[PLUG] Xconq GIS Help Wanted

[D. Cooper Stevenson]

On Wednesday 20 October 2004 12:43, Steve Bonds wrote:
> On Tue, 19 Oct 2004 22:59:31 -0700, AthlonRob wrote:
> > My first question for you is... what made you chose C?

I agree completely with you guys on this. If I had my way I would write this 
in Perl or shell script in a "New York Minute." 

Xconq is cross-platform so we have to write it in C. This ensures that Windows 
users will have the ability to import GIS data.

Let me stress, however, that I am absolutely in your corner on the script 
thing, If I had a choice I would absolutely use a scripting language. If you 
have some ideas on how I may get away with using Perl et al. while 
maintaining cross-platform compatibility, please let me know.


-Coop
> >
> > IMHO, C is really going a few steps too low when what you're really
> > doing is advanced text manipulation.  It seems to me a scripting
> > language such as Python (or perl) might be better suited to this task
> > (and less prone to sloppy code causing problems).
>
> I agree with Rob on this one.  Converting/resampling the GIS data
> using perl should be pretty easy.  Here is a model for most of my
> "translation" perl scripts:
>
> + figure out what the input looks like, put it in a comment
> + figure out what the output should look like, put it in a comment
> + read the whole input file into a complex hash of hashes data
> structure suitable for the input data
>  o In the Old Days this was a big no-no because of the memory
> required, nowadays memory is fairly plentiful so even those monster
> 20MB GIS files probably won't overload your PC
>  o If there's a perl module to parse the data input, use it
> + pull out the data you want from this monster hash and manipulate it
> as needed into a new data structure suitable for your output data
>  o Sometimes I'll omit this second data structure and just print it,
> if the output is line-oriented
> + iterate over the output structure to send the info to the right
> place (screen, CSV file, HTML, XML, database, etc.)
>  o If a perl module is available for the API I need, use it!
>
> For your project, this should be maybe 1 or 2 days' work for someone
> marginally familiar with perl, or maybe a week for a total newbie.
>
> The hard part is getting good data structures figured out.
>
>   -- Steve
>
> _______________________________________________
> PLUG mailing list
> PLUG at lists.pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug