- fix a use after free
- disable tls since we don't support it yet
- BSD endian fix from author Robert Lougher <rob.lougher at gmail.com>
testing and okay MANTAINER Frederick C. Druseikis <fdruseikis at sc.edu>
progress, largely based on the gcc port in ports/lang/gcc/4.2.
Requested by jsg@.
It's somewhat usable on i386 (shared lib versions not yet properly
under control). Build on amd64 currently fails with -fPIC problems.
-- --
lvm-gcc is the LLVM C front end. It is a modified version of gcc
that compiles C/C++/ObjC programs into native objects, LLVM bitcode or
LLVM assembly language, depending upon the options.
By default, llvm-gcc compiles to native objects just like GCC does.
If the -emit-llvm option is given then it will generate LLVM bitcode
files instead. If -S (assembly) is also given, then it will generate
LLVM assembly.
Being derived from the GNU Compiler Collection, llvm-gcc has many of
gcc's features and accepts most of gcc's options. It handles a number
of gcc's extensions to the C programming language.
<sthen@zephyr:/usr/ports/mystuff/lang/llvm-gcc4:9>$CVS: ----------------------------------------------------------------------
all secondary compilers were relocated to separate projects so we need a
rakudo port to get the perl6 binary back.
this update also addresses the recent bulk fallout noticed by naddy@
testing by sthen@ and ajacoutot@, thanks!
ANTLR, ANother Tool for Language Recognition, (formerly PCCTS) is a
language tool that provides a framework for constructing recognizers,
compilers, and translators from grammatical descriptions containing
Java, C#, Python, or C++ actions.
Currently installing the precompiled jar since this is needed for
classpath 0.98 as a BUILD_DEPEND and RUN_DEPEND.
From MAINTAINER: Frederick C. Druseikis <fredd@engr.sc.edu>
people from screwing themselves by using libstdc++-3.x which will fail
only in bizarre ways (embarassing how long it took me to debug this)
ok robert@
"finally! ok" todd@
place to change if you need to use a different Tcl/Tk version.
- provide MODTCL_LIB and MODTK_LIB (avoids a possible messy
construct in an individual port's Makefile when they are needed,
allows use of "LDFLAGS=-L${MODTCL_LIBDIR} -l${MODTCL_LIB}").
ok steven@, Stuart Cassoff
Lots of python ports already do this, so better factorize. It will also
help mitigate some build breakages when several python versions are
installed.
Related ports cleanup coming in a few...
"looks ok" wcmaier@
"makes sense" sthen@
"no objection" djm@
with Chez Scheme but uses high-speed threaded interpreter technology in
place of Chez Scheme's incremental native-code compiler. Petite Chez
Scheme may be used without license, fee or royalty for any purpose,
including for resale as part of a commercial product.
Submitted by (a very patient) Aaron W. Hsu <arcfide at sacrideo dot us>
with license help and sanity checking from sthen@.
a few patches to deal with shared libraries.
there is lisp code to deal with recognizing .so, so until someone dives
in and adapts it for OpenBSD, keep a libecl.so...
work on OpenBSD, and exceptions are hevaily used by OpenOffice.Org.
Backport PR libstdc++/31481 from GCC repository because this fix is needed
by openoffice:
PR libstdc++/31481
* include/ext/type_traits.h (__numeric_traits): Move...
* include/ext/numeric_traits.h: ... here; fix type of
__max_digits10.
* include/ext/pb_ds/detail/type_utils.hpp: Include
<ext/numeric_traits.h> too.
* include/tr1/random: Likewise.
Tested with both openoffice and webkit. bump needed PKGNAMEs;
2.4.4 => 2.4.8
2.5.2 => 2.5.4
2.6 => 2.6.1
Python 2.4 and 2.5 lose their build knobs to match 2.6.
Removes no longer needed Python 2.5 security patches backported
from the release25-maint SVN branch.
Remove the -bz2 subpackage from all three versions. It is silly
to make a subpackage to avoid depending on something tiny and
compatibly licensed.
Python 2.4 and 2.5 lose their -expat subpackages; expat has been
in base for some time.
Python 2.5 loses its sqlite subpackge. Again, sqlite is tiny,
compatibly licensed and is depended upon by more and more
applications. This brings it into line with the 2.6 version.
Rework all three version's handling of setup.py. Rather than regex
replacing LOCALBASE and X11BASE into setup.py post-configure, these
are passed in though environment variables. Will save hours of
frustrated cursing familiar to anyone who has accidently used the
update-patches target after configure and had to go back and redo
all the substitutions.
Rework the patching of setup.py for 2.4 and 2.5 to be more like
what we do for 2.6. I.e. keep the diff minimal and avoid deleting
huge blocks of code, so the diff has a chance of applying without
massive hand-editing each patch release.
Fix .py paths in installed .pyc files (patch from eric@)
feedback from several, particularly eric@, ajacoutot@ and Ingo
Schwarze; "get it in" ajacoutot@
