursor_key, val, ex=24 * 60 * 60)
return list(map(int, data))
@staticmethod
def get_city_scale(city_id):
try:
c = CityScale.objects.get(city_id=city_id)
x, y, z, m = c.native, c.nearby, c.nation, c.megacity
except CityScale.DoesNotExist:
try:
c = City.objects.get(id=city_id)
if c.level in (CITY_LEVEL.SUPER, CITY_LEVEL.ONE):
x, y, m, z = 4, 3, 0, 3
elif c.level == CITY_LEVEL.TWO:
x, y, m, z = 3, 3, 0, 3
elif c.level == CITY_LEVEL.THREE:
x, y, m, z = 1, 4, 0, 5
else:
x, y, m, z = 0, 0, 0, 10
except City.DoesNotExist:
x, y, m, z = 0, 0, 0, 10
return x, y, m, z
@staticmethod
def get_scale_data(local, nearby, nation, megacity, cx, cy, cm, cz, x, y, z, m, size):
"""
:param local: local diary queue
:param nearby: nearby diary queue
:param nation: nation diary queue
:param megacity: megacity diary queue
:param cx: seen local diary offset
:param cy: seen nearby diary offset
:param cz: seen nation diary offset
:param cm: seen megacity diary offset
:param x: local diary scale factor
:param y: nearby diary scale factor
:param z: nation diary scale factor
:param m: megacity diary scale factor
:param size: nubmer of diary
:return:
"""
# 本地 临近 特大城市 全国 四个层级 都按照的是四舍五入取得方式
# 针对出现的问题,本次相应的优化是:
# 1、如果出现两个层级为零,且有剩余坑位时,则按照本地 临近 全国的优先级,先给优先级高且为零的层级一个坑位。
# 2、如果所有层级都非零,且有剩余坑位时,则优先给权重占比大的层级一个坑位。
# 3、如果只有一个层级为零,且有剩余坑位时,则优先填充权重占比大的层级一个坑位。
nx = int(round(x * 1.0 / (x + y + z + m) * size))
ny = int(round(y * 1.0 / (x + y + z + m) * size))
nz = int(round(z * 1.0 / (x + y + z + m) * size))
nm = int(round(m * 1.0 / (x + y + z + m) * size))
nxyz = [nx, ny, nm, nz]
xyz = [x, y, m, z]
counter = Counter([nx, ny, nm, nz])
if counter[0] == 2:
nxyz[nxyz.index(0)] += size - sum(nxyz)
else:
nxyz[xyz.index(max(xyz))] += size - sum(nxyz)
nx, ny, nm, nz = nxyz
slocal = local[cx:cx + nx]
cx = min(cx + nx, len(local))
ny += (nx - len(slocal))
snearby = nearby[cy:cy + ny]
cy = min(cy + ny, len(nearby))
nm += (ny - len(snearby))
smegacity = megacity[cm: cm + nm]
cm = min(cm + nm, len(megacity))
nz += (nm - len(smegacity))
snation = nation[cz:cz + nz]
cz = min(cz + nz, len(nation))
return chain(slocal, snearby, smegacity, snation), cx, cy, cm, cz
......@@ -23,7 +23,7 @@ def filter_history(device_id,cid_list):
r.sadd(all_key, *r.smembers(today_key))
def get_data():
def get_dairy():
device_id = "868080041007173"
r = redis.StrictRedis.from_url("redis://redis.paas-test.env:6379/0")
......@@ -35,7 +35,7 @@ def get_data():
user_portrait_diary_key = 'user_portrait_recommend_diary_queue:device_id:%s:%s' % \
(device_id, datetime.datetime.now().strftime('%Y-%m-%d'))
user_cids = list(range(2,20))
user_cids = list(range(2,6))
user_cids = [str(i) for i in user_cids]
r.hset(user_portrait_diary_key,'diary_queue',json.dumps(user_cids))
r.hset(user_portrait_diary_key, 'cursor', "0")
......@@ -44,7 +44,7 @@ def get_data():
print(r.hgetall(user_portrait_diary_key))
search_diary_recommend_key = "TS:search_recommend_diary_queue:device_id:" + str(device_id)
serach_cids = list(range(20,39))
serach_cids = list(range(20,26))
serach_cids = [str(i) for i in serach_cids]
r.hset(search_diary_recommend_key, 'diary_queue', json.dumps(serach_cids))
print("search")
......@@ -52,7 +52,7 @@ def get_data():
diary_recommend_key = "TS:recommend_diary_queue:device_id:" + str(device_id)
ts_cids = list(range(40,60))
ts_cids = list(range(40,46))
ts_cids = [str(i) for i in ts_cids]
r.hset(diary_recommend_key, 'diary_queue', json.dumps(ts_cids))
print("ts")
......@@ -72,6 +72,42 @@ def get_data():
print(r.hgetall(personal_key))
def get_qa():
device_id = "868080041007173"
r = redis.StrictRedis.from_url("redis://redis.paas-test.env:6379/0")
dislike_key = str(device_id) + "_dislike_qa"
dislike_cids = [529401,529412,529403]
r.sadd(dislike_key, *dislike_cids)
print("不喜欢")
print(r.smembers(dislike_key))
search_qa_recommend_key = "TS:search_recommend_answer_queue:device_id:" + str(device_id)
r.hset(search_qa_recommend_key,'answer_queue',json.dumps(list(range(529401,529408))))
print(r.hgetall(search_qa_recommend_key))
def get_topic():
device_id = "868080041007173"
r = redis.StrictRedis.from_url("redis://redis.paas-test.env:6379/0")
dislike_key = str(device_id) + "_dislike_tractate"
dislike_cids = [2,37]
r.sadd(dislike_key, *dislike_cids)
print("不喜欢")
print(r.smembers(dislike_key))
search_topic_recommend_key = "TS:search_recommend_tractate_queue:device_id:" + str(device_id)
r.hset(search_topic_recommend_key,'tractate_queue',json.dumps(list(range(1,4))))
print(r.hgetall(search_topic_recommend_key))
def yanzheng():
device_id = "E417C286-40A4-42F6-BDA9-AEEBD8FEC3B6"
r = redis.Redis(host="172.16.40.135", port=5379, db=2, socket_timeout=2000)
......@@ -84,7 +120,9 @@ def get_data():
if __name__ == "__main__":
# cid = [16,18,20]
# filter_history("hello",cid)
get_data()
get_topic()
......
......@@ -14,6 +14,7 @@ def get_esmm_users():
"where stat_date = '{}'".format(stat_date)
cursor.execute(sql)
result = list(cursor.fetchall())
db.close()
return result
......@@ -37,7 +38,23 @@ def get_user_profile():
return tags
def get_searchworlds_to_tagid():
try:
sql = 'select id, name from api_tag where is_online = 1 and tag_type < 4'
db = pymysql.connect(host='172.16.30.141', port=3306, user='work',
passwd='BJQaT9VzDcuPBqkd', db='zhengxing')
cursor = db.cursor()
cursor.execute(sql)
tag_id = cursor.fetchall()
db.close()
searchworlds_to_tagid = {}
for i in tag_id:
searchworlds_to_tagid[i[1]] = i[0]
print(searchworlds_to_tagid)
return searchworlds_to_tagid
except Exception as e:
print(e)
......@@ -80,7 +97,7 @@ if __name__ == "__main__":
#
# total_samples.todf
# to_data_base(df)
get_user_profile()
get_searchworlds_to_tagid()
......
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