• https://rubygems.org/downloads/ (Download)

• ruby25
• ruby26
• ruby27

• (2019-11-12) Updated to version: passenger-6.0.4
• (2017-11-04) Updated to version: passenger-5.1.11
• (2016-11-05) Updated to version: passenger-5.0.30
• (2015-07-26) Updated to version: passenger-5.0.14
• (2014-06-25) Updated to version: passenger-4.0.44
• (2011-12-14) Updated to version: passenger-3.0.11
• (2011-05-17) Updated to version: ruby-passenger-3.0.7
• (2010-11-09) Updated to version: ruby-passenger-3.0.0
• (2010-06-25) Updated to version: ruby-passenger-2.2.5p1
• (2010-01-28) Updated to version: ruby-passenger-2.2.5p0

Log message:
Make use of "find -exec {} +" (which is POSIX) and "find -delete"
(which is not) throughout the ports Makefiles.
* Replace find|xargs with find -exec {} +
* Replace -exec {} \; with -exec {} + if applicable.
* Use the -delete operator to remove files and empty directories.
* Combine and tweak some find(1) invocations while here.
ok kn@ rsadowski@ espie@

Log message:
fix typo to unbreak bulks

Log message:
This won't build on any but i386, amd64, arm, and aarch64. We've already
marked nginx-passenger ONLY_FOR_ARCHES. Do the same for www/ruby-passenger
itself.
ok jeremy@

Log message:
Update to passenger 6.0.4

Log message:
replace simple PERMIT_PACKAGE_CDROM=Yes with PERMIT_PACKAGE=Yes

Log message:
bump REVISION in a bunch of ports known or likely to use struct kinfo_proc,
there may be some missing as my unpacked ports source is a little out of date
but this should catch the main things people might run into
the struct was reordered a second time in sysctl.h r1.192 to improve
compatibility but amd64 snapshot packages made it out before that happened
so the bumps are still needed

Log message:
fix build with curl >=7.62.0

Log message:
Add COMPILER lines to c++ ports which currently use the default. Adjust
some existing COMPILER lines with arch restrictions etc. In the usual
case this is now using "COMPILER = base-clang ports-gcc base-gcc" on
ports with c++ libraries in WANTLIB.
This is basically intended to be a noop on architectures using clang
as the system compiler, but help with other architectures where we
currently have many ports knocked out due to building with an unsuitable
compiler -
- some ports require c++11/newer so the GCC version in base that is used
on these archirtectures is too old.
- some ports have conflicts where an executable is built with one compiler
(e.g. gcc from base) but a library dependency is built with a different
one (e.g. gcc from ports), resulted in mixing incompatible libraries in the
same address space.
devel/gmp is intentionally skipped as it's on the path to building gcc -
the c++ library there is unused in ports (and not built by default upstream)
so intending to disable building gmpcxx in a future commit.

Log message:
Don't pickup gmkdir during libuv build
Should fix issue that hit espie@.  No bump as this only affects building.

Log message:
Remove BROKEN-sparc64
Built repeatedly just fine. This reenables www/nginx.
Switch to TLS while here.
OK sthen

Log message:
Add OpenBSD comment to rubygems-ext.PLIST
Bump ruby gem ext ports as this changes the package.
Requested by espie@

Log message:
Bump the remaining ports to trigger updates following cpu stats
changes (CPUSTATES/etc). Last part.
- some of these might be in code that's not actually built or rarely
used but it's not worth the hassle evaluating that, i'm just bumping &
moving on.
- as mentioned before, there might still be runtime problems relating
to the addition or renumbering.

Log message:
Now that arm has switched to clang the base compiler has atomic builtins
and accepts -mfpu=neon.
ok jca@ sthen@

Log message:
Force use of awk
Problem noticed by ajacoutot@

Log message:
bump LIBCXX/LIBECXX/COMPILER_LIBCXX ports.

Log message:
use COMPILER_LIBCXX where applicable

Log message:
use LIBCXX

Log message:
Don't have embedded libev configure script pick up gmkdir
Pointed out by espie@

Log message:
Mark as BROKEN on arm, undefined reference to __sync_val_compare_and_swap_4
The __sync_val_compare_and_swap_4 reference comes from the internal
libuv copy.  Using lang/gcc doesn't seem to help.

Log message:
Update to passenger 5.0.30
Update embedded nginx to 1.10.2.
Support nginx dynamic modules.
All files that were patched previously got moved, add patches for the
new locations.
Remove all of the pointless hiding of build steps behind progress bars.
OK robert@

Log message:
sync WANTLIB (and in one case, add gettext to LIB_DEPENDS) in dependent ports
now that internationalised domain name support has been removed from net/curl

Log message:
Add BDEP on gsed, since embedded libuv picks it up at runtime
problem noticed by naddy@

Log message:
remove SHARED_ONLY from some stragglers

Log message:
fails on sparc64 like on hppa: rake blows when linking PassengerHelperAgent

Log message:
sync WANTLIB (curl+nghttp2)

Log message:
Update to passenger 5.0.14
Remove subpackages, revert to a single package, since standalone package
is now only 10% the size of the main package.

Log message:
Remove MODRUBY_RAKE_DEPENDS.  Only necessary on ruby 1.8, and none of
the ports that use it are still built with ruby 1.8.
OK ajacoutot@

Log message:
Make ruby22 the default FLAVOR

Log message:
Remove obsolete/useless patch: our sigaltstack() implementation
does the right thing since we switched to rthreads, as confirmed
by miod@
ok jeremy@

Log message:
Support ruby22 FLAVOR for ruby-passenger
Add patch to make passenger not fail at runtime on ruby 2.2.

Log message:
fix compatibility with nginx 1.7.9; pulled from upstream git

Log message:
Yet another bunch of ports that reliably fail to build on hppa.
BROKEN-hppa =	SIGILL when running lilypond
BROKEN-hppa =	libjson-c.so.0.0: undefined reference to __sync_val_compare_and_swap_4
BROKEN-hppa=	missing atomic ops
BROKEN-hppa=	./iodev/hdimage/hdimage.cc:807: error: 'bx_bswap32' was not declared in this scope
BROKEN-hppa =	cc1plus: error: unrecognized command line option '-std=c++11'
BROKEN-hppa =	src/bin/jp2/convert.c:192: error: expected identifier or '(' before '__extension__'
BROKEN-hppa=	libpt.so.7.0: undefined reference to __sync_fetch_and_or_4
BROKEN-hppa=	ykpers-json.c:42: error: redefinition of typedef 'json_bool'
BROKEN-hppa=	OOM on src/lxml/lxml.etree.c
BROKEN-hppa =	undefined reference to `__sync_val_compare_and_swap_4' in libjson-c
BROKEN-hppa =	rake fails when linking PassengerHelperAgent
BROKEN-hppa =	kstars/kstars/tools/scriptbuilder.cpp:103: internal compiler error: Segmentation fault
BROKEN-hppa=	build stuck on virt-viewer-window.c
See http://build-failures.rhaalovely.net/hppa/

Log message:
Bump after change to remove OpenBSD version from ruby SITEARCHDIR.
OK landry@, jcs@

Log message:
bump revision for plist changes 4.9 -> 5.0

Log message:
Fix broken PLIST.  Update WANTLIB while here and bump REVISION.
OK william@

Log message:
Update to passenger 3.0.7
Ignore EINVAL from sigaltstack, seems to be what ruby itself does in 1.9
committing on behalf of jeremy@, who did all the work,
tested on i386 and tested packaging on amd64.  All so I can commit a
simple nginx update... THANKS.
ok jasper

Log message:
bump revision for plist changes 4.8 -> 4.9

Log message:
- fix build on at least mips64el by syncing the endian definitions, tested
on both big and little endian machines.

Log message:
REVISION bump all ruby ports due to a change in the pkgspec.
Previously, we were using ruby->=1.8,<=1.9, instead of
ruby->=1.8,<1.9.  While this wouldn't cause an issue, since
our ruby-1.9.2 package isn't included in ruby->=1.8,<=1.9,
it's still wrong and should be fixed.  This also fixes the
following minor issues:
Switch from using FLAVOR to MODRUBY_FLAVOR for *_DEPENDS.
Currently we don't have a ruby port that uses FLAVORs that
would differ from MODRUBY_FLAVOR, but it's possible we will
in the future.
Switch from BASE_PKGPATH to BUILD_PKGPATH in a few cases in
REGRESS_DEPENDS.  This probably is not strictly necessary, but
BUILD_PKGPATH is used in more cases, so it is good for
consistency.
Switch to new style *_DEPENDS, with the version specification
at the end.  The remaining cases where this is not done is
because a specific version is used.
Some FULLPKGNAME added to REGRESS_DEPENDS, to make sure that if
the old version is installed when you run a regress test, it
will install the new version first.
Some conversion of spaces to tabs for consistency.
OK landry@

Log message:
new depends

Log message:
- add missing dependency

Log message:
Update ruby-passenger to 3.0.0.  Switch the directory name from
phusion_passenger to phusion-passenger, since that's the name used
by upstream. Also tested by Pierre-Yves Ritschard.
OK landry@

Log message:
Handle the rest of the fallout from the splitting of MODRUBY_LIB_DEPENDS
into MODRUBY_WANTLIB and using CONFIGURE_STYLE = ruby gem ext.  Use the
lang/ruby module for all dependent ports, setting
MODRUBY_{BUILD,RUN}DEP=No if necessary.
ok landry, phessler, sthen

Log message:
Remove MASTER_SITE_RUBYFORGE from network.conf.template.
It's only remaining user is www/ruby-passenger, so hardcode
it in that port's Makefile.
ok landry

Log message:
Remove Bernd as maintainer of ruby-passenger
OK bernd@

Log message:
Bump all ruby ports not already handled.
In the upgrade from ruby 1.8.6 to 1.8.7, the PLISTs changed
due to differences in how RDoc processes files.
This also has a number of changes to the regress tests to
work with the changes to devel/ruby-rake.  It moves most of
the regress tests to use MODRUBY_REGRESS.
OK jcs@, landry@, jasper@, sthen@

Log message:
The old version of boost included in this port assumes that GCC 4.1 and later
always provide the new atomic builtins.  In reality those are only available
on hardware that provides the necessary atomic instructions.  PA-RISC doesn't
have a suitable set of atomic instructions so work around it on hppa.
ok espie@, bernd@, landry@

Log message:
bump PKGNAME for PLIST changes 4.7 -> 4.8

Log message:
Fix build with nginx->=0.7.66
from upstream commit e973b93e8998edbbb4c6
ok bernd@
found by naddy@, thanks!

Log message:
bump PKGNAME for PLIST changes 4.6 -> 4.7