https://bz.apache.org/bugzilla/show_bug.cgi?id=62328
--- Comment #3 from [email protected] --- Just out of curiosity though - the idea behind the streaming api is to keep a fixed amount of rows in memory for random access, but get rid of anything that has already been flushed to disk, right? So where's the harm in allowing access to the first few rows of a template just read in, provided that the random access window is sufficiently big, that we read in from the template? It wouldn't be overwriting the same file cause we'd be saving that to a new one, and the streaming api creates it's own temp files anyways. -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
