recurvedata-lib 0.1.492__tar.gz → 0.1.496__tar.gz

{recurvedata_lib-0.1.492 → recurvedata_lib-0.1.496}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: recurvedata-lib
-Version: 0.1.492
+Version: 0.1.496
 Summary: Common Library for ReOrc Data Platform
 Author-email: Reorc Team <contact@recurvedata.com>
 Requires-Dist: croniter
@@ -77,6 +77,7 @@ Requires-Dist: dbt-duckdb==1.8.4; extra == 'dbt'
 Requires-Dist: dbt-extractor==0.5.1; extra == 'dbt'
 Requires-Dist: dbt-fabric==1.9.4; extra == 'dbt'
 Requires-Dist: dbt-impala==1.8.0; extra == 'dbt'
+Requires-Dist: dbt-mysql==1.8.0+recurve.1; extra == 'dbt'
 Requires-Dist: dbt-postgres==1.8.2; extra == 'dbt'
 Requires-Dist: dbt-redshift==1.8.1; extra == 'dbt'
 Requires-Dist: dbt-semantic-interfaces==0.5.1; extra == 'dbt'

{recurvedata_lib-0.1.492 → recurvedata_lib-0.1.496}/pyproject.toml

@@ -114,298 +114,6 @@ type = "confirm"
 name = "show_message"
 message = "Do you want to add body message in commit?"
 
-[dependency-groups]
-public = [
-    "adlfs==2025.8.0",
-    "aenum==3.1.16",
-    "agate==1.9.1",
-    "aiobotocore==2.24.2",
-    "aiohappyeyeballs==2.6.1",
-    "aiohttp==3.10.5",
-    "aioitertools==0.12.0",
-    "aiooss2==0.2.11",
-    "aiosignal==1.4.0",
-    "aiosmtplib==4.0.2",
-    "alembic==1.16.5",
-    "aliyun-log-python-sdk==0.9.32",
-    "aliyun-python-sdk-core==2.16.0",
-    "aliyun-python-sdk-kms==2.16.5",
-    "annotated-types==0.7.0",
-    "anyio==4.11.0",
-    "apache-airflow==2.10.3",
-    "apache-airflow-providers-common-compat==1.7.4",
-    "apache-airflow-providers-common-io==1.6.3",
-    "apache-airflow-providers-common-sql==1.28.1",
-    "apache-airflow-providers-fab==1.5.3",
-    "apache-airflow-providers-ftp==3.13.2",
-    "apache-airflow-providers-http==5.3.2",
-    "apache-airflow-providers-imap==3.9.2",
-    "apache-airflow-providers-smtp==2.3.1",
-    "apache-airflow-providers-sqlite==4.1.2",
-    "apispec==6.8.4",
-    "argcomplete==3.6.2",
-    "asgiref==3.9.2",
-    "asn1crypto==1.5.1",
-    "astronomer-cosmos==1.6.0",
-    "asynch==0.2.4",
-    "attrs==25.3.0",
-    "azure-core==1.35.1",
-    "azure-datalake-store==0.0.53",
-    "azure-identity==1.25.0",
-    "azure-storage-blob==12.26.0",
-    "babel==2.17.0",
-    "backoff==2.2.1",
-    "bcrypt==5.0.0",
-    "beautifulsoup4==4.13.5",
-    "bitarray==3.7.1",
-    "blinker==1.9.0",
-    "boto3==1.40.18",
-    "botocore==1.40.18",
-    "cachelib==0.13.0",
-    "cachetools==5.5.2",
-    "certifi==2025.8.3",
-    "cffi==2.0.0",
-    "chardet==5.2.0",
-    "charset-normalizer==3.4.3",
-    "ciso8601==2.3.3",
-    "click==8.3.0",
-    "clickclick==20.10.2",
-    "clickhouse-cityhash==1.0.2.4",
-    "clickhouse-driver==0.2.9",
-    "clickhouse-sqlalchemy==0.2.9",
-    "colorama==0.4.6",
-    "colorlog==6.9.0",
-    "configupdater==3.2",
-    "connexion==2.14.2",
-    "cos-python-sdk-v5==1.9.38",
-    "crcmod==1.7",
-    "cron-descriptor==2.0.6",
-    "croniter==6.0.0",
-    "cryptography==46.0.1",
-    "cssselect==1.3.0",
-    "cssutils==2.11.1",
-    "cython==3.1.4",
-    "cytoolz==1.0.1",
-    "daff==1.4.2",
-    "dateparser==1.2.2",
-    "db-dtypes==1.4.3",
-    "dbt-adapters==1.9.0",
-    "dbt-bigquery==1.8.3",
-    "dbt-common==1.12.0",
-    "dbt-core==1.8.0",
-    "dbt-doris==0.4.0",
-    "dbt-duckdb==1.8.4",
-    "dbt-extractor==0.6.0",
-    "dbt-fabric==1.9.6",
-    "dbt-impala==1.8.0",
-    "dbt-postgres==1.8.2",
-    "dbt-redshift==1.8.1",
-    "dbt-semantic-interfaces==0.5.1",
-    "dbt-starrocks==1.7.0",
-    "deepdiff==7.0.1",
-    "deprecated==1.2.18",
-    "dill==0.4.0",
-    "distlib==0.4.0",
-    "dnspython==2.8.0",
-    "duckdb==1.4.0",
-    "elastic-transport==9.1.0",
-    "elasticsearch==9.1.1",
-    "email-validator==2.3.0",
-    "emails==0.6",
-    "et-xmlfile==2.0.0",
-    "fastapi==0.117.1",
-    "filelock==3.19.1",
-    "flask==2.2.5",
-    "flask-appbuilder==4.5.3",
-    "flask-babel==2.0.0",
-    "flask-caching==2.3.1",
-    "flask-jwt-extended==4.7.1",
-    "flask-limiter==3.13",
-    "flask-login==0.6.3",
-    "flask-session==0.5.0",
-    "flask-sqlalchemy==2.5.1",
-    "flask-wtf==1.2.2",
-    "frozenlist==1.7.0",
-    "fsspec[adl,s3,sftp]==2025.9.0",
-    "future==1.0.0",
-    "google-api-core==2.24.0",
-    "google-auth==2.40.3",
-    "google-auth-oauthlib==1.2.2",
-    "google-cloud-bigquery-storage==2.33.1",
-    "google-cloud-bigquery[bqstorage]==3.38.0",
-    "google-cloud-core==2.4.3",
-    "google-cloud-dataproc==5.22.0",
-    "google-cloud-storage==2.19.0",
-    "google-crc32c==1.7.1",
-    "google-re2==1.1.20250805",
-    "google-resumable-media==2.7.2",
-    "googleapis-common-protos==1.70.0",
-    "greenlet==3.2.4 ; platform_machine == 'AMD64' or platform_machine == 'WIN32' or platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'ppc64le' or platform_machine == 'win32' or platform_machine == 'x86_64'",
-    "grpc-google-iam-v1==0.14.2",
-    "grpcio==1.75.0",
-    "grpcio-status==1.62.3",
-    "gspread==6.2.1",
-    "gunicorn==23.0.0",
-    "h11==0.16.0",
-    "httpcore==1.0.9",
-    "httpx==0.28.1",
-    "humanize==4.13.0",
-    "idna==3.10",
-    "importlib-metadata==6.11.0",
-    "impyla==0.18.0",
-    "inflection==0.5.1",
-    "invoke==2.2.0",
-    "isodate==0.6.1",
-    "itsdangerous==2.2.0",
-    "jinja2==3.1.6",
-    "jmespath==0.10.0",
-    "jsonschema==4.25.1",
-    "jsonschema-specifications==2025.9.1",
-    "larksuite-oapi==1.0.34",
-    "lazy-object-proxy==1.12.0",
-    "leather==0.4.0",
-    "leb128==1.0.8",
-    "limits==5.5.0",
-    "linkify-it-py==2.0.3",
-    "lockfile==0.12.2",
-    "logbook==1.5.3",
-    "loguru==0.7.3",
-    "lxml==6.0.2",
-    "lz4==4.4.4",
-    "mako==1.3.10",
-    "markdown-it-py==4.0.0",
-    "markupsafe==3.0.2",
-    "marshmallow==3.26.1",
-    "marshmallow-oneofschema==3.2.0",
-    "marshmallow-sqlalchemy==0.28.2",
-    "mashumaro==3.16",
-    "mdit-py-plugins==0.5.0",
-    "mdurl==0.1.2",
-    "methodtools==0.4.7",
-    "minimal-snowplow-tracker==0.0.2",
-    "more-itertools==10.8.0",
-    "msal==1.34.0",
-    "msal-extensions==1.3.1",
-    "msgpack==1.1.1",
-    "multidict==6.6.4",
-    "mysql-connector-python==8.2.0",
-    "networkx==3.5",
-    "numpy==2.3.3",
-    "oauthlib==3.3.1",
-    "openpyxl==3.1.5",
-    "opentelemetry-api==1.37.0",
-    "opentelemetry-exporter-otlp==1.15.0",
-    "opentelemetry-exporter-otlp-proto-grpc==1.15.0",
-    "opentelemetry-exporter-otlp-proto-http==1.15.0",
-    "opentelemetry-proto==1.15.0",
-    "opentelemetry-sdk==1.37.0",
-    "opentelemetry-semantic-conventions==0.58b0",
-    "ordered-set==4.1.0",
-    "orjson==3.11.3",
-    "oss2==2.18.4",
-    "ossfs==2025.5.0",
-    "packaging==25.0",
-    "pandas==2.3.2",
-    "pandas-gbq==0.29.2",
-    "paramiko==4.0.0",
-    "parsedatetime==2.6",
-    "pathspec==0.12.1",
-    "pendulum==3.1.0",
-    "platformdirs==4.4.0",
-    "pluggy==1.6.0",
-    "premailer==3.10.0",
-    "prison==0.2.1",
-    "propcache==0.3.2",
-    "proto-plus==1.26.1",
-    "protobuf==4.21.12",
-    "psutil==7.1.0",
-    "psycopg2-binary==2.9.10",
-    "pure-sasl==0.6.2",
-    "pyarrow==21.0.0",
-    "pyasn1==0.6.1",
-    "pyasn1-modules==0.4.2",
-    "pycparser==2.23 ; implementation_name != 'PyPy'",
-    "pycryptodome==3.23.0",
-    "pydantic==2.11.9",
-    "pydantic-core==2.33.2",
-    "pydata-google-auth==1.9.1",
-    "pygments==2.19.2",
-    "pyhive==0.7.0",
-    "pyjwt==2.10.1",
-    "pymongo==4.15.1",
-    "pymysql==1.1.2",
-    "pynacl==1.6.0",
-    "pyocclient==0.6",
-    "pyodbc==5.2.0",
-    "python-daemon==3.1.2",
-    "python-dateutil==2.9.0.post0",
-    "python-decouple==3.8",
-    "python-nvd3==0.16.0",
-    "python-slugify==8.0.4",
-    "pytimeparse==1.1.8",
-    "pytz==2025.2",
-    "pywebhdfs==0.4.1",
-    "pyyaml==6.0.3",
-    "redis==6.4.0",
-    "redshift-connector==2.0.917",
-    "referencing==0.36.2",
-    "regex==2025.9.18",
-    "requests==2.32.5",
-    "requests-oauthlib==2.0.0",
-    "requests-toolbelt==1.0.0",
-    "rfc3339-validator==0.1.4",
-    "rich==14.1.0",
-    "rich-argparse==1.7.1",
-    "rpds-py==0.27.1",
-    "rsa==4.9.1",
-    "s3fs==2025.9.0",
-    "s3transfer==0.13.1",
-    "scramp==1.4.6",
-    "setproctitle==1.3.7",
-    "setuptools==80.9.0",
-    "shellingham==1.5.4",
-    "six==1.17.0",
-    "sniffio==1.3.1",
-    "soupsieve==2.8",
-    "sqlalchemy==1.4.54",
-    "sqlalchemy-bigquery==1.15.0",
-    "sqlalchemy-doris[pymysql]==0.2.2",
-    "sqlalchemy-jsonfield==1.0.2",
-    "sqlalchemy-redshift==0.8.14",
-    "sqlalchemy-utils==0.42.0",
-    "sqlglot==27.18.0",
-    "sqlparse==0.5.3",
-    "sshtunnel==0.4.0",
-    "starlette==0.48.0",
-    "tabulate==0.9.0",
-    "tenacity==8.5.0",
-    "termcolor==3.1.0",
-    "text-unidecode==1.3",
-    "thrift==0.16.0",
-    "thrift-sasl==0.4.3",
-    "toolz==1.0.0",
-    "typer==0.19.2",
-    "typing-extensions==4.15.0",
-    "typing-inspection==0.4.1",
-    "tzdata==2025.2",
-    "tzlocal==5.3.1",
-    "uc-micro-py==1.0.3",
-    "ujson==5.11.0",
-    "universal-pathlib==0.2.6",
-    "urllib3==1.26.20",
-    "uvicorn==0.37.0",
-    "virtualenv==20.34.0",
-    "werkzeug==2.2.3",
-    "win32-setctime==1.2.0 ; sys_platform == 'win32'",
-    "wirerope==1.0.0",
-    "wrapt==1.17.3",
-    "wtforms==3.2.1",
-    "xmltodict==1.0.2",
-    "yarl==1.20.1",
-    "zipp==3.23.0",
-    "zstd==1.5.7.2",
-]
-
 [project.optional-dependencies]
 # Dependencies are managed through requirements files and automatically updated
 # See requirements/worker.in and requirements/dbt.in
@@ -655,6 +363,7 @@ dbt = [
     "dbt-redshift==1.8.1",
     "dbt-semantic-interfaces==0.5.1",
     "dbt-starrocks==1.7.0",
+    "dbt-mysql==1.8.0+recurve.1",
     "deepdiff==7.0.1",
     "deprecated==1.2.15",
     "distlib==0.3.9",

recurvedata_lib-0.1.496/recurvedata/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.1.496"

{recurvedata_lib-0.1.492 → recurvedata_lib-0.1.496}/recurvedata/client/client.py

@@ -1,8 +1,6 @@
 import json
 import logging
-import os
 import time
-from pathlib import Path
 from typing import Any, TypeVar, overload
 
 import httpx
@@ -26,90 +24,13 @@ class Client:
             config = AgentConfig.load()
         self.set_config(config)
 
-    @property
-    def is_offline_mode(self) -> bool:
-        """Check if offline mode is enabled via environment variable"""
-        return os.environ.get("RECURVE_OFFLINE_MODE", "").lower() in ("true", "1")
-
-    @property
-    def offline_data_path(self) -> Path:
-        """Get the offline data directory path"""
-        offline_path = os.environ.get("RECURVE_OFFLINE_DATA_PATH", "offline_data")
-        return Path(offline_path)
-
     def set_config(self, config: AgentConfig):
         self._config = config
-        # Only create HTTP client if not in offline mode
-        if not self.is_offline_mode:
-            self._client = httpx.Client(
-                base_url=config.server_url,
-                timeout=config.request_timeout,
-                headers={"User-Agent": f"RecurveLib/{__version__}"},
-            )
-        else:
-            self._client = None
-
-    def _resolve_offline_file_path(self, path: str, **kwargs) -> Path:
-        """Convert API path to local file path with parameterized support"""
-        # Remove leading /api/ prefix: /api/executor/connection -> executor/connection
-        if path.startswith("/api/"):
-            clean_path = path[5:]  # Remove "/api/" prefix
-        else:
-            clean_path = path.lstrip("/")
-        
-        # Extract parameters from kwargs
-        params = kwargs.get("params", {})
-        
-        # CORE OPERATOR APIs with parameter-based file structure:
-        
-        # 1. get_connection() API - parameterized by project_id + connection_name
-        if clean_path == "executor/connection":
-            project_id = params.get("project_id", "0")
-            connection_name = params.get("name", "default")
-            return self.offline_data_path / "executor/connection" / str(project_id) / f"{connection_name}.json"
-            
-        # 2. get_py_conn_configs() API - parameterized by project_id + project_connection_name  
-        elif clean_path == "executor/python-conn-configs":
-            project_id = params.get("project_id", "0")
-            # Python configs use project_connection_name as the key (fallback to other param names for compatibility)
-            # Handle empty strings properly - treat them as None/missing
-            project_connection_name = (params.get("project_connection_name") or 
-                                     params.get("project_conn_name") or 
-                                     params.get("pyenv_name") or 
-                                     "default")
-            return self.offline_data_path / "executor/python-conn-configs" / str(project_id) / f"{project_connection_name}.json"
-        
-        # For any other APIs, raise error do not support offline mode
-        raise APIError(f"Offline mode: {path} is not supported")
-
-    def _read_offline_data(self, method: str, path: str, response_model_class: type[ResponseModelType] | None = None, **kwargs) -> Any:
-        """Read API response from local JSON file"""
-        file_path = self._resolve_offline_file_path(path, **kwargs)
-        
-        logger.info(f"🔌 Offline mode: Reading from {file_path}")
-        
-        try:
-            if not file_path.exists():
-                logger.error(f"Offline data file not found: {file_path}")
-                raise APIError(f"Offline mode: Required data file not found: {file_path}")
-            
-            with open(file_path, 'r') as f:
-                resp_content = json.load(f)
-            
-            # Handle response model validation (same logic as online mode)
-            if response_model_class is not None:
-                if "code" in resp_content:
-                    return response_model_class.model_validate(resp_content["data"])
-                return response_model_class.model_validate(resp_content)
-            
-            return resp_content.get("data", resp_content)
-            
-        except APIError:
-            raise  # Re-raise APIError as-is
-        except Exception as e:
-            logger.error(f"Error reading offline data from {file_path}: {e}")
-            raise APIError(f"Offline mode: Failed to read data file {file_path}: {e}")
-
+        self._client = httpx.Client(
+            base_url=config.server_url,
+            timeout=config.request_timeout,
+            headers={"User-Agent": f"RecurveLib/{__version__}"},
+        )
 
     @overload
     def request(self, method: str, path: str, response_model_class: None = None, retries: int = 3, **kwargs) -> Any:
@@ -135,11 +56,6 @@ class Client:
         retries: int = 1,
         **kwargs,
     ) -> Any:
-        # Route to offline mode if enabled
-        if self.is_offline_mode:
-            return self._read_offline_data(method, path, response_model_class, **kwargs)
-        
-        # Original online mode logic
         self.prepare_header(kwargs)
         pre_err: httpx.HTTPStatusError | None = None
         for attempt in range(retries):
@@ -227,11 +143,8 @@ class Client:
                 )
 
     def close(self):
-        if self._client:
-            self._client.close()
+        self._client.close()
 
     @property
     def base_url(self) -> str:
-        if self.is_offline_mode:
-            return "offline://localhost"
-        return str(self._client.base_url) if self._client else ""
+        return str(self._client.base_url)

{recurvedata_lib-0.1.492 → recurvedata_lib-0.1.496}/recurvedata/connectors/config_schema.py

@@ -1570,6 +1570,26 @@ ALL_CONFIG_SCHEMA_DCT = {
         "group": ["destination"],
         "test_required": True,
     },
+    "recurvedata.connectors.connectors.wecom": {
+        "type": "wecom",
+        "ui_type": "WeCom",
+        "category": ["others"],
+        "config_schema": {
+            "type": "object",
+            "properties": {
+                "webhook_url": {"type": "string", "title": _l("WeCom Webhook URL")},
+            },
+            "order": [
+                "webhook_url",
+            ],
+            "required": [
+                "webhook_url",
+            ],
+        },
+        "enabled": True,
+        "group": ["integration"],
+        "test_required": True,
+    },
 }
 
 # auto generated finish

recurvedata_lib-0.1.496/recurvedata/connectors/connectors/wecom.py

@@ -0,0 +1,66 @@
+from urllib.parse import urlparse
+
+import requests
+
+from recurvedata.connectors._register import register_connector_class
+from recurvedata.connectors.base import RecurveConnectorBase
+from recurvedata.consts import ConnectorGroup
+from recurvedata.core.translation import _l
+
+CONNECTION_TYPE = "wecom"
+UI_CONNECTION_TYPE = "WeCom"
+
+
+@register_connector_class([CONNECTION_TYPE, UI_CONNECTION_TYPE])
+class WeCom(RecurveConnectorBase):
+    connection_type = CONNECTION_TYPE
+    ui_connection_type = UI_CONNECTION_TYPE
+    group = [ConnectorGroup.INTEGRATION]
+
+    config_schema = {
+        "type": "object",
+        "properties": {
+            "webhook_url": {"type": "string", "title": _l("WeCom Webhook URL")},
+        },
+        "order": [
+            "webhook_url",
+        ],
+        "required": [
+            "webhook_url",
+        ],
+    }
+
+    @property
+    def webhook_url(self):
+        return self.conf["webhook_url"]
+
+    @property
+    def wecom_conf(self):
+        if hasattr(self, "_wecom_conf"):
+            return self._wecom_conf
+        self.init_config()
+        return self._wecom_conf
+
+    def test_connection(self):
+        """
+        Test WeCom webhook connection.
+        Validate URL format and network connectivity without sending actual messages.
+        """
+        # Validate URL format
+        parsed_url = urlparse(self.webhook_url)
+        if not parsed_url.scheme or not parsed_url.netloc:
+            raise ValueError("Invalid webhook URL format")
+
+        if parsed_url.scheme != "https":
+            raise ValueError("WeCom webhook URL must use HTTPS")
+
+        # Verify WeCom domain
+        if "qyapi.weixin.qq.com" not in parsed_url.netloc:
+            raise ValueError("Invalid WeCom webhook URL domain")
+
+        # Test network connectivity (use HEAD request without sending message body)
+        try:
+            requests.head(self.webhook_url, timeout=10, allow_redirects=True)
+            # Connection test only, status code not checked (HEAD may return 405)
+        except requests.exceptions.RequestException as e:
+            raise ValueError(f"Failed to connect to WeCom webhook: {str(e)}")

{recurvedata_lib-0.1.492 → recurvedata_lib-0.1.496}/recurvedata/connectors/dbapi.py

@@ -397,19 +397,27 @@ class DBAPIBase(RecurveConnectorBase):
         set_env_dbt_password(self.password or "")
 
     @classmethod
-    def order_sql(cls, sql: str, orders: list[dict[str, str]] = None, return_sql: bool = True):
-        """
-        order the sql by the orders
-        """
+    def get_dialect(cls):
         # dialect impala -> hive, cuz there is no dialect 'impala' in sqlglot
-        dialect = "hive" if cls.connection_type == "impala" else (cls.connection_type or None)
+        return "hive" if cls.connection_type == "impala" else (cls.connection_type or None)
+
+    @classmethod
+    def clean_sql(cls, sql):
+        dialect = cls.get_dialect()
         # Parse the SQL query
         parsed = sqlglot.parse_one(sql, read=dialect)
         # since some sql dialects have special identifier, we need to use the dialect to generate the clean sql
-        clean_sql = parsed.sql(dialect=dialect, comments=False)
+        return parsed.sql(dialect=dialect, comments=False)
+
+    @classmethod
+    def order_sql(cls, sql: str, orders: list[dict[str, str]] = None, return_sql: bool = True):
+        """
+        order the sql by the orders
+        """
+        dialect = cls.get_dialect()
         # Wrap the entire query with a subquery
         alias = "_recurve_limit_subquery"
-        subquery = exp.Subquery(this=clean_sql, alias=alias)
+        subquery = exp.Subquery(this=cls.clean_sql(sql), alias=alias)
 
         # Create a new SELECT statement with the subquery and the LIMIT clause
         outer_select = exp.select("*").from_(subquery)
@@ -436,7 +444,7 @@ class DBAPIBase(RecurveConnectorBase):
         no validation on sql.
         If the sql is DML, then execute it will raise an error.
         """
-        dialect = "hive" if cls.connection_type == "impala" else (cls.connection_type or None)
+        dialect = cls.get_dialect()
 
         outer_select = cls.order_sql(sql, orders, return_sql=False)
 
@@ -456,7 +464,7 @@ class DBAPIBase(RecurveConnectorBase):
         no validation on sql.
         If the sql is DML, then execute it will raise an error.
         """
-        return f"SELECT COUNT(1) FROM ({sql}) AS cnt_subquery"
+        return f"SELECT COUNT(1) FROM ({cls.clean_sql(sql)}) AS cnt_subquery"
 
 
 @dataclass

{recurvedata_lib-0.1.492 → recurvedata_lib-0.1.496}/recurvedata/schedulers/airflow_db_process.py

@@ -148,12 +148,15 @@ class AirflowDbService:
         max_execution_date: datetime.datetime | None,
         failed_only: bool,
     ):
-        drs: list[DagRun] = DagRun.find(
-            dag_id=dag.dag_id,
-            run_id=run_id,
-            execution_start_date=min_execution_date,
-            execution_end_date=max_execution_date,
-        )
+        kwargs = {
+            "dag_id": dag.dag_id,
+            "run_id": run_id,
+            "execution_start_date": min_execution_date,
+            "execution_end_date": max_execution_date,
+        }
+        if min_execution_date or max_execution_date:
+            kwargs.pop("run_id")
+        drs: list[DagRun] = DagRun.find(**kwargs)
         if not drs:
             logger.info(f"skip rerun, no dag_run found for {dag.dag_id} at {run_id}")
             return
@@ -182,14 +185,19 @@ class AirflowDbService:
         include_downstream: bool,
         failed_only: bool,
     ):
-        drs: list[DagRun] = DagRun.find(
-            dag_id=dag.dag_id,
-            run_id=run_id,
-            execution_start_date=min_execution_date,
-            execution_end_date=max_execution_date,
-        )
+        kwargs = {
+            "dag_id": dag.dag_id,
+            "run_id": run_id,
+            "execution_start_date": min_execution_date,
+            "execution_end_date": max_execution_date,
+        }
+        if min_execution_date or max_execution_date:
+            kwargs.pop("run_id")
+        drs: list[DagRun] = DagRun.find(**kwargs)
         if not drs:
-            logger.info(f"skip rerun, no dag_run found for {dag.dag_id} at {run_id}")
+            logger.info(
+                f"skip rerun, no dag_run found for {dag.dag_id} at {run_id}, or between {min_execution_date} and {max_execution_date}"
+            )
             return
         clear_start_date = min([dr.execution_date for dr in drs])
         clear_end_date = max([dr.execution_date for dr in drs])
@@ -205,14 +213,16 @@ class AirflowDbService:
                 for task_id in clear_task_ids:
                     if task_id in dag.task_dict:
                         task = dag.task_dict[task_id]
-                        upstream_task_ids = [t.task_id for t in task.upstream_list]
+                        # Use Airflow's built-in method to get all upstream tasks
+                        upstream_task_ids = task.get_flat_relative_ids(upstream=True)
                         expanded_task_ids.update(upstream_task_ids)
 
             if include_downstream:
                 for task_id in clear_task_ids:
                     if task_id in dag.task_dict:
                         task = dag.task_dict[task_id]
-                        downstream_task_ids = [t.task_id for t in task.downstream_list]
+                        # Use Airflow's built-in method to get all downstream tasks
+                        downstream_task_ids = task.get_flat_relative_ids(upstream=False)
                         expanded_task_ids.update(downstream_task_ids)
 
         clear_task_ids = list(expanded_task_ids)

recurvedata_lib-0.1.492/recurvedata/__version__.py

@@ -1 +0,0 @@
-__version__ = "0.1.492"