def bulk_del_blackword(shop_id, word_list, camp_id = None, adg_id_list = None, opter = 1, opter_name = ''):
del_id_list = []
if (not camp_id) and (not adg_id_list):
return del_id_list
word_list_escaped = []
for word in word_list:
for sign in Const.COMMON_RE_SIGN:
word = word.replace(sign, '\%s' % sign)
word_list_escaped.append(word)
start_time = date_2datetime(datetime.date.today() - datetime.timedelta(15))
filter_dict = {'shop_id':shop_id,
'$or':[{'word':re.compile(word)} for word in word_list_escaped],
'rpt_list':{'$not':{'$elemMatch':{'date':{'$gte':start_time},
'$or':[{'directpaycount':{'$gt':0}}, {'indirectpaycount':{'$gt':0}}]
}}}
}
if adg_id_list:
filter_dict.update({'adgroup_id':{'$in':adg_id_list}})
elif camp_id:
filter_dict.update({'campaign_id':camp_id})
else:
return del_id_list
kw_cursor = kw_coll.find(filter_dict, {'campaign_id':1, 'adgroup_id':1, '_id':1, 'word':1})
kw_arg_dict = {}
for keyword in kw_cursor:
campaign_id, adgroup_id, keyword_id, word = keyword['campaign_id'], keyword['adgroup_id'], keyword['_id'], keyword['word']
if not kw_arg_dict.has_key(campaign_id):
kw_arg_dict[campaign_id] = []
kw_arg_list = kw_arg_dict[campaign_id]
kw_arg_list.append([adgroup_id, keyword_id, word, 0, 0, '包含屏蔽词'])
for campaign_id, kw_arg_list in kw_arg_dict.items():
sub_del_id_list = delete_keywords(shop_id = shop_id, campaign_id = campaign_id, kw_arg_list = kw_arg_list, data_type = 404, opter = opter, opter_name = opter_name)
del_id_list.extend(sub_del_id_list)
return del_id_list
def get_garbage_words(shop_id):
'''获取整店所有无展现词 (无展现词定义：连续15天展现为0且未激活）'''
start_time = date_2datetime(datetime.datetime.now() - datetime.timedelta(days = 15))
result = []
mnt_campid_list = MntMnger.get_longtail_camp_ids(shop_id = shop_id)
kw_cur = kw_coll.find({'shop_id': shop_id, 'is_garbage':True, 'campaign_id':{'$ne':mnt_campid_list}}, {'campaign_id':1, 'adgroup_id':1, 'word':1, 'rpt_list': {'$slice':-15}})
for kw in kw_cur:
impr = 0
if kw.has_key('rpt_list'):
for rpt in kw['rpt_list']:
if rpt['date'] >= start_time:
impr += rpt['impressions']
if impr == 0:
result.append({'kw_id':kw['_id'], 'word':kw['word'], 'camp_id':kw['campaign_id'], 'adg_id':kw['adgroup_id']})
return result
def filter_dupl_words(shop_id, condition, word_list):
'''获取一批重复词按条件过滤后的所有关键词信息'''
del_day, del_level, statistics_type, del_offline = condition['del_day'], condition['del_level'], condition['del_statistics_type'], condition['del_offline']
mnt_campid_list = MntMnger.get_longtail_camp_ids(shop_id = shop_id)
if not word_list:
return []
kw_list = list(kw_coll.find({'shop_id': shop_id, 'campaign_id':{'$ne':mnt_campid_list}, 'word': {'$in':word_list}}, {'word':1, 'qscore':1, 'campaign_id':1, 'adgroup_id':1, 'rpt_list': {'$slice':-del_day}}).sort('word'))
if del_offline:
camp_id_list, adg_id_list = [], []
for kw in kw_list:
camp_id_list.append(kw['campaign_id'])
adg_id_list.append(kw['adgroup_id'])
camp_cur = camp_coll.find({'_id':{'$in':camp_id_list}}, {'online_status':1})
camp_status_dict = {camp['_id']:camp['online_status'] for camp in camp_cur}
adg_cur = adg_coll.find({'shop_id':shop_id, '_id':{'$in':adg_id_list}}, {'online_status':1})
adg_status_dict = {adg['_id']:adg['online_status'] for adg in adg_cur}
kw_list.append({'word':None})
temp_word, temp_list, result, count = '', [], [], 0
start_time = date_2datetime(datetime.datetime.now() - datetime.timedelta(days = del_day))
for kw in kw_list:
# 当一个重复词遍历完后，排序，过滤后加入到result, 并初始化temp_word, temp_list
if temp_word != kw['word']:
temp_list = sorted(temp_list, key = itemgetter('ctr', 'impr', 'qscore'), reverse = True) # 按点击率、展现量、质量得分降序排序
count = 0
for obj in temp_list:
count += 1
if count > del_level and (statistics_type == '0' or obj[statistics_type] == 0) and (del_offline == 0 or obj['camp_status'] == obj['adg_status'] == 'online'):
result.append({'kw_id':obj['kw_id'], 'word':temp_word, 'camp_id':obj['camp_id'], 'adg_id':obj['adg_id']})
temp_word, temp_list = kw['word'], []
# 将当前词的相关信息存入 temp_list
if kw['word']:
impr, click = 0, 0
if del_offline and kw.has_key('rpt_list'):
for rpt in kw['rpt_list']:
if rpt['date'] >= start_time:
impr += rpt['impressions']
click += rpt['click']
temp_list.append({'kw_id': kw['_id'], 'impr': impr, 'click': click, 'camp_id': kw['campaign_id'], 'adg_id': kw['adgroup_id'], 'qscore':kw['qscore'],
'ctr': impr and click / impr,
'adg_status':del_offline and adg_status_dict[kw['adgroup_id']] or '',
'camp_status': del_offline and camp_status_dict[kw['campaign_id']] or '',
})
return result
def kw_list(self):
if not hasattr(self, '_kw_list'):
kw_list = list(
kw_coll.find({
'shop_id': self.shop_id,
'campaign_id': self.campaign_id,
'adgroup_id': self.adgroup_id,
'mnt_opt_type': {
'$ne': 2
},
'_id': {
'$in': self.kw_rtrpt_dict.keys()
}
}))
self._kw_list = trans_batch_dict_2document(src_dict_list=kw_list,
class_object=Keyword)
return self._kw_list
def get_dupl_word_info(shop_id, last_day, word):
'''获取一个重复词的所有关键词信息'''
mnt_campid_list = MntMnger.get_longtail_camp_ids(shop_id = shop_id)
kw_list = list(kw_coll.find({'shop_id': shop_id, 'word': word, 'campaign_id':{'$nin':mnt_campid_list}}, {'rpt_list':{'$slice':-last_day}}))
camp_id_list, adg_id_list = [], []
for kw in kw_list:
camp_id_list.append(kw['campaign_id'])
adg_id_list.append(kw['adgroup_id'])
camp_cur = camp_coll.find({'_id':{'$in':camp_id_list}}, {'title':1})
camp_dict = {camp['_id']:camp['title'] for camp in camp_cur}
adg_cur = adg_coll.find({'shop_id':shop_id, '_id':{'$in':adg_id_list}}, {'item_id':1})
adg_dict = {adg['_id']:adg['item_id'] for adg in adg_cur}
item_cur = item_coll.find({'shop_id':shop_id, '_id':{'$in':adg_dict.values()}}, {'title':1, 'pic_url':1, 'price':1})
item_dict = {item['_id']:(item['title'], item['price'], item['pic_url']) for item in item_cur }
result = []
start_time = date_2datetime(datetime.datetime.now() - datetime.timedelta(days = last_day))
no_item_tuple = ('该宝贝可能不存在或者下架，请尝试同步数据', 0, '/site_media/jl/img/no_photo') # 当item不存在时，使用默认
for kw in kw_list:
cost, impr, click, pay = 0, 0, 0, 0
if kw.has_key('rpt_list'):
for rpt in kw['rpt_list']:
if rpt['date'] >= start_time:
cost += rpt['cost']
impr += rpt['impressions']
click += rpt['click']
pay += rpt['directpay'] + rpt['indirectpay']
temp_key = kw['adgroup_id']
if temp_key in adg_dict:
item_id = adg_dict[kw['adgroup_id']]
temp_item_info = item_dict.get(item_id, no_item_tuple)
result.append({'kw_id':kw['_id'], 'max_price': kw['max_price'], 'qscore': kw['qscore'], 'impr': impr, 'click': click, 'pay': pay,
'cpc': click and cost / click, 'ctr': impr and click / impr, 'camp_title': camp_dict[kw['campaign_id']],
'item_title': temp_item_info[0], 'item_price':temp_item_info[1], 'pic_url': temp_item_info[2], 'item_id':item_id
})
result = sorted(result, key = itemgetter('ctr', 'impr', 'qscore', 'max_price'), reverse = True) # 按点击率、展现量、质量得分、出价降序排序
return result