fieldswhere

Fieldswhere is a lightweight, domain-specific language proposed for filtering structured data by field values. It is a conceptual tool used in discussions and mock implementations, intended to be portable across storage formats such as JSON documents, relational tables, and log records.

A fieldswhere expression specifies predicates on one or more fields and is typically evaluated by a data processing pipeline or search layer. A fieldswhere expression consists of one or more comparisons involving field identifiers and literals, combined with logical operators such as and, or, and not. Fields may be nested using dot notation or bracket notation to access subfields. Supported operators include equality, inequality, comparison, set membership, and substring checks. The language emphasizes portability, allowing expressions to be translated into underlying query engines or evaluated directly in memory.

Examples: status = 'active' and region in ['EU','APAC']; user.age >= 18 and (user.role = 'admin' or user.role = 'manager'); tags contains 'premium'.

Typical use cases include filtering datasets in ETL jobs, constraining API query parameters, and indexing or searching documents in document stores. Fieldswhere expressions are designed to be deterministic and side-effect free, enabling reproducible filtering results across environments.

Implementation notes: Real-world adoption varies by platform. The specification favors clear parsing rules, lexical tokens, and a stable operator set. Libraries may provide a parser, an evaluator, and helpers to translate expressions into native queries.