Query : Usage of Sylvan Libraries about sylvan HOT 2 CLOSED

1. I'm assuming your file has a header row? Otherwise, you wouldn't be able to map my name alone as the columns don't have names. You can use the Schema type to map the named columns to their own name: Schema.Parse("Name>Name,LastName>LastName,BirthDay>BirthDay:DateTime"). This ">" syntax maps the "base column name" (the name in the file) to the name that it will be exposed as through the CsvDataReader. You could also implement your own mapping mechanism by implementing your own CsvSchemaProvider.

2. Using the above mapping technique, you can check via GetOrdinal which will throw an exception if the name doesn't exist. Or, you can use the GetColumnSchema method to get the exposed schema and verify that all the expected columns are present.

3. I would expect that to "just work", so maybe I don't understand what you're asking.

4. You can use the Validate or ValidateSchema extension methods (in Sylvan.Data) which allow you to handle data validations. These methods allow you to provide a validation method, which will be called for all rows (Validate), or only called for rows that don't match the expected schema (ValidateSchema). ValidateSchema is probably what you want; your validation method will only be called when the date value can't be parsed correctly. In your validation method you can try to "repair" the data. The DataValidationContext parameter has methods to determine what columns have errors (GetErrors/GetException), and methods to allow you to provide a value (SetValue). You can return true if you could repair the data and want it to be yielded from the reader, or false to skip the row altogether, and maybe log the failed row somewhere.

Hopefully that helps, but I'm not sure I was understanding your issues so let me know if I did.

from sylvan.

I'll assume, at this point, that you got things working...

from sylvan.