Created attachment 148512 [details] math_ump.diff
hmm, I have same question about lines 32-34 on this PR. Why was this changed? A) it doesn't do anything different B) it arguably changes the style from "desireable" to "less desirable" In other words, avoiding a compound command joined with "&&" and isolated with parenthesis is desired when possible (assuming it doesn't make the line really long). So the change seems unnecessary and not really wanted.
ping
I'm tired of looking at this. Let's see if I can work with it.
A commit references this bug: Author: marino Date: Fri Feb 6 18:25:06 UTC 2015 New revision: 378549 URL: https://svnweb.freebsd.org/changeset/ports/378549 Log: math/ump: Finish original "fix on current" Back on 10 OCT 2014 when the PR was opened, this unmaintained port didn't build on current. The PR stagnanted due to an unanswered question and the port was eventually fixed by amdmi3 on 21 NOV. Let's go ahead and make the remaining changes which include adding a license (GPLv2), defining desktop entries, tweaking pkg-descr, and make real_frac function match its changed header (for clang). PR: 194497 Submitted by: Ports Fury Changes: head/math/ump/Makefile head/math/ump/files/patch-Makefile head/math/ump/files/patch-src__real.cpp head/math/ump/pkg-descr