From lechner  Tue Oct 15 03:06:53 1996
Subject: Re: chgen on a PC - excellent! (what version/path to source did you use?)
To: Owen_McGonagle_at_notes@smtp.cspi.com
Date: Tue, 15 Oct 1996 03:05:36 -0400 (EDT)
RJLRef: ~/96f522/chgen2PC_owenMcG.hlp

Please tell the path of your starting point.
I presume you could also compile the pr_util content produced by chgen?
CAn you identify the (.c or .h) files containing your fixes?
Or, better, upload your version  so we can do diff on it. 

> 
> 
>   Hello Prof. Lechner,
> 
>   Just for your information. I got chgen running on a PC using the Borland
>   C/C++ . I am not yet a windows
>   programmer -- the version I have running is a 16-bit version. I had to
>   make minor changes to the code
>   to get it to build and execute properly (the major problem was the 64k
>   data section size restriction):
> 
>        1. A macro incorrectly commenting a variable as "var" instead of
>   #var (the preprocessor needs to do its       thing).
>        2. A procedure returning a type "int" instead of "struct *" (the
>   default 16-bit return int was not getting
>            converted properly).
>        3. To get it to fit into the 64k data space, the Borland compiler
>   luckily includes a compiler switch that
>                       stores string literals into text space. Otherwise,
>   the fprintf(fp,".................") type calls put the literals in data
>            space and immediately cause chgen to exceed 64k.
>        4. The Borland compiler is ANSI and I could not figure out how to
>   keep it from complaining about
>             "old sytle" procedure call syntax. Therefore, I had to convert
>   a few routines to get the
>            compiler to stop complaining.
> 
>   I successfully built and ran the student/roster state model example.
> 
>   I plan on moving on the timer and JPsim state model examples.
> 
>   Maybe another student in the class would like to get it up and running on
>   a PC?
>