Martin Persson wrote:
>
> the make step doesnt put the
> binaries in the proper subdirectories (that is
> assuming for instance that the example executables
> should be in the examples subdirectory)
That is the way it worked at one time, but hasn't been that way since
2.2.0. Whether this is "proper" or not isn't important. It's just a
side effect of a change in the build system which has more benefits than
problems, so it stays.
> No separate debug and release versions of the .a libraries are
> built.
How many projects that have a Unixy build system can you name that do do
this? I can't think of any. Even if there are some, they're certainly
quite rare, which makes me wonder how you have come by the expectation
that this should be automatic.
MySQL++'s build system does allow you to _ask_ it for a separate build
of either a debug or release version. See http://bakefile.sf.net/ and
mysql++.bkl in the MySQL++ distribution.
But before you chase that, you should realize that the idea of separate
debug and release versions of a binary is a Windowsism. Unixy programs
(like MySQL++) generally don't require that. They usually get built
with debugging information by default, which you can then strip out
later to create a release version.
> Running the final install.bat script ...
I've half a mind to just remove that file.... It's not very useful, and
as you've seen, it usually doesn't get updated to track build system
changes.
> C:\mysql++>resetdb [localhost [dba [dba[]]]]
Sigh...
Brackets denote optional parameters. Nesting denotes position-sensitive
parameters. Thus, "[host [user [pass]]]" means that all three are
optional, and if you want to give "user", you must give "host" first.
This is *very* common notation. The fact that you don't recognize it
means you need to do a lot of catching up.
