Re: [groff] Loss of MSVC support

[Eli Zaretskii]

> Date: Thu, 14 Feb 2019 12:50:09 +0000
> From: Colin Watson <address@hidden>
> 
> On Wed, Feb 13, 2019 at 06:58:18PM +0200, Eli Zaretskii wrote:
> > Just to make what Keith says (and I concur) crystal clear: there's a
> > need to distinguish between C99 compliance of the compiler and the
> > C99/Posix compliance of the C runtime.  We can assume the former,
> > certainly when using MinGW GCC, but we cannot assume the latter when
> > building a native MS-Windows port (as opposed to Cygwin port) of
> > Groff.
> 
> If it's just the runtime, then Gnulib should be able to paper over a
> pretty fair number of the differences, and groff already uses that.

Up to a degree.  There's no fork for Windows, for example, and many
other functions are missing.

> (It's possible that some of the _WIN32 conditionals can be supplied by
> Gnulib these days, but there's also no great urgency to remove them,
> IMO.)

I doubt that.  The vast majority of those I see in the current sources
deal with issues that Gnulib cannot fix:

  . absence of SIGPIPE
  . backslashes as directory separators
  . differences in how argv[0] is presented to 'main'
  . different names of environment variables, like TEMP vs TMPDIR
  . quoting of command arguments 'like this' that isn't supported on
    Windows