Dear list members, I'm trying to extract the results of a query that returns fairly large amounts of data (100-200Mb) as quickly as possible using the c/c++ api.
Is there any way to avoid unnecessarily duplicating the data coming from the following functions? const unsigned char *sqlite3_column_text(sqlite3_stmt*, int iCol); const void *sqlite3_column_blob(sqlite3_stmt*, int iCol); I need access to the results of these functions beyond subsequent calls to sqlite3_step(). Or is this duplication unavoidable? The naive half of my brain wants my application to take responsibility for the memory that the text or blob occupies. But I'm not optimistic this is either a) possible or b) worthwhile because the marginally less naive half of my brain presumes that the results of these calls are pointers to memory within some highly optimised caching/paging system that can't hand over responsibility for sub-chunks of memory willy-nilly, and were there even to be some way I could take ownership of the memory via the sqlite api, it would just shift responsibility for allocating new chunks of memory for further row results away from my code and into sqlite.c so the gain would be limited to avoiding an unnecessary copy step from sqlite's memory into mine. But even that's likely to give a worthwhile speed-up, if it's possible. Many thanks, Nathaniel _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
