From: <bt...@vi...> - 2016-12-15 06:28:41
|
I just posted another question. An example of a Unicon program compiled by unicon -C converted to allow gcc to create a .o file that is linked into a COBOL program. COBOL calling Unicon and getting at some data on the way out. Works well, but currently fragile. I'd like opinions and advice on how/if unicon -C can be slightly tweaked to allow Unicon to generate code that other compilers and programming environments can use without fuss. (how/if, as in, if it should). https://sourceforge.net/p/unicon/discussion/contributions/thread/d15b4085/ (https://sourceforge.net/p/unicon/discussion/contributions/thread/d15b4085/) I get a sense that most people on the list don't bother visiting the SourceForge project space yet, but I'd really appreciate some feedback on this one. It could be big, (or it could fizzle and spin around on the floor to become dust), but I'm aiming for big. Unicon generating object files that can be used in C, C++, Fortran, COBOL, Vala, Python, Perl, any language that can link to C ABI object code. Two small changes to the output from unicon -C and some assumptions in a Makefile. Cheers, Brian |