def update(self, area_name: str, update_values: Dict[rules.ColumnNames,
str]):
"""
Change a value in the database.
"""
# Get index for area_name
index_for_area = self.columns[rules.ColumnNames.AREA.value].index(
area_name)
# Iterate over keys in update_values dict
for key in update_values:
# Get the current values
column_values = self.columns[key.value]
# Update value at index
column_values[index_for_area] = update_values[key]
# Make copy of instance dataframe
database = self.database.copy()
# Update column in dataframe with updated value
database[key.value] = column_values
# Validate
database = rules.database_schema().validate(database)
# Set new database
self.database = database
# Reset columns cached property
self._columns = None
def read_database_csv(path: Path) -> pd.DataFrame:
"""
Read database csv.
"""
csv = pd.read_csv(path, index_col=0, sep=rules.DATABASE_CSV_SEP, dtype=str)
assert isinstance(csv, pd.DataFrame)
csv = rules.database_schema().validate(csv)
return csv
def test_read_database_csv(database_csv):
"""
Test read_database_csv.
"""
path = database_csv
database_csv = repo.read_database_csv(path=path)
validated = rules.database_schema().validate(database_csv)
assert isinstance(validated, pd.DataFrame)
def __post_init__(self):
"""
Post initialization steps.
"""
assert self.tracerepository_path.exists()
self.database = rules.database_schema().validate(self.database)
self.unorganized_folder = (self.tracerepository_path /
rules.PathNames.UNORGANIZED.value)
def write_database_csv(path: Path, database: pd.DataFrame):
"""
Write database.csv to disk.
"""
database = rules.database_schema().validate(database)
database.to_csv(
path_or_buf=path,
sep=rules.DATABASE_CSV_SEP,
index=True,
index_label=rules.ColumnNames.AREA.value,
)
def scaffold(tracerepository_path: Path) -> pd.DataFrame:
"""
Make scaffold start for a repo.
"""
unorganized_path = tracerepository_path / rules.PathNames.UNORGANIZED.value
data_path = tracerepository_path / rules.PathNames.DATA.value
# Make unorganized folder
unorganized_path.mkdir(exist_ok=True)
data_path.mkdir(exist_ok=True)
# # Make other default folders
# for path in rules.folder_structure():
# path.mkdir(exist_ok=True, parents=True)
# Create dataframe for relations between datasets
df = scaffold_database()
assert df is not None
# Validate dataframe with pandera
df = rules.database_schema().validate(df)
return df