Beginning the porting work (Well, began...)


Got to the point of actually compiling the first project last night. I chose SimpleParse because it's a fairly straightforward extension but it's not stupid-simple like the various accellerator extensions. I obviously forgot to check in the changes from the last time I ported it to Win32 :( . I fixed one or two glaring errors in the setup.py script, but it's still failing on LINK errors telling me I have to define an entry point. Will have to experiment with other projects to see whether this is a problem with my setup or the project.

Today is a billable-work day (late as I am getting to it due to being up wrestling with C last night), so probably not going to get anything more done today. I did get the DVD burned off with backups of the full downloads for all of the Toolkit components, so at least in the future there won't be a 4+ hour delay in the middle of the process.

Oh, while I was waiting for the compiles I realised something: instead of having binary extensions for Numeric and numarray I should hack a static version similar to the dynamic version. That would eliminate even the optional extension modules while still having full data-type support. Eventually we will have C accellerator modules to compile, the idea here is that those will always have Python backups for pure-Python operation without losing functionality (though we will lose speed), to make that happen we want all functionality to be available with the pure-python version.

Comments

Comments are closed.

Pingbacks

Pingbacks are closed.