On Sat, 5 Jul 2014 20:08:02 +0000 David Holland <dholland-d...@netbsd.org> wrote:
> Another possibility is to write a PDF driver for the groff we have. > This is still work, but possibly not that much (e.g. maybe one could > rewrite the upstream gplv3 Perl groff driver in Lua...); Translation from Perl to C++ is straightforward. A couple of years ago, as an experiment, I re-coded the examples from the Camel book in C++. When C++ is supplied with appropriate libraries (simple things that I wrote myself) the converted linecount is roughly 1:1. That would be a 3200-line gropdf written in C++. (The only exception was Perl's eval, decidedly non-trivial in C++!) I participate in the groff list and admire what Deri James hath wrought. If there is interest here, I could broach the subject of whether a C++ version would be acceptable upstream. I'm not volunteering, but I do think it would make a nice SoC project. > on the other hand the artefact it produces is of limited long-term > value. And it's not like groff is doing a super job of typesetting > the articles... Which is the worst one, in your opinion? I guess I haven't read every one, because I don't remember any typesetting problems. You mention semantic markup more than once, as though it's valuable. What do you hope to do with it? As far as I'm concerned it's a lost cause *and* unimportant. The promise of "write one, render anywhere" remains, after two decades, unfulfilled. Once you acknowledge that, the use of -ms seems downright precient. > and there's long been a desire to kick groff out of base. I don't understand the desire to kick groff out of base unless it's to be replaced by heirloom troff. I realize troff qua troff is not fashionable anymore but, then, neither is NetBSD, er, anymore. I'm not interested in a markup-language tug of war. I declare flatly that no equivalent to groff exists. For one, show me the equivalent of pic. For two, show me a graphing package with a built-in typesetter. By the way, groff recently added an implementation of the IDEAL language, an intriguing improvement on its successors, bar none. --jkl
