snowshu.adapters.source_adapters.snowflake_adapter¶

class snowshu.adapters.source_adapters.snowflake_adapter.SnowflakeAdapter(preserve_case: bool = False)¶

The Snowflake Data Warehouse source adapter.

Parameters: preserve_case – By default the adapter folds case-insensitive strings to uppercase. If preserve_case is True,SnowShu will __not__ alter cases (dangerous!).

DATA_TYPE_MAPPINGS = {'array': json, 'bigint': bigint, 'binary': binary, 'boolean': boolean, 'char': char, 'character': char, 'date': date, 'datetime': datetime, 'decimal': decimal, 'double': float, 'double precision': float, 'float': float, 'float4': float, 'float8': float, 'int': bigint, 'integer': bigint, 'number': bigint, 'numeric': numeric, 'object': json, 'real': float, 'smallint': bigint, 'string': varchar, 'text': varchar, 'time': time, 'timestamp': timestamp_ntz, 'timestamp_ltz': timestamp_tz, 'timestamp_ntz': timestamp_ntz, 'timestamp_tz': timestamp_tz, 'varbinary': binary, 'varchar': varchar, 'variant': json}¶

SUPPORTED_SAMPLE_METHODS = (<class 'snowshu.samplings.sample_methods.bernoulli_sample_method.BernoulliSampleMethod'>,)¶

static analyze_wrap_statement(sql: str, relation: snowshu.core.models.relation.Relation) → str¶

check_count_and_query(query: str, max_count: int, unsampled: bool) → pandas.core.frame.DataFrame¶: checks the count, if count passes returns results as a dataframe.

create_table(query: str, name: str, schema: str, database: str = 'SNOWSHU')¶

directionally_wrap_statement(sql: str, relation: snowshu.core.models.relation.Relation, sample_type: Optional[BaseSampleMethod]) → str¶

drop_schema(name: str, database: str = 'SNOWSHU')¶

Drop a schema and all of its contained objects (tables, views, stored procedures)

Args:
name: The name of the schema to drop database: The database name where the schema is located.

Defaults to SNOWSHU.

generate_schema(name: str, database: str = 'SNOWSHU')¶

Create a schema in the specified database.

Parameters

name – The name of the schema to create.
database – The database where the schema will be created. Defaults to ‘SNOWSHU’.

get_connection(database_override: Optional[str] = None, schema_override: Optional[str] = None) → sqlalchemy.engine.base.Engine¶

Creates a connection engine without transactions.

By default, uses the instance credentials unless database or schema override are provided.

static polymorphic_constraint_statement(relation: snowshu.core.models.relation.Relation, analyze: bool, local_key: str, remote_key: str, local_type: str, local_type_match_val: str = None) → str¶

static population_count_statement(relation: snowshu.core.models.relation.Relation) → str¶

creates the count * statement for a relation

Parameters: relation – the Relation to create the statement for.
Returns: a query that results in a single row, single column, integer value of the unsampled relation population size

static predicate_constraint_statement(relation: snowshu.core.models.relation.Relation, analyze: bool, local_key: str, remote_key: str) → str¶: builds ‘where’ strings

sample_statement_from_relation(relation: snowshu.core.models.relation.Relation, sample_type: Optional[BaseSampleMethod]) → str¶: builds the base sample statment for a given relation.

static union_constraint_statement(subject: snowshu.core.models.relation.Relation, constraint: snowshu.core.models.relation.Relation, subject_key: str, constraint_key: str, max_number_of_outliers: int) → str¶: Union statements to select outliers. This does not pull in NULL values.

static unsampled_statement(relation: snowshu.core.models.relation.Relation) → str¶

static upstream_constraint_statement(relation: snowshu.core.models.relation.Relation, local_key: str, remote_key: str) → str¶: builds upstream where constraints against downstream full population

static view_creation_statement(relation: snowshu.core.models.relation.Relation) → str¶