Merge branch 'offic' into 'master'

Offic

See merge request !62

eda/smart_rank/es_tool.py

 from elasticsearch import Elasticsearch as Es
+import json
 
 
 def get_es():
@@ -38,4 +39,14 @@ def es_mquery(doc, body, es=None):
         es = get_es()
     index = es_index_adapt(index_prefix='gm-dbmw', doc_type=doc, rw='read')
     res = es.msearch(body, index=index)
-    return res
\ No newline at end of file
+    return res
+
+
+def es_insert_device_info(device_id, body, es=None, rw=None):
+    if es is None:
+        es = get_es()
+    index = es_index_adapt(index_prefix="gm-dbmw", doc_type="device", rw=None)
+    bulk_head = '{"index": {"_id":"%s"}}' % device_id
+    data_str = json.dumps(body, ensure_ascii=False)
+    bulk_one_body = bulk_head + "\n" + data_str + "\n"
+    return es.bulk(index=index, doc_type="device", body=bulk_one_body)

eda/smart_rank/tag3_update_user_portrait_offline.py

@@ -4,13 +4,12 @@ import json
 import operator
 from collections import Counter
 
-import redis
-
 import pymysql
 from pyspark import SparkConf
 from pyspark.sql import SparkSession
-from tool import (get_doris_prod, get_jerry_test, get_tag3_user_log, send_email, write_user_portrait,
-                  write_user_portrait_by_event, write_user_portrait_doris)
+
+from tool import (get_doris_prod, get_redis_client, get_tag3_user_log, get_user_portrait_tag3_with_score, send_email,
+                  write_user_portrait, write_user_portrait_by_event, write_user_portrait_doris)
 
 
 # [{'激光': 1.949194898204873}, {'手术': 1.949194898204873}, {'手术': 1.949194898204873}, {'手术': 1.949194898204873, '植发际线': 7.1}]
@@ -255,7 +254,10 @@ def update_tag3_user_portrait(cl_id):
             res.update(tmp_res)
 
         key = "doris:user_portrait:tag3:device_id:" + str(cl_id)
-        redis_client = redis.StrictRedis.from_url("redis://:ReDis!GmTx*0aN9@172.16.40.173:6379")
+        redis_client = get_redis_client()
+
+        anecdote_tags_scores = get_user_portrait_tag3_with_score(cl_id).get("anecdote_tags", {})
+        res["anecdote_tags"] = anecdote_tags_scores
 
         if (len(first_demands_score.keys()) > 0) or (len(second_demands_score.keys()) > 0) or \
            (len(first_solutions_score.keys()) > 0) or (len(second_solutions_score.keys()) > 0) or \
@@ -269,6 +271,15 @@ def update_tag3_user_portrait(cl_id):
                                 ",".join(first_positions_score.keys()), ",".join(second_positions_score.keys()),
                                 ",".join(projects_score.keys()))
 
+            # body = {}
+            # for (k, v) in res.items():
+            #     body[k] = list(v.keys())
+
+            # body["device_id"] = cl_id
+            # body["last_modified"] = datetime.datetime.strftime(datetime.datetime.now(pytz.timezone("Asia/Shanghai")),
+            #                                                    "%Y-%m-%dT%H:%M:%S.%f")[:-7] + "Z"
+            # es_insert_device_info(cl_id, body)
+
             # write_user_portrait_doris(cl_id, ",".join(first_solutions_score.keys()), ",".join(second_solutions_score.keys()),
             #                           ",".join(first_demands_score.keys()), ",".join(second_demands_score.keys()),
             #                           ",".join(first_positions_score.keys()), ",".join(second_positions_score.keys()),
@@ -299,6 +310,7 @@ def consume_kafka():
         spark = SparkSession.builder.config(conf=sparkConf).enableHiveSupport().getOrCreate()
         spark.sparkContext.setLogLevel("WARN")
         spark.sparkContext.addPyFile("/srv/apps/ffm-baseline_git/eda/smart_rank/tool.py")
+        # spark.sparkContext.addPyFile("/srv/apps/ffm-baseline_git/eda/smart_rank/es_tool.py")
 
         device_ids_lst_rdd = spark.sparkContext.parallelize(device_ids_lst, numSlices=1000)
 
@@ -318,9 +330,8 @@ def consume_kafka():
 if __name__ == "__main__":
     start = datetime.datetime.now()
 
+    # update_tag3_user_portrait("862460044588666")
     # update_tag3_user_portrait("androidid_a25a1129c0b38f7b")
-    # device_id = "862460044588666"
-    # update_tag3_user_portrait(device_id)
 
     consume_kafka()
     end = datetime.datetime.now()

eda/smart_rank/tool.py

@@ -503,3 +503,45 @@ def write_user_portrait_by_event(cl_id, first_solutions, second_solutions, first
     except Exception as e:
         print("write db error")
         print(e)
+
+
+def get_redis_client():
+    return redis.StrictRedis.from_url('redis://:ReDis!GmTx*0aN9@172.16.40.173:6379')
+
+
+def get_user_portrait_tag3_from_redis(device_id, limit_score=0):
+    def items_gt_score(d):
+        new_d = dict(sorted(d.items(), key=lambda x: x[1], reverse=True))
+        res = {tag: float(score) for tag, score in new_d.items() if float(score) >= limit_score}
+        return list(res.keys())
+    portrait_key = "doris:user_portrait:tag3:device_id:" + str(device_id)
+    redis_client = get_redis_client()
+    if redis_client.exists(portrait_key):
+        user_portrait = json.loads(redis_client.get(portrait_key))
+        first_demands = items_gt_score(user_portrait.get("first_demands", {}))
+        second_demands = items_gt_score(user_portrait.get("second_demands", {}))
+        first_solutions = items_gt_score(user_portrait.get("first_solutions", {}))
+        second_solutions = items_gt_score(user_portrait.get("second_solutions", {}))
+        first_positions = items_gt_score(user_portrait.get("first_positions", {}))
+        second_positions = items_gt_score(user_portrait.get("second_positions", {}))
+        projects = items_gt_score(user_portrait.get("projects", {}))
+        anecdote_tags = items_gt_score(user_portrait.get("anecdote_tags", {}))
+        return {
+            "first_demands": first_demands,
+            "second_demands": second_demands,
+            "first_solutions": first_solutions,
+            "second_solutions": second_solutions,
+            "first_positions": first_positions,
+            "second_positions": second_positions,
+            "projects": projects,
+            "anecdote_tags": anecdote_tags
+        }
+    return {}
+
+
+def get_user_portrait_tag3_with_score(device_id):
+    portrait_key = "doris:user_portrait:tag3:device_id:" + str(device_id)
+    redis_client = get_redis_client()
+    if redis_client.exists(portrait_key):
+        return json.loads(redis_client.get(portrait_key).decode("utf-8"))
+    return {}