From: Warren Young Date: May 13 2010 8:45pm Subject: v3.1.0 imminent List-Archive: http://lists.mysql.com/plusplus/8942 Message-Id: <4BEC64FE.3070400@etr-usa.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Since it's taking too long to get around to finishing SSQLS v2, I've decided I'm going to put a cap on v3.1.0 soon. I want to squeeze a few more features into 3.1 before releasing it: o Can we fix "if (!query)" without changing the ABI? http://lists.mysql.com/plusplus/8594 o Fix pedantic warning: lib/null.h:55: warning: type qualifiers ignored on function return type o Can we bring examples/vstudio into the top-level build system? Would be nice to get MYSQL_WIN_DIR support, and to make it link against the built MySQL++ instead of requiring it to be installed first. o Add -lmysqlclient for Xcode case o Xcode project needs to ignore config.h when not doing an autotools build. Put conditions in lib/common.h. o Result::field_num() should be case-insensitive; column names aren't case-sensitive in SQL. MySQL++ is currently trying to do this by smashing field name case to lowercase in FieldNames, calling str_to_lwr() on all received values. This doesn't help with SSQLS because it's not smashing field name case we get as macro arguments in the SSQLS definition. Probably better all around to remove the case smashing and do case-insensitive comparisons instead. o Raise tquery and SSQLS field limits in RPM build process, since the very fact that you're building an RPM means we know the limits of the minimum supported platform. It needs to be smart enough to not override a value given when one hand-configures a source tree, possibly by parsing it out of ssqls.h. Otherwise, use the known floor value. o Check that it works with Connector/C and MySQL 5.1 o Document sql_*_null in userman o Add to datetime.h: inline DateTime NOW() { return DateTime(); } MySQL++ v3.2.0 will pretty much be dedicated to finishing SSQLS v2, then we're back to more general feature releases after that. MySQL++ is maturing, so it's no surprise the release cycle is slowing down, but 15 months between releases is too long. Comments? Complaints? Kudos?