Skip to content
PostgreSQL database dump obfuscator
Go
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.circleci
example
.gitignore
LICENSE
README.md
misty.go
misty_test.go

README.md

Misty

PostgreSQL database dump obfuscator

CircleCI Go Report Card License

This project was inspired by pgdump-obfuscator.

This project needs a plain-text formatted dump to work, if the dump is compressed you can create a plain-text version of it by using pg_restore:

pg_restore your_dump > dump.sql

To work with misty you need an io.Reader (eg: a dump file), an io.Writer (could be os.Stdout) and a list of misty.Target you want to obfuscate.

Altering column values

target := &misty.Target{
    TableName: "public.users",
    Columns: []*misty.TargetColumn{
        {
            Name:  "username",
            Value: func(_ []byte) []byte {
                return []byte("mistery_user")
            },
        },
    },
}

The target above will set username value in all the records of the public.users table to be "mistery_user". As you can see column values accept any function returning a slice of bytes, so you can get as fancy as you like with those obfuscators.

Using existing column value

In case you want to do something with the existing column value (reusing, condition something else, etc.), this is is passed in form of a []byte as the argument of your TargetColumn functions.

target := &misty.Target{
    TableName: "public.users",
    Columns: []*misty.TargetColumn{
        {
            Name:  "username",
            Value: func(oldVal []byte) []byte {
              if string(oldVal) == "rentziass" {
                return []("maintainer")
              }
              return []byte("mistery_user")
            },
        },
    },
}

This example will turn all the usernames to mistery_user, unless the real username was rentziass, in which case it'll be changed to maintainer.

Incremental values

If you need to have incremental values, to maintain uniqueness for example, you can achieve that creating counters that are then used in your functions:

usernameCounter := 0
target := &misty.Target{
    TableName: "public.users",
    Columns: []*misty.TargetColumn{
        {
            Name:  "username",
            Value: func(oldVal []byte) []byte {
              usernameCounter++
              return []byte(fmt.Sprintf("user_%v", usernameCounter))
            },
        },
    },
}

These functions are called once for every single record in the target table, so if three users were found in public.users, their usernames would be user_1, user_2 and user_3.

You can play around with the example, providing a plain-text dump named dump.sql.

Deleting rows

You can delete rows from your original dump by adding DeleteRules to your targets, like so:

target := &misty.Target{
    TableName: "public.users",
    DeleteRowRules: []*misty.DeleteRule{
        {
            ColumnName: "username",
            ShouldDelete: func(u []byte) bool {
              username := string(u)
              return string.HasPrefix(username, "admin")
            },
        },
    },
}

This will delete all the rows from public.users table where username begins with "admin". You can have multiple conditions for the same table.

If you want to delete all the rows from a table you can achieve that with something like

target := &misty.Target{
    TableName: "public.users",
    DeleteRowRules: []*misty.DeleteRule{
        {
            ColumnName: "id",
            ShouldDelete: func(_ []byte) bool {
              return true
            },
        },
    },
}

Just make sure the column name you're using actually exists or you'll get an error.

You can’t perform that action at this time.