recurvedata-lib 0.1.487__tar.gz → 0.1.489__tar.gz

{recurvedata_lib-0.1.487 → recurvedata_lib-0.1.489}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: recurvedata-lib
-Version: 0.1.487
+Version: 0.1.489
 Summary: Common Library for ReOrc Data Platform
 Author-email: Reorc Team <contact@recurvedata.com>
 Requires-Dist: croniter
@@ -475,7 +475,7 @@ Description-Content-Type: text/markdown
 # Recurve Libraries
 
 For the unified maintenance of public components of the Recurve platform,
-these codes may be used in both Server and Executor (Worker) environments. 
+these codes may be used in both Server and Executor (Worker) environments.
 
 Only Python 3.11+ are supported.
 

{recurvedata_lib-0.1.487 → recurvedata_lib-0.1.489}/README.md

@@ -1,7 +1,7 @@
 # Recurve Libraries
 
 For the unified maintenance of public components of the Recurve platform,
-these codes may be used in both Server and Executor (Worker) environments. 
+these codes may be used in both Server and Executor (Worker) environments.
 
 Only Python 3.11+ are supported.
 

{recurvedata_lib-0.1.487 → recurvedata_lib-0.1.489}/pyproject.toml

@@ -61,9 +61,6 @@ asyncio_mode = "auto"
 [tool.hatch.version]
 path = "recurvedata/__version__.py"
 
-[tool.hatch.publish.index]
-recurve-pypi = "https://repo.tool.reorc.cloud/repository/recurve_pypi_internal/"
-
 [tool.hatch.build.targets.sdist]
 include = ["/recurvedata"]
 
@@ -117,6 +114,232 @@ 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",
+    "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",
+    "asgiref==3.9.2",
+    "asn1crypto==1.5.1",
+    "astronomer-cosmos==1.6.0+recurve.1",
+    "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",
+    "boto3==1.40.18",
+    "botocore==1.40.18",
+    "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",
+    "clickhouse-cityhash==1.0.2.4",
+    "clickhouse-driver==0.2.9",
+    "clickhouse-sqlalchemy==0.2.9",
+    "colorama==0.4.6",
+    "cos-python-sdk-v5==1.9.38",
+    "crcmod==1.7",
+    "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==0.3.4+recurve.11",
+    "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",
+    "distlib==0.4.0",
+    "dnspython==2.8.0",
+    "duckdb==1.4.0",
+    "elastic-transport==9.1.0",
+    "elasticsearch==9.1.1",
+    "emails==0.6",
+    "et-xmlfile==2.0.0",
+    "fastapi==0.117.1",
+    "filelock==3.19.1",
+    "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-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",
+    "invoke==2.2.0",
+    "isodate==0.6.1",
+    "jinja2==3.1.6",
+    "jmespath==0.10.0",
+    "jsonschema==4.25.1",
+    "jsonschema-specifications==2025.9.1",
+    "larksuite-oapi==1.0.34",
+    "leather==0.4.0",
+    "leb128==1.0.8",
+    "logbook==1.5.3",
+    "loguru==0.7.3",
+    "lxml==6.0.2",
+    "lz4==4.4.4",
+    "markdown-it-py==4.0.0",
+    "markupsafe==3.0.2",
+    "mashumaro==3.16",
+    "mdurl==0.1.2",
+    "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==9.4.0",
+    "networkx==3.5",
+    "numpy==2.3.3",
+    "oauthlib==3.3.1",
+    "openpyxl==3.1.5",
+    "opentelemetry-api==1.37.0",
+    "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",
+    "premailer==3.10.0",
+    "propcache==0.3.2",
+    "proto-plus==1.26.1",
+    "protobuf==4.25.5",
+    "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-dateutil==2.9.0.post0",
+    "python-decouple==3.8",
+    "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",
+    "rich==14.1.0",
+    "rpds-py==0.27.1",
+    "rsa==4.9.1",
+    "s3fs==2025.9.0",
+    "s3transfer==0.13.1",
+    "scramp==1.4.6",
+    "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-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",
+    "tenacity==8.5.0",
+    "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",
+    "ujson==5.11.0",
+    "urllib3==1.26.20",
+    "uvicorn==0.37.0",
+    "virtualenv==20.34.0",
+    "win32-setctime==1.2.0 ; sys_platform == 'win32'",
+    "wrapt==1.17.3",
+    "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

recurvedata_lib-0.1.489/recurvedata/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.1.489"

{recurvedata_lib-0.1.487 → recurvedata_lib-0.1.489}/recurvedata/client/client.py

@@ -1,6 +1,8 @@
 import json
 import logging
+import os
 import time
+from pathlib import Path
 from typing import Any, TypeVar, overload
 
 import httpx
@@ -24,13 +26,90 @@ 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
-        self._client = httpx.Client(
-            base_url=config.server_url,
-            timeout=config.request_timeout,
-            headers={"User-Agent": f"RecurveLib/{__version__}"},
-        )
+        # 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}")
+
 
     @overload
     def request(self, method: str, path: str, response_model_class: None = None, retries: int = 3, **kwargs) -> Any:
@@ -56,6 +135,11 @@ 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):
@@ -143,8 +227,11 @@ class Client:
                 )
 
     def close(self):
-        self._client.close()
+        if self._client:
+            self._client.close()
 
     @property
     def base_url(self) -> str:
-        return str(self._client.base_url)
+        if self.is_offline_mode:
+            return "offline://localhost"
+        return str(self._client.base_url) if self._client else ""

{recurvedata_lib-0.1.487 → recurvedata_lib-0.1.489}/recurvedata/core/config.py

@@ -26,7 +26,7 @@ class Config:
         """load default config from env file or environment variables
 
         file format:
-        RECURVE_HOST = https://recurve.test.recurvedata.com
+        RECURVE_HOST = https://abc.test.domain.com
         RECURVE_USERNAME = foo
         RECURVE_PASSWORD = pwd123
         """

{recurvedata_lib-0.1.487 → recurvedata_lib-0.1.489}/recurvedata/operators/task.py

@@ -181,7 +181,7 @@ class BaseTask(Configurable, LineageTaskMixin):
 
     @property
     def node_url(self) -> str:
-        # https://dev-test.recurve.test.recurvedata.com/datawork/workflow?p_id=257942399102349312&wf_id=258282502478635008&open_drawer=true&node_key=D2f0I
+        # https://abc.env.name.domain.com/datawork/workflow?p_id=257942399102349312&wf_id=258282502478635008&open_drawer=true&node_key=D2f0I
         host = context.client.base_url  # todo: correct it
         query_string = urllib.parse.urlencode(
             {"p_id": self.dag.project_id, "job_id": self.dag.id, "node_key": self.node.node_key, "open_drawer": "true"}

recurvedata_lib-0.1.487/recurvedata/__version__.py

@@ -1 +0,0 @@
-__version__ = "0.1.487"