def describe_formatted(
self,
name: str,
database: str | None = None,
) -> pd.DataFrame:
"""Retrieve the results of a `DESCRIBE FORMATTED` command.
See Impala documentation for more.
Parameters
----------
name
Table name. Can be fully qualified (with database)
database
Database name
"""
from ibis.backends.impala.metadata import parse_metadata
stmt = self._table_command(
'DESCRIBE FORMATTED', name, database=database
)
result = self._exec_statement(stmt)
# Leave formatting to pandas
for c in result.columns:
result[c] = result[c].str.strip()
return parse_metadata(result)
def parsed_unpart(unpart_metadata):
return parse_metadata(unpart_metadata)
def parsed_part(part_metadata):
return parse_metadata(part_metadata)
def setUpClass(cls):
cls.spacer = ('', nan, nan)
cls.schema = [
('# col_name', 'data_type', 'comment'),
cls.spacer,
('foo', 'int', nan),
('bar', 'tinyint', nan),
('baz', 'bigint', nan),
]
cls.partitions = [
('# Partition Information', nan, nan),
('# col_name', 'data_type', 'comment'),
cls.spacer,
('qux', 'bigint', nan),
]
cls.info = [
('# Detailed Table Information', nan, nan),
('Database:', 'tpcds', nan),
('Owner:', 'wesm', nan),
('CreateTime:', '2015-11-08 01:09:42-08:00', nan),
('LastAccessTime:', 'UNKNOWN', nan),
('Protect Mode:', 'None', nan),
('Retention:', '0', nan),
(
'Location:',
('hdfs://host-name:20500/my.db' '/dbname.table_name'),
nan,
),
('Table Type:', 'EXTERNAL_TABLE', nan),
('Table Parameters:', nan, nan),
('', 'EXTERNAL', 'TRUE'),
('', 'STATS_GENERATED_VIA_STATS_TASK', 'true'),
('', 'numRows', '183592'),
('', 'transient_lastDdlTime', '1447340941'),
]
cls.storage_info = [
('# Storage Information', nan, nan),
(
'SerDe Library:',
('org.apache.hadoop' '.hive.serde2.lazy.LazySimpleSerDe'),
nan,
),
('InputFormat:', 'org.apache.hadoop.mapred.TextInputFormat', nan),
(
'OutputFormat:',
'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat',
nan,
),
('Compressed:', 'No', nan),
('Num Buckets:', '0', nan),
('Bucket Columns:', '[]', nan),
('Sort Columns:', '[]', nan),
('Storage Desc Params:', nan, nan),
('', 'field.delim', '|'),
('', 'serialization.format', '|'),
]
cls.part_metadata = pd.DataFrame.from_records(
_glue_lists_spacer(
cls.spacer,
[cls.schema, cls.partitions, cls.info, cls.storage_info],
),
columns=['name', 'type', 'comment'],
)
cls.unpart_metadata = pd.DataFrame.from_records(
_glue_lists_spacer(
cls.spacer, [cls.schema, cls.info, cls.storage_info]
),
columns=['name', 'type', 'comment'],
)
cls.parsed_part = parse_metadata(cls.part_metadata)
cls.parsed_unpart = parse_metadata(cls.unpart_metadata)