new project release , which provides an addition to the PostgreSQL DBMS that solves the problem of hiding or replacing confidential data or information that is a trade secret. Data can be hidden on the fly based on specially defined rules and lists of users whose responses to queries need to be anonymized. Code licensed under PostgreSQL.
For example, using the add-on in question, you can provide access to the database to third parties, for example, third-party business intelligence services, automatically cutting out data for them such as phone numbers and credit cards, or using more sophisticated methods, such as replacing customer and company names with fictitious information. In addition to performing anonymization with a direct connection to the DBMS, a mode for creating anonymized SQL dumps is provided (the pg_dump_anon utility is proposed).
PostgreSQL Anonymizer PostgreSQL DDL (Data Definition Language) and allows you to set an anonymization strategy at the schema level that defines the table structure. A large set of functions is provided for processing data to be replaced: randomization, replacement with dummy values, partial , shuffling, noise, etc. The new version adds functions for identifying identifiers, as well as a pseudonymization mode that allows you to generate realistic fictitious values ββ\uXNUMXb\uXNUMXblinked to the source data.
Additionally, we can note the open platform developed by Microsoft for detecting leaks of confidential information. . The platform allows you to identify or remove information in documents, text and images that contains personal and confidential data, such as full name, phone numbers, email, credit card numbers, crypto wallets, addresses, passport numbers, financial data, etc. Processing of various storages (from Amazon S3 to PostgreSQL) and formats is supported. The code is written in Go (there is a variant in Python) and under the MIT license.
Source: opennet.ru