Date: Wed, 13 Jul 2022 18:50:20 +0000 From: bugzilla-noreply@freebsd.org To: chromium@FreeBSD.org Subject: [Bug 265180] www/chromium: actually use the system-installed ICU Message-ID: <bug-265180-28929-AChpmfKs6I@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-265180-28929@https.bugs.freebsd.org/bugzilla/> References: <bug-265180-28929@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D265180 --- Comment #2 from Mikhail T. <mi@ALDAN.algebra.com> --- (In reply to Robert Nagy from comment #1) > built-in ICU has several modifications If these modifications are good and useful, Google can/should submit them to IBM for inclusion into ICU releases. If IBM are too slow to process the contribution, FreeBSD can incorporate the changes ourselves -- they can be added (as a patch) to our devel/icu -- to benefit ALL of ICU-using software. Plenty of precedent there. To have ports use their own versions -- of ANYTHING -- is not right (see li= nk above). The actual work on unbundling may be substantial -- which would explain, why it hasn't happened yet. But that such unbundling is a goal, is beyond dispute. Chromium port is not the first to stumble into this -- OpenOffice.org, for example, was quite notorious too some years ago. > The same goes for all the other deps that are picked up internally This view is held by a number of people, but FreeBSD Ports have a clearly-stated policy to the contrary: https://docs.freebsd.org/doc/10.1-RELEASE/usr/local/share/doc/freebsd/en/bo= oks/porters-handbook/bundled-libs.html Ideally, we'll get to the point, where the entire third_party/ subdirectory= is not even extracted from the upstream's tar-ball. > there have been several incompatibility issues previously Frankly, this is FUD :-) --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-265180-28929-AChpmfKs6I>