The Coralogix team tackled a challenge in translating SQL-like expressions into OpenSearch DSL while maintaining backwards compatibility in their DataPrime query engine. The crux of the task involved handling SQL's null semantics, where expressions can result in TRUE, FALSE, or NULL, and adapting these to OpenSearch, which only recognizes TRUE or FALSE through document filtering. The team explored the logic behind SQL's three-valued logic, focusing on translating expressions using boolean operators like AND, OR, and NOT into OpenSearch's binary system by determining whether an expression is "false or null" or "true or null." This translation demanded creating an intermediate representation of the Elasticsearch DSL to optimize and simplify the conversion process, applying logical principles like De Morgan’s laws. Although OpenSearch DSL's capacity is limited for advanced operations, script queries were used for more complex tasks, highlighting the balance between performance and functionality in achieving accurate translations.