Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!samsung!uunet!tut.cis.ohio-state.edu!uc!cs.umn.edu!aps1.spa.umn.edu!zoo From: zoo@grumpy.spa.umn.edu (david d [zoo] zuhn) Newsgroups: comp.unix.admin Subject: Re: Ditroff Message-ID: Date: 4 Sep 90 00:09:45 GMT References: <10266@hubcap.clemson.edu> <6486@tekgen.BV.TEK.COM> <1990Sep1.143205.17957@mccc.uucp> <6738@tekcrl.LABS.TEK.COM> Sender: iacovou@cs.umn.edu (Danny Iacovou) Distribution: usa Organization: Minnesota Automated Plate Scanner Project Lines: 30 In-Reply-To: markh@squirrel.LABS.TEK.COM's message of 3 Sep 90 21:48:28 GMT Mark == markh@squirrel.LABS.TEK.COM Someone> Isn't groff written in C++? Mark> It sure is. In fact, it is also dependent upon libg++. I expect that Mark> compilation by a C++ compiler other than g++ would be substantial work. Mark> G++ does work with SysV so you might have a chance. Groff does seem to Mark> work quite well and as a bonus includes an X11 previewer. It also has Mark> an option to produce dvi format (as well as postscript, of course) so Mark> it should be usable with all sorts of printers. Sorry. My libg++ is a minimal subset (gnulib3.o and xyzzy.o), which is the same as the libg++ used for development of groff (according to the docs). Everything worked just fine. This is from the groff 0.4 INSTALL file: Groff has been compiled on a Sun 4 under SunOS 4.0.3 with g++ 1.37.1 and with AT&T C++ 2.0, and on a 386 PC under 386/ix 2.0.1 with g++ 1.37.1 configured to use gas and GNU ld. You may encounter problems on other machines that I cannot anticipate. If you are using g++, you will need to install the header files from libg++. The only other parts of libg++ used by groff are contained in the files xyzzy.c and gnulib3.c; the libg++.a that I use contains only xyzzy.o and gnulib3.o. david d [zoo] zuhn Univ. of Minnesota Dept. of Astronomy zoo@aps1.spa.umn.edu Automated Plate Scanner Project