I am using Apache MetaModel to get the schema information. There is one use case, where I need to create CsvDataContext object for csv file with no header. I have column names in a separate data structure (List<String> colNames).
The context object gives column names as "A", "B", "C", etc. I guess metamodel assigns some default column names to the tables with no headers.
Is there any way to modify the schema which is held by the CsvDataContext object?
I believe UpdateableDataContext should work, but the documentation doesn't expose any method that allows modifying the metadata like column name.
How is it possible to achieve this scenario?
When you create your CsvDataContext, you specify a CsvConfiguration. One of the options in the CsvConfiguration is to provide a ColumnNamingStrategy. The default strategy is in deed to use alphabetic characters, A, B, C etc. But you can use a custom naming strategy, like this:
ColumnNamingStrategy columnNamingStrategy =
ColumnNamingStrategies.customNames("id", "foo", "bar", "baz");
CsvConfiguration configuration = new CsvConfiguration(
0, columnNamingStrategy, "UTF-8", ',', '"', '\\', true, false);
return new CsvDataContext(file, configuration);