altimate-datapilot-cli 0.3.4__tar.gz → 0.3.5__tar.gz

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.4
+current_version = 0.3.5
 commit = True
 tag = True
 

{altimate_datapilot_cli-0.3.4/src/altimate_datapilot_cli.egg-info → altimate_datapilot_cli-0.3.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: altimate-datapilot-cli
-Version: 0.3.4
+Version: 0.3.5
 Summary: Assistant for Data Teams
 Home-page: https://github.com/AltimateAI/datapilot-cli
 Author: Altimate Inc

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/README.md

@@ -42,6 +42,35 @@ The [--catalog-path] is an optional argument. If you don't specify a catalog pat
 
 The [--config-path] is an optional argument. You can provide a yaml file with overrides for the default behavior of the insights.
 
+#### Verbose / Debug Logging
+
+By default, API failures are reported as a short summary such as `Error in uploading the manifest.`
+To see the underlying HTTP status codes and the error bodies returned by the API, enable debug
+logging with either the `--debug` flag or the `DATAPILOT_DEBUG` environment variable:
+
+```shell
+datapilot dbt onboard --debug ...
+
+# Or, for CI/CD pipelines where the command line is generated:
+export DATAPILOT_DEBUG=1
+datapilot dbt onboard ...
+```
+
+This turns the generic message into an actionable one:
+
+```
+DEBUG:APIClient:Sending GET request for tenant acme at url: https://api.myaltimate.com/dbt/v1/signed_url
+DEBUG:APIClient:Request params: {'dbt_core_integration_id': '2', 'dbt_core_integration_environment_type': 'DEV', 'file_type': 'manifest'}
+DEBUG:APIClient:HTTP Error: {'detail': 'dbt_core_integration with id:2 and env:DEV not found'} - Status code: 400
+Error in uploading the manifest.
+```
+
+Note that `--dbt_core_integration_environment` is matched exactly, including case, against the
+environments configured for your integration.
+
+Debug output is safe to share: the API token is never logged, and the credentials in presigned
+upload URLs are redacted to `?<redacted>`.
+
 #### Generating Manifest and Catalog Files for dbt Projects
 
 1. Generate Manifest File (manifest.json). Open your dbt project's root directory in a terminal or command prompt. Run `dbt compile`. This command generates manifest.json in the target folder under your dbt project directory structure.

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/docs/conf.py

@@ -15,7 +15,7 @@ project = "datapilot-cli"
 year = "2024"
 author = "Altimate Inc."
 copyright = f"{year}, {author}"
-version = release = "0.3.4"
+version = release = "0.3.5"
 
 pygments_style = "trac"
 templates_path = ["."]

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/setup.py

@@ -13,7 +13,7 @@ def read(*names, **kwargs):
 
 setup(
     name="altimate-datapilot-cli",
-    version="0.3.4",
+    version="0.3.5",
     license="MIT",
     description="Assistant for Data Teams",
     long_description="{}\n{}".format(

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5/src/altimate_datapilot_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: altimate-datapilot-cli
-Version: 0.3.4
+Version: 0.3.5
 Summary: Assistant for Data Teams
 Home-page: https://github.com/AltimateAI/datapilot-cli
 Author: Altimate Inc

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/src/altimate_datapilot_cli.egg-info/SOURCES.txt

@@ -193,6 +193,7 @@ src/datapilot/schemas/constants.py
 src/datapilot/schemas/nodes.py
 src/datapilot/schemas/sql.py
 src/datapilot/utils/__init__.py
+src/datapilot/utils/logging_utils.py
 src/datapilot/utils/utils.py
 src/datapilot/utils/formatting/__init__.py
 src/datapilot/utils/formatting/utils.py
@@ -252,8 +253,11 @@ src/vendor/dbt_artifacts_parser/resources/run-results/run-results_v6.json
 src/vendor/dbt_artifacts_parser/resources/sources/sources_v1.json
 src/vendor/dbt_artifacts_parser/resources/sources/sources_v2.json
 src/vendor/dbt_artifacts_parser/resources/sources/sources_v3.json
+tests/cli/__init__.py
+tests/cli/test_debug_option.py
 tests/clients/__init__.py
 tests/clients/altimate/__init__.py
+tests/clients/altimate/test_client_redaction.py
 tests/clients/altimate/test_utils.py
 tests/core/__init__.py
 tests/core/platform/__init__.py
@@ -282,4 +286,5 @@ tests/test_vendor/test_catalog_v1.py
 tests/test_vendor/test_run_results_pre_v6.py
 tests/test_vendor/test_run_results_v6.py
 tests/utils/__init__.py
+tests/utils/test_logging_utils.py
 tests/utils/test_utils.py

altimate_datapilot_cli-0.3.5/src/datapilot/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.3.5"

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/src/datapilot/cli/decorators.py

@@ -9,6 +9,27 @@ from typing import Optional
 import click
 from dotenv import load_dotenv
 
+from datapilot.utils.logging_utils import DEBUG_ENV_VAR
+from datapilot.utils.logging_utils import configure_logging
+
+
+def debug_option(f):
+    """Decorator adding a --debug flag, also settable via DATAPILOT_DEBUG."""
+
+    @click.option(
+        "--debug",
+        is_flag=True,
+        default=False,
+        envvar=DEBUG_ENV_VAR,
+        help=f"Enable verbose debug logging, including API status codes and error responses. Can also be set with {DEBUG_ENV_VAR}=1.",
+    )
+    @wraps(f)
+    def wrapper(*args, **kwargs):
+        configure_logging(debug=kwargs.pop("debug", False))
+        return f(*args, **kwargs)
+
+    return wrapper
+
 
 def load_config_from_file() -> Optional[Dict]:
     """Load configuration from ~/.altimate/altimate.json if it exists."""

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/src/datapilot/clients/altimate/client.py

@@ -6,6 +6,8 @@ from requests.exceptions import HTTPError
 from requests.exceptions import RequestException
 from requests.exceptions import Timeout
 
+from datapilot.utils.logging_utils import redact_url
+
 
 class APIClient:
     def __init__(self, api_token="", base_url="", tenant=""):
@@ -36,6 +38,8 @@ class APIClient:
 
         try:
             self.logger.debug(f"Sending GET request for tenant {self.tenant} at url: {url}")
+            if params:
+                self.logger.debug(f"Request params: {params}")
             response = requests.get(url, headers=headers, params=params, timeout=timeout)
 
             # Check if the response was successful
@@ -67,7 +71,8 @@ class APIClient:
     def put(self, endpoint, data, timeout=None):
         url = f"{self.base_url}{endpoint}"
 
-        self.logger.debug(f"Sending PUT request for tenant {self.tenant} at url: {url}")
+        # Presigned upload URLs carry AWS credentials in the query string.
+        self.logger.debug(f"Sending PUT request for tenant {self.tenant} at url: {redact_url(url)}")
         response = requests.put(url, data=data, timeout=timeout)
         self.logger.debug(f"Received PUT response with status: {response.status_code}")
         return response

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/src/datapilot/clients/altimate/utils.py

@@ -8,6 +8,7 @@ from requests import Response
 
 from datapilot.clients.altimate.client import APIClient
 from datapilot.clients.altimate.constants import SUPPORTED_ARTIFACT_TYPES
+from datapilot.utils.logging_utils import redact_url
 
 
 def check_token_and_instance(
@@ -89,7 +90,7 @@ def onboard_file(api_token, tenant, dbt_core_integration_id, dbt_core_integratio
     if signed_url_data:
         signed_url = signed_url_data.get("url")
         file_id = signed_url_data.get("dbt_core_integration_file_id")
-        api_client.log(f"Received signed URL: {signed_url}")
+        api_client.log(f"Received signed URL: {redact_url(signed_url)}")
         api_client.log(f"Received File ID: {file_id}")
 
         upload_response = upload_content_to_signed_url(file_path, signed_url)

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/src/datapilot/core/mcp_utils/mcp.py

@@ -1,6 +1,5 @@
 import asyncio
 import json
-import logging
 import shutil
 from dataclasses import dataclass
 
@@ -10,7 +9,9 @@ from mcp import ClientSession
 from mcp import StdioServerParameters
 from mcp.client.stdio import stdio_client
 
-logging.basicConfig(level=logging.INFO)
+from datapilot.utils.logging_utils import configure_logging
+
+configure_logging()
 
 
 @dataclass

{altimate_datapilot_cli-0.3.4 → altimate_datapilot_cli-0.3.5}/src/datapilot/core/platforms/dbt/cli/cli.py

@@ -1,8 +1,7 @@
-import logging
-
 import click
 
 from datapilot.cli.decorators import auth_options
+from datapilot.cli.decorators import debug_option
 from datapilot.clients.altimate.utils import check_token_and_instance
 from datapilot.clients.altimate.utils import get_all_dbt_configs
 from datapilot.clients.altimate.utils import onboard_file
@@ -21,9 +20,11 @@ from datapilot.core.platforms.dbt.utils import load_manifest
 from datapilot.core.platforms.dbt.utils import load_run_results
 from datapilot.core.platforms.dbt.utils import load_sources
 from datapilot.utils.formatting.utils import tabulate_data
+from datapilot.utils.logging_utils import configure_logging
 from datapilot.utils.utils import map_url_to_instance
 
-logging.basicConfig(level=logging.INFO)
+# Honour DATAPILOT_DEBUG even for code paths that never reach a command callback.
+configure_logging()
 
 
 # New dbt group
@@ -33,6 +34,7 @@ def dbt():
 
 
 @dbt.command("project-health")
+@debug_option
 @auth_options
 @click.option(
     "--manifest-path",
@@ -134,6 +136,7 @@ def project_health(
 
 
 @dbt.command("onboard")
+@debug_option
 @auth_options
 @click.option(
     "--dbt_core_integration_id",

altimate_datapilot_cli-0.3.5/src/datapilot/utils/logging_utils.py

@@ -0,0 +1,79 @@
+"""Logging configuration for the datapilot CLI.
+
+Verbose output is opt-in and can be turned on two ways:
+
+- the ``DATAPILOT_DEBUG`` environment variable, which is the option to reach for
+  in CI/CD pipelines where the command line is generated and hard to edit
+- the ``--debug`` flag on individual commands
+
+Both raise the root logger to ``DEBUG``, which surfaces the HTTP status codes and
+API error bodies that ``APIClient`` already records but normally discards.
+"""
+
+import logging
+import os
+from typing import Optional
+from urllib.parse import urlsplit
+from urllib.parse import urlunsplit
+
+DEBUG_ENV_VAR = "DATAPILOT_DEBUG"
+
+# Values that count as "on" for DATAPILOT_DEBUG. Anything else (including "0",
+# "false" and the empty string) leaves debug logging off.
+_TRUTHY_VALUES = frozenset({"1", "true", "yes", "on"})
+
+# Debug mode is sticky: a command-level `--debug` must not be undone by a later
+# call that happens to default to False.
+_debug_enabled = False
+
+
+def debug_enabled_via_env() -> bool:
+    """Return True when DATAPILOT_DEBUG is set to a truthy value."""
+    return os.environ.get(DEBUG_ENV_VAR, "").strip().lower() in _TRUTHY_VALUES
+
+
+def configure_logging(debug: bool = False) -> bool:
+    """Configure root logging for the CLI and return whether debug mode is on.
+
+    Safe to call more than once; the group callback and the command callback both
+    call it, and the more verbose of the two wins.
+    """
+    global _debug_enabled
+    _debug_enabled = _debug_enabled or debug or debug_enabled_via_env()
+    level = logging.DEBUG if _debug_enabled else logging.INFO
+
+    root = logging.getLogger()
+    # Only install a handler when nothing else has, so embedding applications
+    # (and pytest's caplog) keep theirs. Setting the level is what actually
+    # decides whether the DEBUG records get through.
+    if not root.handlers:
+        logging.basicConfig(level=level)
+    root.setLevel(level)
+
+    # urllib3 logs each request line in full, which for a presigned S3 upload means
+    # the AWS key and signature. Our own client logs the status codes and request
+    # params, so nothing diagnostic is lost by holding urllib3 at INFO.
+    logging.getLogger("urllib3").setLevel(logging.INFO)
+
+    return _debug_enabled
+
+
+def is_debug_enabled() -> bool:
+    """Return whether debug logging is currently enabled."""
+    return _debug_enabled
+
+
+def redact_url(url: Optional[str]) -> str:
+    """Strip a URL's query string so it is safe to log.
+
+    Presigned upload URLs carry AWS credentials and a signature in the query
+    string, and debug output routinely gets pasted into support tickets.
+    """
+    if not url:
+        return ""
+
+    parts = urlsplit(url)
+    if not parts.query:
+        return url
+
+    return urlunsplit((parts.scheme, parts.netloc, parts.path, "<redacted>", ""))

altimate_datapilot_cli-0.3.5/tests/cli/test_debug_option.py

@@ -0,0 +1,65 @@
+import logging
+
+import pytest
+from click.testing import CliRunner
+
+from datapilot.cli.main import datapilot
+from datapilot.utils import logging_utils
+from datapilot.utils.logging_utils import DEBUG_ENV_VAR
+
+
+@pytest.fixture(autouse=True)
+def _reset_logging_state(monkeypatch):
+    monkeypatch.setattr(logging_utils, "_debug_enabled", False)
+    monkeypatch.delenv(DEBUG_ENV_VAR, raising=False)
+    original_level = logging.getLogger().level
+    yield
+    logging.getLogger().setLevel(original_level)
+
+
+def invoke_project_health(env=None, extra_args=()):
+    """project-health runs fully offline, so it exercises the flag without network access."""
+    return CliRunner().invoke(
+        datapilot,
+        ["dbt", "project-health", "--manifest-path", "tests/data/manifest_v11.json", *extra_args],
+        env=env or {},
+    )
+
+
+class TestDebugFlag:
+    def test_flag_is_advertised_in_help(self):
+        result = CliRunner().invoke(datapilot, ["dbt", "onboard", "--help"])
+
+        assert result.exit_code == 0
+        assert "--debug" in result.output
+        assert DEBUG_ENV_VAR in result.output
+
+    def test_flag_enables_debug_logging(self):
+        result = invoke_project_health(extra_args=["--debug"])
+
+        assert result.exit_code == 0
+        assert logging_utils.is_debug_enabled() is True
+        assert logging.getLogger().level == logging.DEBUG
+
+    def test_without_flag_stays_at_info(self):
+        result = invoke_project_health()
+
+        assert result.exit_code == 0
+        assert logging_utils.is_debug_enabled() is False
+        assert logging.getLogger().level == logging.INFO
+
+
+class TestDebugEnvVar:
+    def test_env_var_enables_debug_logging(self):
+        result = invoke_project_health(env={DEBUG_ENV_VAR: "1"})
+
+        assert result.exit_code == 0
+        assert logging_utils.is_debug_enabled() is True
+        assert logging.getLogger().level == logging.DEBUG
+
+    def test_env_var_off_stays_at_info(self):
+        result = invoke_project_health(env={DEBUG_ENV_VAR: "0"})
+
+        assert result.exit_code == 0
+        assert logging_utils.is_debug_enabled() is False
+        assert logging.getLogger().level == logging.INFO

altimate_datapilot_cli-0.3.5/tests/clients/altimate/test_client_redaction.py

@@ -0,0 +1,66 @@
+import logging
+from unittest.mock import patch
+
+import pytest
+from requests import Response
+
+from datapilot.clients.altimate.client import APIClient
+from datapilot.utils import logging_utils
+from datapilot.utils.logging_utils import DEBUG_ENV_VAR
+from datapilot.utils.logging_utils import configure_logging
+
+PRESIGNED_URL = (
+    "https://altimate-datapilot-freemium-prod.s3.amazonaws.com/prd/tenant%3Delastic/manifest.json"
+    "?AWSAccessKeyId=AKIAVXVSOK5H3JQFSSU4&Signature=2LV9nU5JHPVOFz&Expires=1785271367"
+)
+
+
+@pytest.fixture(autouse=True)
+def _reset_logging_state(monkeypatch):
+    monkeypatch.setattr(logging_utils, "_debug_enabled", False)
+    monkeypatch.delenv(DEBUG_ENV_VAR, raising=False)
+    original_level = logging.getLogger().level
+    yield
+    logging.getLogger().setLevel(original_level)
+
+
+def make_response(status_code=200):
+    response = Response()
+    response.status_code = status_code
+    response._content = b""
+    return response
+
+
+class TestPresignedUrlRedaction:
+    def test_put_does_not_log_aws_credentials(self, caplog):
+        client = APIClient()
+
+        with caplog.at_level(logging.DEBUG), patch("requests.put", return_value=make_response()):
+            client.put(PRESIGNED_URL, data=b"{}")
+
+        assert "AKIAVXVSOK5H3JQFSSU4" not in caplog.text
+        assert "2LV9nU5JHPVOFz" not in caplog.text
+        # The object path is still logged, which is the part that aids debugging.
+        assert "manifest.json" in caplog.text
+        assert "<redacted>" in caplog.text
+
+    def test_get_logs_request_params(self, caplog):
+        """Params identify the integration id/env, and contain no secrets."""
+        client = APIClient(api_token="secret-token", base_url="https://api.myaltimate.com", tenant="elastic")  # noqa: S106
+        params = {"dbt_core_integration_id": "2", "dbt_core_integration_environment_type": "PROD"}
+
+        with caplog.at_level(logging.DEBUG), patch("requests.get", return_value=make_response()):
+            client.get("/dbt/v1/signed_url", params=params)
+
+        assert "dbt_core_integration_environment_type" in caplog.text
+        assert "PROD" in caplog.text
+        assert "secret-token" not in caplog.text
+
+
+class TestUrllib3Silencing:
+    def test_urllib3_is_held_at_info_in_debug_mode(self):
+        """urllib3 logs whole request lines, presigned query string included."""
+        configure_logging(debug=True)
+
+        assert logging.getLogger().level == logging.DEBUG
+        assert logging.getLogger("urllib3").level == logging.INFO

altimate_datapilot_cli-0.3.5/tests/utils/test_logging_utils.py

@@ -0,0 +1,99 @@
+import logging
+
+import pytest
+
+from datapilot.utils import logging_utils
+from datapilot.utils.logging_utils import DEBUG_ENV_VAR
+from datapilot.utils.logging_utils import configure_logging
+from datapilot.utils.logging_utils import debug_enabled_via_env
+from datapilot.utils.logging_utils import is_debug_enabled
+from datapilot.utils.logging_utils import redact_url
+
+
+@pytest.fixture(autouse=True)
+def _reset_logging_state(monkeypatch):
+    """Debug mode is sticky by design, so reset it between tests."""
+    monkeypatch.setattr(logging_utils, "_debug_enabled", False)
+    monkeypatch.delenv(DEBUG_ENV_VAR, raising=False)
+    original_level = logging.getLogger().level
+    yield
+    logging.getLogger().setLevel(original_level)
+
+
+class TestDebugEnabledViaEnv:
+    @pytest.mark.parametrize("value", ["1", "true", "TRUE", "yes", "on", " True "])
+    def test_truthy_values(self, monkeypatch, value):
+        monkeypatch.setenv(DEBUG_ENV_VAR, value)
+
+        assert debug_enabled_via_env() is True
+
+    @pytest.mark.parametrize("value", ["0", "false", "no", "off", "", "   "])
+    def test_falsy_values(self, monkeypatch, value):
+        monkeypatch.setenv(DEBUG_ENV_VAR, value)
+
+        assert debug_enabled_via_env() is False
+
+    def test_unset(self):
+        assert debug_enabled_via_env() is False
+
+
+class TestConfigureLogging:
+    def test_default_is_info(self):
+        assert configure_logging() is False
+        assert is_debug_enabled() is False
+        assert logging.getLogger().level == logging.INFO
+
+    def test_debug_argument_sets_debug_level(self):
+        assert configure_logging(debug=True) is True
+        assert logging.getLogger().level == logging.DEBUG
+
+    def test_env_var_sets_debug_level(self, monkeypatch):
+        monkeypatch.setenv(DEBUG_ENV_VAR, "1")
+
+        assert configure_logging() is True
+        assert logging.getLogger().level == logging.DEBUG
+
+    def test_env_var_off_stays_at_info(self, monkeypatch):
+        monkeypatch.setenv(DEBUG_ENV_VAR, "0")
+
+        assert configure_logging() is False
+        assert logging.getLogger().level == logging.INFO
+
+    def test_debug_is_sticky_across_calls(self):
+        """An import-time configure_logging() must not undo a --debug flag, or vice versa."""
+        configure_logging(debug=True)
+        configure_logging(debug=False)
+
+        assert is_debug_enabled() is True
+        assert logging.getLogger().level == logging.DEBUG
+
+    def test_debug_records_reach_a_handler(self, caplog):
+        configure_logging(debug=True)
+
+        with caplog.at_level(logging.DEBUG):
+            logging.getLogger("APIClient").debug("HTTP Error: boom")
+
+        assert "HTTP Error: boom" in caplog.text
+
+
+class TestRedactUrl:
+    def test_url_without_query_is_unchanged(self):
+        url = "https://api.myaltimate.com/dbt/v1/signed_url"
+
+        assert redact_url(url) == url
+
+    def test_presigned_url_credentials_are_removed(self):
+        url = (
+            "https://altimate-datapilot-freemium-prod.s3.amazonaws.com/prd/tenant%3Delastic/manifest.json"
+            "?AWSAccessKeyId=AKIAVXVSOK5H3JQFSSU4&Signature=2LV9nU5JHPVOFz&Expires=1785271367"
+        )
+
+        redacted = redact_url(url)
+
+        assert "AKIAVXVSOK5H3JQFSSU4" not in redacted
+        assert "2LV9nU5JHPVOFz" not in redacted
+        assert redacted == ("https://altimate-datapilot-freemium-prod.s3.amazonaws.com/prd/tenant%3Delastic/manifest.json?<redacted>")
+
+    def test_empty_values(self):
+        assert redact_url("") == ""
+        assert redact_url(None) == ""

altimate_datapilot_cli-0.3.4/src/datapilot/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.3.4"