linliu-code opened a new issue, #20477: URL: https://github.com/apache/hudi/issues/20477
Hudi tables often hold free-form text (logs, support conversations, product descriptions, audit events), but no Hudi index can tell an engine which files contain a given word. A query such as "rows whose `message` holds both `timeout` and `kubernetes`" scans every file of the table. RFC-110 (claimed in #19614) proposes a full-text index stored in the metadata table: per base file, a coverage marker plus token postings with row positions, used to skip files and rows for token predicates without changing query results. This issue tracks the RFC design document. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
