The stages we have just discussed are implemented in a Python 2 module written in C/C++. Generally speaking it was quite important to make the parsing very quick because it opens a possibility to draw a flowchart nearly as the user types the code – via catching pauses – and not introducing irritating hang ups. I believe that the goal was reached successfully. Here are some numbers for my not such a new laptop. Certainly the python installation files are of different size but nevertheless the average time of around 1 millisecond for a file is more than acceptable.