Skip to content
This repository has been archived by the owner on Sep 26, 2023. It is now read-only.
/ anon Public archive

A UNIX Command To Anonymise Data

License

Notifications You must be signed in to change notification settings

intenthq/anon

Repository files navigation

Anon — A UNIX Command To Anonymise Data

Build Status Go Report Card License GitHub release

Anon is a tool for taking delimited files and anonymising or transforming columns until the output is useful for applications where sensitive information cannot be exposed.

Installation

Releases of Anon are available as pre-compiled static binaries on the corresponding GitHub release. Simply download the appropriate build for your machine and make sure it's in your PATH (or use it directly).

Usage

anon [--config <path to config file, default is ./config.json>]
     [--output <path to output to, default is STDOUT>]

Anon is designed to take input from STDIN and by default will output the anonymised file to STDOUT:

anon < some_file.csv > some_file_anonymised.csv

Configuration

In order to be useful, Anon needs to be told what you want to do to each column of the CSV. The config is defined as a JSON file (defaults to a file called config.json in the current directory):

{
  "csv": {
    "delimiter": ","
  },
  // Optionally define a number of rows to randomly sample down to.
  // To do it, it will hash (using FNV-1 32 bits) the column with the ID
  // in it and will mod the result by the value specified to decide if the
  // row is included or not -> include = hash(idColumn) % mod == 0
  "sampling": {
    // Number used to mod the hash of the id and determine if the row
    // has to be included in the sample or not
    "mod": 30000
    // Specify in which a column a unique ID exists on which the sampling can
    // be performed. Indices are 0 based, so this would sample on the first
    // column.
    "idColumn": 0
  },
  // An array of actions to take on each column - indices are 0 based, so index
  // 0 in this array corresponds to column 1, and so on.
  //
  // There must be an action for every column in the CSV.
  "actions": [
    {
      // The no-op, leaves the input unchanged.
      "name": "nothing"
    },
    {
      // Takes a UK format postcode (eg. W1W 8BE) and just keeps the outcode
      // (eg. W1W).
      "name": "outcode"
    },
    {
      // Hash (SHA1) the input.
      "name": "hash",
      // Optional salt that will be appened to the input.
      // If not defined, a random salt will be generated
      "salt": "salt"
    },
    {
      // Given a date, just keep the year.
      "name": "year",
      "dateConfig": {
        // Define the format of the input date here.
        "format": "YYYYmmmdd"
      }
    },
    {
      // Summarise a range of values.
      "name": "range",
      "rangeConfig": {
        "ranges": [
          // For example, this will take values between 0 and 100, and convert
          // them to the string "0-100".
          // You can use one of (gt, gte) and (lt, lte) but not both at the
          // same time.
          // You also need to define at least one of (gt, gte, lt, lte).
          {
            "gte": 0,
            "lt": 100,
            "output": "0-100"
          }
        ]
      }
    }
  ]
}

Contributing

Any contribution will be welcome, please refer to our contributing guidelines for more information.

License

This project is licensed under the MIT license.

The icon is by Pixel Perfect from Flaticon, and is licensed under a Creative Commons 3.0 BY license.