добавляет dag

This commit is contained in:
2025-11-04 00:17:47 +03:00
parent 319a7baa8b
commit 745f1ef2de
17 changed files with 425 additions and 4 deletions

0
dags/task_01/__init__.py Normal file
View File

View File

@@ -0,0 +1,22 @@
from task_01.target_db.repositories import TargetDBRepo
from task_01.tasks import UserActionTransfer
class Adapters:
def __init__(self, driver):
self.driver = driver
@property
def target_repo(self):
return TargetDBRepo(self.driver)
class UserActionTransferTask:
def __init__(self, driver):
self.adapters = Adapters(driver)
@property
def run(self):
return UserActionTransfer(target_db=self.adapters.target_repo, )

8
dags/task_01/dto.py Normal file
View File

@@ -0,0 +1,8 @@
from dataclasses import dataclass
@dataclass
class ConnectionInfo:
uri: str
username: str
password: str

View File

View File

@@ -0,0 +1,41 @@
import logging
from neo4j import Driver
from pandas import DataFrame
BATCH_SIZE = 1000
class TargetDBRepo:
def __init__(self, driver: Driver):
self.driver = driver
self.log = logging.getLogger(__name__)
def save_users(self, users: DataFrame) -> None:
query = """
UNWIND $rows AS row
CREATE (u:User {
user_id: row.user_id,
action: row.action,
timestamp: row.timestamp
})
"""
with self.driver.session() as session:
for i in range(0, len(users), BATCH_SIZE):
batch = users.iloc[i:i + BATCH_SIZE]
records = batch.to_dict(orient="records")
session.run(query, {"rows": records})
self.log.info("rows saved %s", i + BATCH_SIZE)
def get_number_of_users(self) -> int:
with self.driver.session() as session:
result = session.run(
"MATCH (u:User) RETURN count(u) as user_count")
return result.single()["user_count"]
def check_connection(self):
with self.driver.session() as session:
result = session.run('RETURN "Connection successful" AS message')
return result.single()["message"]

View File

@@ -0,0 +1,21 @@
from airflow.hooks.base import BaseHook
from neo4j import Driver, GraphDatabase
from task_01.dto import ConnectionInfo
def get_neo4j_connection(conn_id: str) -> ConnectionInfo:
conn = BaseHook.get_connection(conn_id)
uri = f'bolt://{conn.host}:{conn.port}'
return ConnectionInfo(
uri,
username=conn.login,
password=conn.password,
)
def get_driver(conn_id: str) -> Driver:
connection_info = get_neo4j_connection(conn_id)
return GraphDatabase.driver(
uri=connection_info.uri,
auth=(connection_info.username, connection_info.password),
)

84
dags/task_01/tasks.py Normal file
View File

@@ -0,0 +1,84 @@
import logging
import os
import tempfile
from datetime import datetime
from random import randint
import pandas as pd
from airflow import AirflowException
from airflow.models import Variable
from task_01.target_db.repositories import TargetDBRepo
log = logging.getLogger(__name__)
class UserActionTransfer:
def __init__(
self,
target_db: TargetDBRepo,
):
self.target_db = target_db
self.log = logging.getLogger(__name__)
def generate_sample_data(self):
"""Генерация тестовых данных и сохранение в CSV файл"""
actions = self.get_fake_user_action_data()
csv_file_path = self.get_csv_file_path()
actions.to_csv(csv_file_path, index=False)
# Сохраняем путь к файлу в переменную DAG
Variable.set("user_action_data_path", csv_file_path)
logging.info("Sample data generated and saved to: %s", csv_file_path)
logging.info(f"Data preview: %s", actions.head())
def load_data_to_neo4j(self) -> None:
csv_file_path = Variable.get("user_action_data_path")
if not csv_file_path or not os.path.exists(csv_file_path):
raise AirflowException("CSV file not found: %s", csv_file_path)
# Чтение CSV файла
user_actions = pd.read_csv(csv_file_path)
logging.info("Loaded CSV data with %s rows", len(user_actions))
self.target_db.save_users(user_actions)
total_rows = self.target_db.get_number_of_users()
logging.info("Total rows: %s", total_rows)
# Очистка временного файла
os.remove(csv_file_path)
logging.info("Temporary CSV file cleaned up")
def check_neo4j_connection(self):
"""Проверка соединения с БД"""
try:
result = self.target_db.check_connection()
log.info(f"Neo4j message: {result}")
log.info("Neo4j connection is healthy")
except Exception as e:
log.error(f"Neo4j connection failed: {e}")
raise
return result
@staticmethod
def get_csv_file_path() -> str:
temp_dir = tempfile.gettempdir()
return os.path.join(temp_dir, "user_action_data.csv")
@staticmethod
def get_fake_user_action_data() -> pd.DataFrame:
actions = ["login", "purchase", "view", "logout", "search"]
ids = list(range(1, 10001))
action = []
timestamp = []
for _ in ids:
action.append(actions[randint(0, len(actions) - 1)])
timestamp.append(datetime.now())
sample_data = {
"user_id": ids,
"action": action,
"timestamp": timestamp
}
return pd.DataFrame(sample_data)