Hi, Otto, > It would be best if you submitted your patch directly upstream at > https://github.com/mariadb-corporation/mariadb-connector-c/
Thank you; I have done that at https://jira.mariadb.org/browse/CONC-487 However, given the developer's response on https://jira.mariadb.org/browse/CONC-289 I don't know if they will accept the patch. > They have the best knowledge to assess if this change is OK or if it > has potential regressions. Well, the patch simply sets pointers to NULL after the memory that they point to has been freed. They already do that in one other place (the memset call in mysql_close sets mysql->options.extension to NULL) with a comment "/* Clear pointers for better safety */" so I hope that they (and Debian) will accept the patch. Regards, Dianne. better safety */" so I hope that they (and Debian) will accept the patch. Regards, Dianne.
