[norm.olsen@autodesk.com: RE: [MetaCRS] Using CS-Map to retrieve WKT for EPSG CRS codes: Help?]

[Christopher Schmidt]

(Forwarding on to the list...)

I don't know if CS-Map specfic discussion is becoming overbearing for
the MetaCRS list, we can perhaps create a CS-Map list for discussion of
CS-Map specifically...

-- Chris

----- Forwarded message from Norm Olsen <norm.olsen at autodesk.com> -----

From: Norm Olsen <norm.olsen at autodesk.com>
To: Christopher Schmidt <crschmidt at metacarta.com>
Date: Wed, 17 Dec 2008 10:22:31 -0800
Subject: RE: [MetaCRS] Using CS-Map to retrieve WKT for EPSG CRS codes: Help?

Hello Christopher . . .

Thanks for the offer.  I have a Vista machine on order, and all new machines are of the 64 bit variety.  It'll have a huge disk drive with four partitions.  One of those partitions will, hopefully, become a Linux 64 installation; assuming I can figure out how to do that and keep Vista running in another partition.  Let's wait and see if I can achieve that.

It has always been an objective of the library to be usable on any machine which supports a C compiler.  C++ usage crept in with the MFC and WKT stuff and is getting more prevalent.  A quick look at the first couple hundred lines of the cs_map.h file will show the original intent to resolve all compiler differences in one source module.  (You'll also see a lot of historical baggage there as well.)  The hardware/software markets overcame my ability to keep up long ago.

I think it will be up to the community to extend the scope of this ability to compile and run anywhere.  I believe there will be strong support from the project committee for such efforts.

Norm

-----Original Message-----
From: Christopher Schmidt [mailto:crschmidt at metacarta.com]
Sent: Wednesday, December 17, 2008 11:09 AM
To: Norm Olsen
Cc: metacrs at lists.osgeo.org
Subject: Re: [MetaCRS] Using CS-Map to retrieve WKT for EPSG CRS codes: Help?

On Wed, Dec 17, 2008 at 09:53:23AM -0800, Norm Olsen wrote:
> Hello Christopher . . .
>
> This information helps a lot.  CS-MAP's origin goes back to the days
> when file names and function names had to be 8 characters or less and
> there were such things as near, far, and huge pointers.  So there is a
> lot of historical baggage in the library.  The long32_t and ulong32_t
> were invented to handle the problem of a 64 bit "long" on some
> machines/compilers.  It's good to know that Linux 64 is one of them.
> Without having a machine to test on, its easy for these errors to
> creep in.  Some day soon I will have access to a 64 bit machine and
> hope to be able to put Linux on it.

Norm,

Offer to set you up with a Linux shell account on an OSGeo server
running 64-bit Linux stands, if you're interested :)

Is compiling on less common platforms, generally speaking, a goal of the
library? I don't want to flood you with bug reports about lack of
compilation on Mac if the answer is just "we have no intention of
supporting macs". Thus far, it seems that the changes to support
different machines I've tried are relatively minor, so I don't forsee
this being a major issue. (It's easy for things to atrophy when you
don't have access to a particular type of machine/setup!)

Regards,
--
Christopher Schmidt
MetaCarta

----- End forwarded message -----

-- 
Christopher Schmidt
MetaCarta