By default, the output data can be returned as a Pandas Dataframe for further manipulation.
df = wrangles.recipe.run('recipe.wrgl.yml')
It is possible to define the returned dataframe within the write section. If omitted, all columns will be included.
# Return a Dataframe with specified columns
write:
- dataframe:
columns:
- ID Number
- Product Properties
- Product Name in French
A variety of connectors are available to write the Wrangled data out to different targets such as files or databases.
Multiple write targets can be specified.
write:
# To an Excel file
- file:
name: file.xlsx
sheet_name: sheet_name
columns:
- ID Number
- Product Properties
- Product Name in French
# To a Microsoft SQL Server
- mssql:
host: sql.domain
database: database
table: table
user: user
password: password
It is possible to include a where condition within any write to filter the data. This uses SQL syntax to specify the condition for the rows.
# Save only the rows where the value
# in my_column is greater than 10
write:
- file:
name: output.xlsx
where: my_column > 10