This course covers a selection of Python and programming topics aimed at researchers in mathematics. As such it assumes a fairly high willingness to self-teach and read documentation, and the topics it covers reflect as much my interests as an attempt to cover 'the important stuff'.
A few things are missing - the tex file for the first object lesson vanished
and I think I must have rm'ed it by accident at some point. Also, for the
scripting session you should download fresh wikipedia page dump files and use
GNU shuf (or something) to retrieve some random subset. I would not suggest
using the wiki page dumps without vetting them since the page titles can be
highly inappropriate...
I have realised a couple of mistakes crept in during writing this. Some are small and inconsequential, a few are important and worth correcting. I will update this as they come to mind.
__repr__()is not supposed to be a printable form of an object for a user to see. It is better to use__str__()for that purpose, and in particularprintlooks for__str__()first.
__repr__() should be used to give a printable description of how the object
is made. Standard library types have the property that the return of
__repr__() should be runnable Python code that would recreate the same object
- i.e.
eval(repr(x)) == xshould be true. This is a guiding principle and its as always up to you how you treat it.