tocsv

Easing my life

Tool to automate the transformation of a really ugly file with lots of inconsistent data to CSV format.

NOTE: This a very specific tool to help me with a really trivial task. It's not in any way generic. However, I'm leaving this open source in case it's useful for anyone as a reference, maybe.

Installation

go get -u github.com/racsoraul/tocsv

Usage

tocsv <path to file>

It'll produce an output.csv file in the current directory.

Run tests

go test ./... -v -bench=. -benchmem

Format of the input file

Ideally, each line was expected to have the following data:

CORRELATIVE_NUMBER      NIT     DUI

NIT and DUI are documents of identification in my country (El Salvador). NIT is expected to have 14 numbers, and DUI is expected to have 9 numbers.

Reality

However, the file has lines with the following data:

• Empty line. No data at all
• Both DUI and NIT are NULL:

123     NULL        NULL

• The data is incomplete. The correlative number is the only data:

123

• The data is incomplete. Either NIT or DUI is missing, and the other one is NULL. Difficult to know for which that NULL stands for:

123     NULL

• The data is incomplete. It has the correlative number and NIT or DUI:

123     12345678901234

• The data is complete. All 3 values are there:

123     12345678901234      123456789

• The data is complete or incomplete, but NIT or DUI or both are wrong and do not have the right amount of digits:

123     123456789012      123456789
123           1234567

• The data is complete or incomplete, and has random extra spaces at the beggining, end or in the middle of the values:

      123        12345678901234      123456789   
      123               123456789   

• There are more than 3 values:

123        1234567890123 4      123456789   

Lines to be skipped

The lines where the data is wrong are skipped for the final CSV file:

• NIT, DUI or both exist but haven't the right amount of digits
• Empty line
• Line with more than 3 values