def run(self):
"""
Extracts table objects from the scope, then creates new representation
objects which are placed into the DWRepresentation
:return: A DWRepresentation object for the given scope
"""
# Gets all table objects in the scope
pygrametl = self.scope['pygrametl']
tables = pygrametl._alltables
# Creates representation objects
for table in tables:
# If the table is a dimension.
if self.check_table_type(table, DIM_CLASSES):
if isinstance(table, TypeOneSlowlyChangingDimension):
dim = SCDType1DimRepresentation(table, self.dw_conn)
elif isinstance(table, SlowlyChangingDimension):
dim = SCDType2DimRepresentation(table, self.dw_conn)
else:
dim = DimRepresentation(table, self.dw_conn)
self.dim_reps.append(dim)
# If the table is a fact table
elif self.check_table_type(table, FT_CLASSES):
ft = FTRepresentation(table, self.dw_conn)
self.fts_reps.append(ft)
# From the scope, gets all SnowflakedDimensions.
# These are used to re-create the referencing structure of the DW,
# when instantiating DWRepresentation.
snowflakes = []
for x, value in self.scope.items():
if isinstance(value, SnowflakedDimension):
snowflakes.append(value)
dw_rep = DWRepresentation(self.dim_reps, self.dw_conn, self.fts_reps,
snowflakes)
# Clears the list of tables as its contents may otherwise be retained,
# when a new Case is executed. This is because the list is mutable.
pygrametl._alltables.clear()
return dw_rep
special_snowflake = SnowflakedDimension(references=[(dim1,
[dim2, dim3]), (dim2,
dim4)])
for row in data:
special_snowflake.insert(row)
conn.commit()
dim1_rep = DimRepresentation(dim1, conn)
dim2_rep = DimRepresentation(dim2, conn)
dim3_rep = DimRepresentation(dim3, conn)
dim4_rep = DimRepresentation(dim4, conn)
snow_dw_rep = DWRepresentation([dim1_rep, dim2_rep, dim3_rep, dim4_rep],
conn,
snowflakeddims=(special_snowflake, ))
snow_dw_rep.connection.cursor().execute("DELETE FROM dim2 WHERE key2 > 3")
# quick overview of the keys we have left
for dim in snow_dw_rep.dims:
allatts = dim.all.copy()
lookupatts = dim.lookupatts.copy()
for attr in lookupatts:
allatts.remove(attr)
for row in dim.itercolumns(allatts):
print(dim.name, row)
ref_tester = ReferentialIntegrityPredicate(refs={
}, {
'attr1': 25
}, {
'attr1': None
}, {
'attr1': 5
}, {
'attr1': None
}, {
'attr1': None
}]
wrapper = pygrametl.ConnectionWrapper(connection=null_conn)
dim1 = Dimension(
name='dim1',
key='key1',
attributes=['attr1'],
)
for row in data:
dim1.insert(row)
dim_rep = DimRepresentation(dim1, null_conn)
notnull_tester = ColumnNotNullPredicate('dim1')
null_rep = DWRepresentation([dim_rep], null_conn)
print(notnull_tester.run(null_rep))
null_conn.close()
AGE INT NOT NULL,
ADDRESS CHAR(50),
VERSION INT);''')
company_info = [('Anders', 43, 'Denmark', 1.0), ('Charles', 50, 'Texas', 1),
('Wolf', 28, 'Sweden', 1), ('Hannibal', 45, 'America', 1),
('Anders', 43, 'Denmark', 2.0)]
# ... and inserting the necessary data.
c.executemany(
"INSERT INTO COMPANY (NAME,AGE,ADDRESS,VERSION) VALUES (?,?,?,?)",
company_info)
conn.commit()
ConnectionWrapper(conn)
s = SlowlyChangingDimension('COMPANY', 'ID',
['NAME', 'AGE', 'ADDRESS', 'VERSION'],
['NAME', 'AGE'], 'VERSION')
dim = SCDType2DimRepresentation(s, conn)
dw = DWRepresentation([dim], conn)
a = SCDVersionPredicate('COMPANY', {'NAME': 'Anders', 'AGE': 43}, 2)
b = SCDVersionPredicate('COMPANY', {'NAME': 'Anders', 'AGE': 43}, 3)
print(a.run(dw))
print(b.run(dw))
conn.close()
c.executemany("INSERT INTO FACTTABLE (BOOK,ISSUE) VALUES (?,?)", company_info)
conn.commit()
ConnectionWrapper(conn)
dim = Dimension('COMPANY', 'ID', ['NAME', 'AGE', 'ADDRESS', 'SALARY'], ['NAME'])
dim_rep = DimRepresentation(dim, conn)
dim2 = Dimension('LASTNAME', 'NAME', ['LAST'])
dim_rep2 = DimRepresentation(dim2, conn)
ft = FactTable('FACTTABLE', ['Book'], ['Issue'])
ft_rep = FTRepresentation(ft, conn)
dw = DWRepresentation([dim_rep, ft_rep, dim_rep2], conn)
expected_list1 = [
{'NAME': 'Anders', 'AGE': 43, 'SALARY': 21000.0, 'ADDRESS': 'Denmark',
'ID': 1},
{'NAME': 'Charles', 'AGE': 50, 'SALARY': 25000.0, 'ADDRESS': 'Texas',
'ID': 2},
{'NAME': 'Wolf', 'AGE': 28, 'SALARY': 19000.0, 'ADDRESS': 'Sweden',
'ID': 3},
{'NAME': 'Hannibal', 'AGE': 45, 'SALARY': 65000.0, 'ADDRESS': 'America',
'ID': 4},
{'NAME': 'Buggy', 'AGE': 67, 'SALARY': 2000.0, 'ADDRESS': 'America',
'ID': 5}
]
expected_list2 = expected_list1.copy()
SALARY REAL);''')
company_info = [('Anders', 43, 'Denmark', 21000.00),
('Charles', 50, 'Texas', 25000.00),
('Wolf', 28, 'Sweden', 19000.00),
('Hannibal', 45, 'America', 65000.00),
('Hannibal', 45, 'America', 65000.00),
('Buggy', 67, 'America', 2000), ('Buggy', 67, 'America', 2000),
('Buggy', 67, 'America', 2000), ('Buggy', 67, 'America', 2000),
('Buggy', 67, 'America', 2000), ('Buggy', 67, 'America', 2000)]
# ... and inserting the necessary data.
c.executemany("INSERT INTO COMPANY (NAME,AGE,ADDRESS,SALARY) VALUES (?,?,?,?)",
company_info)
conn.commit()
columns = ['NAME', 'AGE', 'ADDRESS', 'SALARY']
wrapper = ConnectionWrapper(conn)
dim = Dimension(name='COMPANY',
key='ID',
attributes=['NAME', 'AGE', 'ADDRESS', 'SALARY'],
lookupatts=['NAME'])
dim_rep = DimRepresentation(dim, conn)
dw = DWRepresentation([dim_rep], conn)
dup_predicate = DuplicatePredicate('company', ['ID'], True)
print(dup_predicate.run(dw))
conn.close()