pilichoumao opened a new issue, #6404: URL: https://github.com/apache/rocketmq-dashboard/issues/6404
### Studio Version `rocketmq-studio@779b8c0e7aed2396d3cc72743dcd544ea48e9ae6`, built from source. ### Runtime Environment macOS, Node 22.19.0, npm 11.11.1; pinned web/package-lock.json. Reproduced with the actual shared parser and with both page components' File upload flow (jsdom, mocked import services). No broker/database is needed for the parser reproduction. ### Actual Behavior Topic and consumer-group CSV import already rejects more than 100 nonblank data records, but readCsvRows builds every parsed row, parseCsvTable maps every record, and only then is the limit checked. A file that can never be imported still allocates objects for its entire excess inventory. A small discriminating reproduction is `Name` followed by 101 valid names and an unterminated quoted tail. Both pages report the tail's quote error rather than stopping at the exceeded record limit. The same parser regression fails on unchanged production. ### Expected Behavior Stop when the first excess nonblank data record completes, before parsing the remaining CSV and mapping excess records. The header, empty lines and newlines inside quoted cells should not count as extra data records. Retain the current 100-record limit/message and all valid CSV behavior; an oversized upload must not call the import service, and the next valid upload should work. ### Scope / Limits This is the existing record-count policy, not a new file-byte or cell-size policy. File.text() still reads the selected file, and a huge individual cell remains outside this fix. No claim of eliminating all memory risk or of a memory leak. #6069 guards ownership of asynchronous file reads; #5574 handles validation i18n and does not add this early parser boundary. -- 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]
