garf-executors 0.2.2__tar.gz

garf_executors-0.2.2/PKG-INFO

@@ -0,0 +1,72 @@
+Metadata-Version: 2.4
+Name: garf-executors
+Version: 0.2.2
+Summary: Executes queries against API and writes data to local/remote storage.
+Author-email: "Google Inc. (gTech gPS CSE team)" <no-reply@google.com>, Andrei Markin <andrey.markin.ppc@gmail.com>
+License: Apache 2.0
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Operating System :: OS Independent
+Classifier: License :: OSI Approved :: Apache Software License
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: garf-core
+Requires-Dist: garf-io
+Requires-Dist: pyyaml
+Requires-Dist: pydantic
+Requires-Dist: opentelemetry-api
+Requires-Dist: opentelemetry-sdk
+Requires-Dist: opentelemetry-exporter-otlp
+Provides-Extra: bq
+Requires-Dist: garf-io[bq]; extra == "bq"
+Requires-Dist: pandas; extra == "bq"
+Requires-Dist: google-cloud-logging; extra == "bq"
+Provides-Extra: sql
+Requires-Dist: garf-io[sqlalchemy]; extra == "sql"
+Requires-Dist: pandas; extra == "sql"
+Provides-Extra: server
+Requires-Dist: fastapi[standard]; extra == "server"
+Requires-Dist: opentelemetry-instrumentation-fastapi; extra == "server"
+Requires-Dist: typer; extra == "server"
+Provides-Extra: all
+Requires-Dist: garf-executors[bq,server,sql]; extra == "all"
+
+# `garf-executors` - One stop-shop for interacting with Reporting APIs.
+
+`garf-executors` is responsible for orchestrating process of fetching from API and storing data in a storage.
+
+Currently the following executors are supports:
+
+* `ApiExecutor` - fetching data from reporting API and saves it to a requested destination.
+* `BigQueryExecutor` - executes SQL code in BigQuery.
+* `SqlExecutor` - executes SQL code in a SqlAlchemy supported DB.
+
+## Installation
+
+`pip install garf-executors`
+
+## Usage
+
+After `garf-executors` is installed you can use `garf` utility to perform fetching.
+
+```
+garf <QUERIES> --source <API_SOURCE> \
+  --output <OUTPUT_TYPE> \
+  --source.params1=<VALUE>
+```
+
+where
+
+* `<QUERIES>`- local or remote path(s) to files with queries.
+* `<API_SOURCE>`- type of API to use. Based on that the appropriate report fetcher will be initialized.
+* `<OUTPUT_TYPE>` - output supported by [`garf-io` library](../garf_io/README.md).
+
+If your report fetcher requires additional parameters you can pass them via key value pairs under `--source.` argument, i.e.`--source.regionCode='US'` - to get data only from *US*.
+> Concrete `--source` parameters are dependent on a particular report fetcher and should be looked up in a documentation for this fetcher.

garf_executors-0.2.2/README.md

@@ -0,0 +1,32 @@
+# `garf-executors` - One stop-shop for interacting with Reporting APIs.
+
+`garf-executors` is responsible for orchestrating process of fetching from API and storing data in a storage.
+
+Currently the following executors are supports:
+
+* `ApiExecutor` - fetching data from reporting API and saves it to a requested destination.
+* `BigQueryExecutor` - executes SQL code in BigQuery.
+* `SqlExecutor` - executes SQL code in a SqlAlchemy supported DB.
+
+## Installation
+
+`pip install garf-executors`
+
+## Usage
+
+After `garf-executors` is installed you can use `garf` utility to perform fetching.
+
+```
+garf <QUERIES> --source <API_SOURCE> \
+  --output <OUTPUT_TYPE> \
+  --source.params1=<VALUE>
+```
+
+where
+
+* `<QUERIES>`- local or remote path(s) to files with queries.
+* `<API_SOURCE>`- type of API to use. Based on that the appropriate report fetcher will be initialized.
+* `<OUTPUT_TYPE>` - output supported by [`garf-io` library](../garf_io/README.md).
+
+If your report fetcher requires additional parameters you can pass them via key value pairs under `--source.` argument, i.e.`--source.regionCode='US'` - to get data only from *US*.
+> Concrete `--source` parameters are dependent on a particular report fetcher and should be looked up in a documentation for this fetcher.

garf_executors-0.2.2/garf_executors/__init__.py

@@ -0,0 +1,60 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Executors to fetch data from various APIs."""
+
+from __future__ import annotations
+
+import importlib
+
+from garf_executors import executor, fetchers
+from garf_executors.api_executor import ApiExecutionContext, ApiQueryExecutor
+from garf_executors.telemetry import tracer
+
+
+@tracer.start_as_current_span('setup_executor')
+def setup_executor(
+  source: str,
+  fetcher_parameters: dict[str, str | int | bool],
+  enable_cache: bool = False,
+  cache_ttl_seconds: int = 3600,
+) -> type[executor.Executor]:
+  """Initializes executors based on a source and parameters."""
+  if source == 'bq':
+    bq_executor = importlib.import_module('garf_executors.bq_executor')
+    query_executor = bq_executor.BigQueryExecutor(**fetcher_parameters)
+  elif source == 'sqldb':
+    sql_executor = importlib.import_module('garf_executors.sql_executor')
+    query_executor = (
+      sql_executor.SqlAlchemyQueryExecutor.from_connection_string(
+        fetcher_parameters.get('connection_string')
+      )
+    )
+  else:
+    concrete_api_fetcher = fetchers.get_report_fetcher(source)
+    query_executor = ApiQueryExecutor(
+      fetcher=concrete_api_fetcher(
+        **fetcher_parameters,
+        enable_cache=enable_cache,
+        cache_ttl_seconds=cache_ttl_seconds,
+      )
+    )
+  return query_executor
+
+
+__all__ = [
+  'ApiQueryExecutor',
+  'ApiExecutionContext',
+]
+
+__version__ = '0.2.2'

garf_executors-0.2.2/garf_executors/api_executor.py

@@ -0,0 +1,137 @@
+# Copyright 2024 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Module for executing Garf queries and writing them to local/remote.
+
+ApiQueryExecutor performs fetching data from API in a form of
+GarfReport and saving it to local/remote storage.
+"""
+# pylint: disable=C0330, g-bad-import-order, g-multiple-import
+
+from __future__ import annotations
+
+import logging
+
+from garf_core import report_fetcher
+from opentelemetry import trace
+
+from garf_executors import exceptions, execution_context, executor, fetchers
+from garf_executors.telemetry import tracer
+
+logger = logging.getLogger(__name__)
+
+
+class ApiExecutionContext(execution_context.ExecutionContext):
+  """Common context for executing one or more queries."""
+
+  writer: str | list[str] = 'console'
+
+
+class ApiQueryExecutor(executor.Executor):
+  """Gets data from API and writes them to local/remote storage.
+
+  Attributes:
+      api_client: a client used for connecting to API.
+  """
+
+  def __init__(self, fetcher: report_fetcher.ApiReportFetcher) -> None:
+    """Initializes ApiQueryExecutor.
+
+    Args:
+        fetcher: Instantiated report fetcher.
+    """
+    self.fetcher = fetcher
+    super().__init__(
+      preprocessors=self.fetcher.preprocessors,
+      postprocessors=self.fetcher.postprocessors,
+    )
+
+  @classmethod
+  def from_fetcher_alias(
+    cls,
+    source: str,
+    fetcher_parameters: dict[str, str] | None = None,
+    enable_cache: bool = False,
+    cache_ttl_seconds: int = 3600,
+  ) -> ApiQueryExecutor:
+    if not fetcher_parameters:
+      fetcher_parameters = {}
+    concrete_api_fetcher = fetchers.get_report_fetcher(source)
+    return ApiQueryExecutor(
+      fetcher=concrete_api_fetcher(
+        **fetcher_parameters,
+        enable_cache=enable_cache,
+        cache_ttl_seconds=cache_ttl_seconds,
+      )
+    )
+
+  @tracer.start_as_current_span('api.execute')
+  def execute(
+    self,
+    query: str,
+    title: str,
+    context: ApiExecutionContext,
+  ) -> str:
+    """Reads query, extract results and stores them in a specified location.
+
+    Args:
+      query: Location of the query.
+      title: Name of the query.
+      context: Query execution context.
+
+    Returns:
+      Result of writing the report.
+
+    Raises:
+      GarfExecutorError: When failed to execute query.
+    """
+    span = trace.get_current_span()
+    span.set_attribute('fetcher.class', self.fetcher.__class__.__name__)
+    span.set_attribute(
+      'api.client.class', self.fetcher.api_client.__class__.__name__
+    )
+    try:
+      span.set_attribute('query.title', title)
+      span.set_attribute('query.text', query)
+      logger.debug('starting query %s', query)
+      results = self.fetcher.fetch(
+        query_specification=query,
+        args=context.query_parameters,
+        **context.fetcher_parameters,
+      )
+      writer_clients = context.writer_clients
+      if not writer_clients:
+        logger.warning('No writers configured, skipping write operation')
+        return None
+      writing_results = []
+      for writer_client in writer_clients:
+        logger.debug(
+          'Start writing data for query %s via %s writer',
+          title,
+          type(writer_client),
+        )
+        result = writer_client.write(results, title)
+        logger.debug(
+          'Finish writing data for query %s via %s writer',
+          title,
+          type(writer_client),
+        )
+        writing_results.append(result)
+      logger.info('%s executed successfully', title)
+      # Return the last writer's result for backward compatibility
+      return writing_results[-1] if writing_results else None
+    except Exception as e:
+      logger.error('%s generated an exception: %s', title, str(e))
+      raise exceptions.GarfExecutorError(
+        '%s generated an exception: %s', title, str(e)
+      ) from e

garf_executors-0.2.2/garf_executors/bq_executor.py

@@ -0,0 +1,177 @@
+# Copyright 2024 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Executes queries in BigQuery."""
+
+from __future__ import annotations
+
+import contextlib
+import os
+
+try:
+  from google.cloud import bigquery  # type: ignore
+except ImportError as e:
+  raise ImportError(
+    'Please install garf-executors with BigQuery support '
+    '- `pip install garf-executors[bq]`'
+  ) from e
+
+import logging
+
+from garf_core import query_editor, report
+from google.cloud import exceptions as google_cloud_exceptions
+from opentelemetry import trace
+
+from garf_executors import exceptions, execution_context, executor
+from garf_executors.telemetry import tracer
+
+logger = logging.getLogger(__name__)
+
+
+class BigQueryExecutorError(exceptions.GarfExecutorError):
+  """Error when BigQueryExecutor fails to run query."""
+
+
+class BigQueryExecutor(executor.Executor, query_editor.TemplateProcessorMixin):
+  """Handles query execution in BigQuery.
+
+  Attributes:
+      project_id: Google Cloud project id.
+      location: BigQuery dataset location.
+      client: BigQuery client.
+  """
+
+  def __init__(
+    self,
+    project_id: str | None = os.getenv('GOOGLE_CLOUD_PROJECT'),
+    location: str | None = None,
+  ) -> None:
+    """Initializes BigQueryExecutor.
+
+    Args:
+        project_id: Google Cloud project id.
+        location: BigQuery dataset location.
+    """
+    if not project_id:
+      raise BigQueryExecutorError(
+        'project_id is required. Either provide it as project_id parameter '
+        'or GOOGLE_CLOUD_PROJECT env variable.'
+      )
+    self.project_id = project_id
+    self.location = location
+    super().__init__()
+
+  @property
+  def client(self) -> bigquery.Client:
+    """Instantiates bigquery client."""
+    return bigquery.Client(self.project_id)
+
+  @tracer.start_as_current_span('bq.execute')
+  def execute(
+    self,
+    query: str,
+    title: str,
+    context: execution_context.ExecutionContext = (
+      execution_context.ExecutionContext()
+    ),
+  ) -> report.GarfReport:
+    """Executes query in BigQuery.
+
+    Args:
+      query: Location of the query.
+      title: Name of the query.
+      context: Query execution context.
+
+    Returns:
+      Report with data if query returns some data otherwise empty Report.
+    """
+    span = trace.get_current_span()
+    logger.info('Executing script: %s', title)
+    query_text = self.replace_params_template(query, context.query_parameters)
+    self.create_datasets(context.query_parameters.macro)
+    job = self.client.query(query_text)
+    try:
+      result = job.result()
+    except google_cloud_exceptions.GoogleCloudError as e:
+      raise BigQueryExecutorError(
+        f'Failed to execute query {title}: Reason: {e}'
+      ) from e
+      logger.debug('%s launched successfully', title)
+    if result.total_rows:
+      results = report.GarfReport.from_pandas(result.to_dataframe())
+    else:
+      results = report.GarfReport()
+    if context.writer and results:
+      writer_clients = context.writer_clients
+      if not writer_clients:
+        logger.warning('No writers configured, skipping write operation')
+      else:
+        writing_results = []
+        for writer_client in writer_clients:
+          logger.debug(
+            'Start writing data for query %s via %s writer',
+            title,
+            type(writer_client),
+          )
+          writing_result = writer_client.write(results, title)
+          logger.debug(
+            'Finish writing data for query %s via %s writer',
+            title,
+            type(writer_client),
+          )
+          writing_results.append(writing_result)
+        # Return the last writer's result for backward compatibility
+        logger.info('%s executed successfully', title)
+        return writing_results[-1] if writing_results else None
+    logger.info('%s executed successfully', title)
+    span.set_attribute('execute.num_results', len(results))
+    return results
+
+  @tracer.start_as_current_span('bq.create_datasets')
+  def create_datasets(self, macros: dict | None) -> None:
+    """Creates datasets in BQ based on values in a dict.
+
+    If dict contains keys with 'dataset' in them, then values for such keys
+    are treated as dataset names.
+
+    Args:
+      macros: Mapping containing data for query execution.
+    """
+    if macros and (datasets := extract_datasets(macros)):
+      for dataset in datasets:
+        dataset_id = f'{self.project_id}.{dataset}'
+        try:
+          self.client.get_dataset(dataset_id)
+        except google_cloud_exceptions.NotFound:
+          bq_dataset = bigquery.Dataset(dataset_id)
+          bq_dataset.location = self.location
+          with contextlib.suppress(google_cloud_exceptions.Conflict):
+            self.client.create_dataset(bq_dataset, timeout=30)
+            logger.info('Created new dataset %s', dataset_id)
+
+
+def extract_datasets(macros: dict | None) -> list[str]:
+  """Finds dataset-related keys based on values in a dict.
+
+  If dict contains keys with 'dataset' in them, then values for such keys
+  are treated as dataset names.
+
+  Args:
+      macros: Mapping containing data for query execution.
+
+  Returns:
+      Possible names of datasets.
+  """
+  if not macros:
+    return []
+  return [value for macro, value in macros.items() if 'dataset' in macro]

garf_executors-0.2.2/garf_executors/config.py

@@ -0,0 +1,53 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# pylint: disable=C0330, g-bad-import-order, g-multiple-import
+
+"""Stores mapping between API aliases and their execution context."""
+
+from __future__ import annotations
+
+import os
+import pathlib
+
+import pydantic
+import smart_open
+import yaml
+
+from garf_executors.execution_context import ExecutionContext
+
+
+class Config(pydantic.BaseModel):
+  """Stores necessary parameters for one or multiple API sources.
+
+  Attributes:
+    source: Mapping between API source alias and execution parameters.
+  """
+
+  sources: dict[str, ExecutionContext]
+
+  @classmethod
+  def from_file(cls, path: str | pathlib.Path | os.PathLike[str]) -> Config:
+    """Builds config from local or remote yaml file."""
+    with smart_open.open(path, 'r', encoding='utf-8') as f:
+      data = yaml.safe_load(f)
+    return Config(sources=data)
+
+  def save(self, path: str | pathlib.Path | os.PathLike[str]) -> str:
+    """Saves config to local or remote yaml file."""
+    with smart_open.open(path, 'w', encoding='utf-8') as f:
+      yaml.dump(
+        self.model_dump(exclude_none=True).get('sources'), f, encoding='utf-8'
+      )
+    return f'Config is saved to {str(path)}'