class Table(DataClass("Table", [
"name",
"url",
"query_path",
"timestamp"
])):
@property
def columns(self):
return FromESMetadata.singlton.get_columns(table_name=self.name)
Column = DataClass(
"Column",
[
"name",
"table",
"es_column",
"es_index",
"type",
{
"name": "useSource",
"default": False
},
{
"name": "nested_path",
"nulls": True
}, # AN ARRAY OF PATHS (FROM DEEPEST TO SHALLOWEST) INDICATING THE JSON SUB-ARRAYS
{
"name": "relative",
"nulls": True
},
{
"name": "count",
"nulls": True
},
{
"name": "cardinality",
"nulls": True
},
{
"name": "partitions",
"nulls": True
},
{
"name": "last_updated",
"nulls": True
}
])
"timestamp"
])):
@property
def columns(self):
return FromESMetadata.singlton.get_columns(table_name=self.name)
Column = DataClass(
"Column",
[
# "table",
"names", # MAP FROM TABLE NAME TO COLUMN NAME (ONE COLUMN CAN HAVE MULTIPLE NAMES)
"es_column",
"es_index",
# "es_type",
"type",
{"name": "useSource", "default": False},
{"name": "nested_path", "nulls": True}, # AN ARRAY OF PATHS (FROM DEEPEST TO SHALLOWEST) INDICATING THE JSON SUB-ARRAYS
{"name": "count", "nulls": True},
{"name": "cardinality", "nulls": True},
{"name": "partitions", "nulls": True},
{"name": "last_updated", "nulls": True}
]
)
class ColumnList(Container):
"""
OPTIMIZED FOR THE PARTICULAR ACCESS PATTERNS USED
"""
ColumnMapping = DataClass(
"ColumnMapping",
[
{ # EDGES ARE AUTOMATICALLY INCLUDED IN THE OUTPUT, USE THIS TO INDICATE EDGES SO WE DO NOT DOUBLE-PRINT
"name":"is_edge",
"default": False
},
{ # TRACK NUMBER OF TABLE COLUMNS THIS column REPRESENTS
"name":"num_push_columns",
"nulls": True
},
{ # NAME OF THE PROPERTY (USED BY LIST FORMAT ONLY)
"name": "push_name",
"nulls": True
},
{ # PATH INTO COLUMN WHERE VALUE IS STORED ("." MEANS COLUMN HOLDS PRIMITIVE VALUE)
"name": "push_child",
"nulls": True
},
{ # THE COLUMN NUMBER
"name": "push_column",
"nulls": True
},
{ # THE COLUMN NAME FOR TABLES AND CUBES (WITH NO ESCAPING DOTS, NOT IN LEAF FORM)
"name": "push_column_name",
"nulls": True
},
{ # A FUNCTION THAT WILL RETURN A VALUE
"name": "pull",
"nulls": True
},
{ # A LIST OF MULTI-SQL REQUIRED TO GET THE VALUE FROM THE DATABASE
"name": "sql",
"type": list
},
"type", # THE NAME OF THE JSON DATA TYPE EXPECTED
{ # A LIST OF PATHS EACH INDICATING AN ARRAY
"name": "nested_path",
"type": list,
"default": ["."]
},
"column_alias"
],
constraint={"and": [
{"in": {"type": ["null", "boolean", "number", "string", "object"]}},
{"gte": [{"length": "nested_path"}, 1]}
]}
)