I'm Sure There's A Better Way

[Tim Daneliuk]

Bengt Richter wrote:
>

Thanks to all who answered!  A fertitle bunch of minds 'round here....

The response that gets it all (despite my fuzzy definition) was mailed
to me privately.  I deconstructed it (well really, reinvented it from 
scratch myself) to make sure I understood what was going on. 

And the winner is...

^-?((\d+(\.\d{2})?)|(\.\d{2}))$

To be really fussy, the \. should really be a string constant which
is conatenated into the remainder substrings so that other delimiters
could be used when I18Ning the code.
 
Just for the record - I knew when I started that a re was a better way to do
this.  Although I've programmed a lot in other languages, I am still in
the process of absorbing python, and I had not yet attacked its re
capabilities so I decided to hand code it.  Now that I look at the re
implementation, I realize I should have started there because it is so
much more compact and elegant.

In any case, it was a nice tutorial exercise which made me learn an
essential piece of python.

I'm writing a complete (small) application which does keyboard I/O,
saves state to disk, does meaningful arithmetic and so on.  My intention
is to put it up on the web when done so others can use it to see something
a bit less trivial than the examples which are common in tutorials.

Once again, many thanks to all who kindly took the time to respond.

------------------------------------------------------------------------------
Tim Daneliuk
tundra at tundraware.com