Just install the package and run csvmanip <options> <csv_file_input> where the options configure the steps below.
The only limitation is that your input CSV file must only contain , as delimiters (no commas inside columns escaped or anything).
Parses your CSV file.
--ignore_headers N, ignores N lines at the beginning (typically the headers and padding of input file)--limit N, only parse the first N lines of the input
Allows you to plug modifiers, to make the input look prettier. Right now only two modifiers exist:
split1spacetransforms a column in 2 columns by splitting at the first name (e.g. use it for "Name Surname")- 'camelcase` transforms a string BLA in Bla
Use as many --modifier colnum:function as you wish. Be careful if you transform a column into more now columns, each modifier is run sequentially therefore the colnum should be changed accordingly.
Deduplicate records by using one or more column as unique key. Use --dedupe with a number or comma-separated list of numbers.
If you want to ignore records, pass in one or multiple --ignore colnum:regex expressions. Self-explanatory.
By default it will output the current set of records, if you want to change the columns, use --remapping and pass in a list of integers or strings:
- an integer represents a column in the current set of records
- a string will be replace as is at that position in the new records
You can also use --output-headers if you want to add a custom description in the first line of all the new records (comma-separated list of strings)
Eventually sort the records by using --sort colnum where colnum represents the column index in the NEW records to be output.