- Add new PCRE option for use with wesside
- Enable SQLite and PCRE by default
- Use upstream's strip make target instead of stripping manually
- Cleanup Makefile
In some basic benchmarks this sped up 'all-depends-list' about 20%. x11/kde4
went from 52 seconds to 41 seconds. More improvement is expected once
more command executions are cached in the 'export_ports_env' function.
With hat: portmgr
This is only missing the bsd.port.mk pieces to hook it up fully. A blocker
for hooking that up has been sub-packages, even though some implementation
could be made without them. For now just commit what I have so it is not
forgotten.
Obtained from: OneFS
Sponsored by: EMC / Isilon Storage Division
With hat: portmgr
This has been slightly faster in my tests since readelf(1) will fail on the
file much quicker if it doesn't find the ELF headers. This also more directly
finds the symbol table.
With hat: portmgr
sole purpose to avoid using our standard MAKE_ENV.
They were introduced in r279589 as part of "update to 0.0.6" PR 159499 by
Kato (duh!) some four years ago; in r359185 bapt@ had mentioned that "lots
of invocation of MAKE_CMD here are wrong as they do not properly respect
MAKE_ENV" (which is ironic as avoiding MAKE_ENV *is* their only point) but
the the real problem was neither fixed nor rationale for ugly work-around
explained.
The port builds itself through a series of recursive make(1) calls, and is
using variables to pass various bits of internal state to submakes. This
approach typically requires strict discipline and can be hard to implement
correctly, to an extent being considered harmful [Miller 1997].
Incidentally, ${MAKE_ENV} includes variables that are 1) used by the port's
own build logic and 2) are not handled in a robust way by it.
Problem #1: consider the following code from `Makefile.rules.gnu.in':
ifndef LIBDIR
LIBDIR=.
endif
This is roughly equivalent to the following:
ifeq ($(origin LIBDIR), undefined)
LIBDIR=.
else
# use whatever LIBDIR value make(1) can deduce
endif
Knowing that LIBDIR is set to some other value (`..') by inner makefiles,
that code can be rewritten more elaborately like this:
ifeq ($(origin LIBDIR), undefined)
LIBDIR=.
else ifeq ($(origin LIBDIR), file)
# use LIBDIR value set by some Makefile
else
# use whatever LIBDIR value make(1) can deduce
endif
Now, because LIBDIR is passed to make(1) via MAKE_ENV and the code above
does not have "ifeq ($(origin LIBDIR), environment)" check, the build was
affected by unexpected bogus value of it and subsequently failed. Since
the only valid place we can expect "our" LIBDIR to come from is makefiles,
we can inhibit unwanted pollution from the environment by rewriting the
initial code like this:
ifneq ($(origin LIBDIR), file)
LIBDIR=.
endif
Problem #2 is similar: checking for CFLAGS and LDFLAGS to protect their
initial assignment is very fragile as many frameworks akin to the Ports
Collection would provide some default values. While it is usually safe
to append to them, it is almost always a bad idea to use them verbatim.
Apparently, these checks were put there to support resetting CFLAGS and
LDFLAGS in `util/Makefile', but since removing them does not hurt do so
regardless of small pollution in that one case that does not affect the
build in any noticeable way.
- Support multiple values in *_OLD_CMD, i.e. we can now fix both "/usr/bin/python" and "/usr/bin/env python" at the same time
- Default *_OLD_CMD values are now always appended, so you don't need to specify them in individual ports
- Add lua support (depends on USES=lua)
- Add more default values, such as "/usr/bin/env foo" for python, perl, bash, ruby and lua
- Shebangfix now matches whole words, e.g. we will no longer (erroneously) replace "/usr/bin/perl5.005" with "${perl_CMD}5.005" (but "/usr/bin/perl -tt" is still (correctly) replaced with "${perl_CMD} -tt")
Note that *_OLD_CMD items containing spaces must now be quoted (e.g. perl_OLD_CMD=/bin/perl /usr/bin/perl "/usr/bin/env perl")
Update shebangfix usage according to new rules in many ports:
- Remove *_OLD_CMD for patterns now replaced by default
- Quote custom *_OLD_CMD which contain spaces
Fix shebangfix usage in many ports (irrelevant to infrastructure change):
- Remove redundant SHEBANG_LANG (no need to duplicate default langs)
- Remove redundant *_CMD (such as python_CMD=${LOCALBASE}/bin/python${PYTHON_VER} when USES=python is present)
- Never use *_OLD_CMD in REINPLACE_CMD matchers, these should always look for exact string
Approved by: portmgr (bapt)
Differential Revision: D3756
- Use option helpers, GC no longer needed WANT_GNOME, sort USES, and
include b.p.m. bits just once at the end
- Drop explicit "support" for ia64: the port is marked broken on it,
while the architecture itself was never a first-class citizen in
FreeBSD and official killed in -CURRENT for a while now
turnserver will not build if OpenSSL was built without SSLv3
(--no-ssl3).
LibreSSL 2.3.0 has removed SSLv3 support completely.
This change fixes the build when SSLv3 is not available
While I'm here, add LICENSE_FILE
PR: 203700
Submitted by: cpbsdmail gmail com
Approved by: mom040267 gmail com (maintainer)
MFH: 2014Q4
- Add option for v4l [1]
- Use more option helpers [1]
- Removed CUPS option since CUPS requirement has been dropped upstream [1]
While here:
- Convert to USES=localbase
- Regenerate some patches [1]
This version also includes support for Canon LiDE 220 scanners. [2]
PR: 203799 [1], 200788 [2]
Submitted by: Ports Fury [1], avg@ [2]
