On Wed, Jan 18, 2017 at 01:27:24PM -0600, Dirk Eddelbuettel wrote: > ... > | And it appears that the local sources are ephemeral, so I can't > | readily patch those sources to see if I can get it to work (as the > | patched sources wouldn't be used) -- unless I miss the mark.... > | > | Any suggestions for how I might proceed? > > I don't follow. Can you not use 'diff' and 'patch'? I am so old that I have > > tar xvzf foo_1.2.3.tar.gz > mv foo foo.orig > tar xvzf foo_1.2.3.tar.gz > cd foo > /* now do your work */ > cd .. && diff -ru foo.orig foo > foo.patch > > fairly deeply engrained in muscle memory. And R itself is in pretty tip-top > shape and will listen to 'configure; make; make install'. You probably can > even skip make install as you are "just" fighting a failure to build. > ....
OK; thanks (again!) for your hints. :-)
I was able puzzle through things enough to determine that re-invoking
install.packages() as
install.packages("~/R/Rtmp4VjFtY/dhw/openssl", repos = NULL)
(in my case) would allow me to work with my copy of the sources; some
experimentation^Whacking later, I determined that the patch:
--- src/ssl.c.orig 2016-12-30 13:26:04.000000000 -0800
+++ src/ssl.c 2017-01-18 16:31:02.301226000 -0800
@@ -2,6 +2,9 @@
#if !defined(_WIN32) && !defined(__sun) && !defined(_POSIX_C_SOURCE)
#define _POSIX_C_SOURCE 200112L
#endif
+#if defined(__FreeBSD__)
+#define __BSD_VISIBLE 1
+#endif
#include <Rinternals.h>
#include <stdlib.h>
appears to allow the build (& install) to complete successfully.
I have managed to report this to the openssl package author.
Thanks again!
Peace,
david
--
David H. Wolfskill r...@catwhisker.org
How could one possibly "respect" a misogynist, racist, bullying con-man??!?
See http://www.catwhisker.org/~david/publickey.gpg for my public key.
signature.asc
Description: PGP signature
______________________________________________ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
