Should I write a book on Python Metaprogramming ?

[Bjorn Pettersen]

> From: Michael Hudson [mailto:mwh at python.net] 
> 
> mertz at gnosis.cx (David Mertz) writes:
> 
> > Here's the story though: For a number of months, I have had in mind
> > the idea of writing a book called _Metaprogramming in Python_.
> 
> Here's a convenient place to plant a small rant: I object somewhat to
> the notion of Metaprogramming being something special performed by
> wizards with pointy hats and keepers of secrets.
[...]

Completely agree (at least for Python :-), alhtough your background
biases you. If you've ever used a language that can treat programs as
data you're way ahead (or if you've written a
parser/evaluator/interpreter/??). To most C++ programmers, classes
exists when the program starts (period).

The real problem is that they're new, under-documented, probably missing
features and or have excessive limitations, and I'm guessing much less
than 100 examples -- all of which will be fixed with experience.
However, in essense, you're currently requiring a programmer that will
read the source and experiment to figure something out, and that isn't
put off by changes to the feature... (and if we were talking about
MSWord macro features here, they allready do what I need them for, so I
wouldn't be interested <wink>).

-- bjorn