Is your feature request related to a problem? Please describe.
When I import a CSV file in dataprep, the column types are automatically set by appropriate rules. I wish the datasource ingestion have the same behavior.
Describe the solution you'd like
Add the same feature. (hopefully reuse the data preparation's code)
Moreover, the role of the columns (dim/measure) might be set automatically too.
I suggest that the numeric columns to be measures by default.
IMPORTANT : Mis-guessing is not a big pain for dataprep because you can change the rule or add another rule to fix it, but for datasource ingestion, it could be troublesome, for example, if the data have lots of number-like codes. So I suggest a checkbox that controls the auto type conversion feature.
Describe alternatives you've considered
After implementing the metadata management feature (TBD), the types in the metadata should override the guess.
Additional context
Related code is in PrepDatasetFileService.getResponseMapFromCsv().
Is your feature request related to a problem? Please describe. When I import a CSV file in dataprep, the column types are automatically set by appropriate rules. I wish the datasource ingestion have the same behavior.
Describe the solution you'd like Add the same feature. (hopefully reuse the data preparation's code) Moreover, the role of the columns (dim/measure) might be set automatically too. I suggest that the numeric columns to be measures by default.
IMPORTANT : Mis-guessing is not a big pain for dataprep because you can change the rule or add another rule to fix it, but for datasource ingestion, it could be troublesome, for example, if the data have lots of number-like codes. So I suggest a checkbox that controls the auto type conversion feature.
Describe alternatives you've considered After implementing the metadata management feature (TBD), the types in the metadata should override the guess.
Additional context Related code is in PrepDatasetFileService.getResponseMapFromCsv().