def _resolve_query_with_cache(
cache_info: _CacheInfo,
categories: Optional[List[str]],
chunksize: Optional[Union[int, bool]],
use_threads: bool,
session: Optional[boto3.Session],
) -> Union[pd.DataFrame, Iterator[pd.DataFrame]]:
"""Fetch cached data and return it as a pandas DataFrame (or list of DataFrames)."""
_logger.debug("cache_info:\n%s", cache_info)
if cache_info.query_execution_id is None:
raise RuntimeError("Trying to resolve with cache but w/o any query execution ID.")
query_metadata: _QueryMetadata = _get_query_metadata(
query_execution_id=cache_info.query_execution_id,
boto3_session=session,
categories=categories,
query_execution_payload=cache_info.query_execution_payload,
)
if cache_info.file_format == "parquet":
return _fetch_parquet_result(
query_metadata=query_metadata,
keep_files=True,
categories=categories,
chunksize=chunksize,
use_threads=use_threads,
boto3_session=session,
)
if cache_info.file_format == "csv":
return _fetch_csv_result(
query_metadata=query_metadata,
keep_files=True,
chunksize=chunksize,
use_threads=use_threads,
boto3_session=session,
)
raise exceptions.InvalidArgumentValue(f"Invalid data type: {cache_info.file_format}.")
def _resolve_query_without_cache_regular(
sql: str,
database: Optional[str],
data_source: Optional[str],
s3_output: Optional[str],
keep_files: bool,
chunksize: Union[int, bool, None],
categories: Optional[List[str]],
encryption: Optional[str],
workgroup: Optional[str],
kms_key: Optional[str],
wg_config: _WorkGroupConfig,
use_threads: bool,
s3_additional_kwargs: Optional[Dict[str, Any]],
boto3_session: boto3.Session,
) -> Union[pd.DataFrame, Iterator[pd.DataFrame]]:
_logger.debug("sql: %s", sql)
query_id: str = _start_query_execution(
sql=sql,
wg_config=wg_config,
database=database,
data_source=data_source,
s3_output=s3_output,
workgroup=workgroup,
encryption=encryption,
kms_key=kms_key,
boto3_session=boto3_session,
)
_logger.debug("query_id: %s", query_id)
query_metadata: _QueryMetadata = _get_query_metadata(
query_execution_id=query_id,
boto3_session=boto3_session,
categories=categories,
metadata_cache_manager=_cache_manager,
)
return _fetch_csv_result(
query_metadata=query_metadata,
keep_files=keep_files,
chunksize=chunksize,
use_threads=use_threads,
boto3_session=boto3_session,
s3_additional_kwargs=s3_additional_kwargs,
)
def _resolve_query_without_cache_ctas(
sql: str,
database: Optional[str],
data_source: Optional[str],
s3_output: Optional[str],
keep_files: bool,
chunksize: Union[int, bool, None],
categories: Optional[List[str]],
encryption: Optional[str],
workgroup: Optional[str],
kms_key: Optional[str],
wg_config: _WorkGroupConfig,
name: Optional[str],
use_threads: bool,
s3_additional_kwargs: Optional[Dict[str, Any]],
boto3_session: boto3.Session,
) -> Union[pd.DataFrame, Iterator[pd.DataFrame]]:
path: str = f"{s3_output}/{name}"
ext_location: str = "\n" if wg_config.enforced is True else f",\n external_location = '{path}'\n"
sql = (f'CREATE TABLE "{name}"\n'
f"WITH(\n"
f" format = 'Parquet',\n"
f" parquet_compression = 'SNAPPY'"
f"{ext_location}"
f") AS\n"
f"{sql}")
_logger.debug("sql: %s", sql)
try:
query_id: str = _start_query_execution(
sql=sql,
wg_config=wg_config,
database=database,
data_source=data_source,
s3_output=s3_output,
workgroup=workgroup,
encryption=encryption,
kms_key=kms_key,
boto3_session=boto3_session,
)
except botocore.exceptions.ClientError as ex:
error: Dict[str, Any] = ex.response["Error"]
if error[
"Code"] == "InvalidRequestException" and "Exception parsing query" in error[
"Message"]:
raise exceptions.InvalidCtasApproachQuery(
"Is not possible to wrap this query into a CTAS statement. Please use ctas_approach=False."
)
if error[
"Code"] == "InvalidRequestException" and "extraneous input" in error[
"Message"]:
raise exceptions.InvalidCtasApproachQuery(
"Is not possible to wrap this query into a CTAS statement. Please use ctas_approach=False."
)
raise ex
_logger.debug("query_id: %s", query_id)
try:
query_metadata: _QueryMetadata = _get_query_metadata(
query_execution_id=query_id,
boto3_session=boto3_session,
categories=categories,
metadata_cache_manager=_cache_manager,
)
except exceptions.QueryFailed as ex:
msg: str = str(ex)
if "Column name" in msg and "specified more than once" in msg:
raise exceptions.InvalidCtasApproachQuery(
f"Please, define distinct names for your columns OR pass ctas_approach=False. Root error message: {msg}"
)
if "Column name not specified" in msg:
raise exceptions.InvalidArgumentValue(
"Please, define all columns names in your query. (E.g. 'SELECT MAX(col1) AS max_col1, ...')"
)
if "Column type is unknown" in msg:
raise exceptions.InvalidArgumentValue(
"Please, don't leave undefined columns types in your query. You can cast to ensure it. "
"(E.g. 'SELECT CAST(NULL AS INTEGER) AS MY_COL, ...')")
raise ex
return _fetch_parquet_result(
query_metadata=query_metadata,
keep_files=keep_files,
categories=categories,
chunksize=chunksize,
use_threads=use_threads,
s3_additional_kwargs=s3_additional_kwargs,
boto3_session=boto3_session,
)
