Re: Python

[pasty camel <pasty_camel yahoo com au>]

> I suspect a misunderstanding here.  Adding global (non workbook
> specific) _functions_ is possible.  What I'll bet you are looking
> for are _procedures_ (aka macros).  What we support is adding new
> evaluation functions to expressions.  For 1.3 we're planning to work
> on our scripting interface and support macros properly.
> _______________________________________________
> gnumeric-list mailing list
> gnumeric-list gnome org
> http://mail.gnome.org/mailman/listinfo/gnumeric-list

Jody,

Yeah, I think you're right.  Macros are probably the tool that would be
most useful for the sort of workbook specific _procedure_ I'm looking
for.

You suggest that there are plans to get the scripting interface (read
macro interface?) working properly for version 1.3.  Does this suggest
there is some form of support for scripting in the current version
(1.2.1)?

With regard to your question regarding why workbook local (or indeed
sheet local) functions are desirable, I'll first stand at the head of
the class and say "I suffer from a Microsoft Excel dependency" (kind of
feels like an AA meeting).  Excel is the only spreadsheet application
I've looked at in any depth, and my views on what a spreadsheet should
be capable of will be coloured by that.

To the question:
Similar to the effective stiffness of a cracked concrete section, there
methods of determining the bending strength of steel sections which are
useful for various beam configurations, as well as designing columns. 
Because Excel allows for workbook specific macros, I've written a number
of workbooks each of which deal with a different material, timber,
concrete, steel, etc.  Each workbook has it's own macros for dealing
with each of the problems specific to each topic.

I don't intend this to be read as "Make it like Excel, or I wont use it"
'cause that isn't what I mean.  I'm more that happy to work with a
system that's different to what I know, especially if the system is an
improvement.  

Your comment regarding named expression is full of promise, and is worth
exploring.  If the expressions can allow sufficient complexity, many
macros could be dropped altogether.

Frank