update

utils/func_from_es_get_article.py

@@ -3,6 +3,7 @@
 # @File  : func_from_es_get_article.py
 # @email : litao@igengmei.com
 # @author : litao
+import pymysql
 from elasticsearch import Elasticsearch
 exists_es_dic = {}
 es = Elasticsearch([
@@ -14,6 +15,26 @@ es = Elasticsearch([
         'port': 9200,
     }])
 
+def con_sql_jerry_prod(sql):
+    # 从数据库的表里获取数据
+    db = pymysql.connect(host='172.16.40.158', port=4000, user='st_user', passwd='aqpuBLYzEV7tML5RPsN1pntUzFy',
+                         db='jerry_prod')
+    cursor = db.cursor()
+    cursor.execute(sql)
+    result = cursor.fetchall()
+    db.close()
+    return result
+
+def con_sql_doris_prod(sql):
+    # 从数据库的表里获取数据
+    db = pymysql.connect(host='172.16.30.136', port=3306, user='doris', passwd='o5gbA27hXHHm',
+                         db='doris_prod')
+    cursor = db.cursor()
+    cursor.execute(sql)
+    result = cursor.fetchall()
+    db.close()
+    return result
+
 
 def get_device_num_from_es(word):
 
@@ -162,11 +183,19 @@ def get_device_num_from_es(word):
     tractate_content_num = results["hits"]["total"]
     return tractate_content_num
 
+def get_tractate_tags_from_es(doc_id):
 
-def get_es_article_num(tag_dict, allow_tag=["first_demands", "second_demands", "first_solutions", "second_solutions",
-                                            "positions", "second_positions", "tags_v3"]):
-    # {tag_name:(answer_content_num, tractate_content_num, diary_content_num, total_num)}
+    # tractate
+    results = es.get_source(
+        index='gm-dbmw-tractate-read',
+        doc_type='tractate',
+        timeout='10s',
+        size=0,
+        id=doc_id
+    )
+    return results
 
+def get_es_article_num(tag_dict, allow_tag=["first_demands", "second_demands", "first_solutions", "second_solutions","positions", "second_positions", "tags_v3"]):
     article_dict = {
         "first_demands": [],
         "second_demands": [],
@@ -291,3 +320,25 @@ def get_es_article_num(tag_dict, allow_tag=["first_demands", "second_demands", "
             article_dict[tag_type].append(data_dic)
 
     return article_dict
+
+def get_user_post_from_mysql():
+    data_tag_count = {}
+    second_demands_count_dict = {}
+    tags_v3_count_dict = {}
+    sql = """
+    select card_id from strategy_content_exposure_index where card_content_type="user_post" and preciseexposure_num>=50 and ctr>=0.05 and avg_page_stay>=20;
+    """
+    sql_res = con_sql_doris_prod(sql)
+    for card_id in sql_res:
+        es_res = get_tractate_tags_from_es(card_id)
+        for position in es_res["_source"]["second_demands"]:
+            if position in second_demands_count_dict:
+                second_demands_count_dict[position] += 1
+            else:
+                second_demands_count_dict[position] = 1
+        for tag in es_res["_source"]["tags_v3"]:
+            if tag in tags_v3_count_dict:
+                tags_v3_count_dict[tag] += 1
+            else:
+                tags_v3_count_dict[tag] = 1
+    return second_demands_count_dict,tags_v3_count_dict