Saturday, July 22, 2017

Log 072217

So, Thursday I wrote an introduction to the evaluation strategy I employ in the Egel interpreter. I moved that to the main website and started shaving, rewriting, and removing. Lastly, I updated the pictures a few times which took some work because Inkscape isn't very good at keeping a similar layout for different pictures. Now, it's Sunday and I am reasonably content with the result, though it might be somewhat informal to some people. But, ah well, this is the first time I had the opportunity to get half-baked 'feelings' right while loosely relating what I designed to other approaches so I guess I was entitled to wasting some time on it.

Meanwhile, I didn't code. I decided upon a C++ implementation for primitive input/output because reasons -callback hell, the convenience of Unicode support, minimal dependencies, C++ is reasonably portable, etc.- but that also implies a lot of work I can't seem to find the energy for at the moment.

I need to:
  • Implement channels in C++, an abstraction over several iostreams which is going to be pretty verbose in that it needs to support a lot of operations.
  • Write IO combinators employing that abstraction in a dynamically loadable module.
  • Test all of that. The channel abstraction, the combinators, and module loading.
Too. Much. Work.

No comments:

Post a Comment