MySQL progress
I’m made progress with enabling the MySQL backend to Redland in Windows-land.
So far, I’ve managed to put together a reasonable test case that matches the documentation, compile the backend, register and dynamically load the mySQL library, comiled support into the Python bindings and figured out the command line interface to MySQL so I could create a test database.
Unfortunately, Redland crashes when I try to create a storage. After several false starts it looks like context->model is not being initialized properly for some reason. It gets partway through creating the storage before an sprintf call crashes hard trying to print something out (illegal memeory access, usually a sign of an uninitialized pointer). Most of the other context variables seems to be populated correctly, so it looks I’m going to finally decend into CVS to see if this is something that has been fixed (or if the code has changed enough that this is not an issue).
If not, there’s probably something going on specific to Windows that I’ll have to ferrett out. All that said, good progress has been made and I’m still optimistic that I can solve this before the weekend is over.
May 22nd, 2005 at 2:48 am
I’ve seen your vcproj files in redland-1.0.1. Your work is appreciated.
I’m interested in building redland for Windows, but I don’t have Visual Studio, so I’m using the free command-line tools described at http://www.vrplumber.com/programming/mstoolkit/. This means (I think) that the vcproj files are not directly usable by me.
So I think I’ll need NMAKE style makefiles in order to build. I’m thinking of writing a Python script to parse the vcproj file and generate a makefile. Does that sound like the right thing to do? Or is it possible for you to generate the NMAKE style files with Visual Studio and include these in the distro?
Again, thanks for your contributions.
May 22nd, 2005 at 9:57 am
It’s trivial for me to generate the file. I’ll include it in the next patch.