tesera / bawlk Goto Github PK

Many standard CSVs pad white space between commas for readability; so needs to be trimmed before processing.

For some reason 0.00 is getting by the int test and it shouldn't.

For example [0-9]{1,2}/[0-9]{1,2}/[0-9]{4} will get parsed to [0-9]{1. As a workaround use expanded rule set [0-9][0-9]?/[0-9][0-9]?/[0-9][0-9][0-9]?[0-9]?

field_name is missing to correctly describe the field that raised an error. For example:

source_row_index,violating_table,violation_key,violation_type,violation_severity,violation_comment
2,sonde_data.csv,,required,error,date_in value is required but was empty,date_in

This causes the validation checker, which uses csv-parse to bail with incorrect column count. The column headers are appended here https://github.com/tesera/datapackage-validator-awslambda/blob/master/fkeychecks.sh#L35.

Bawlk is not adding field types for "create table" statement in the awk output.