This was specified on USES: iconv:build,lib
This is incorrect; the way iconv.mk is written, "build" and "lib" are
mutually exclusive and "build" takes precedence. This means the library
dependency was not registered.
Moveover, it's failing on FreeBSD 11 and dev-branch of DragonFly because
it uses transliteration. Setting USES=iconv:translit fixes both issues.
Approved by: Just fix it
because the old name wasn't accurate - the class libraries are only a fraction
of the software.
The STEP Class Library (SCL) originated at the National Institute of Standards
and Technology, or NIST. NIST started working with STEP in the 80's and
continued until the late 90's. Some components of SCL were originally written
in Lisp and then re-written in mixed C and C++ in the early 90's.
The rest of SCL was written in C++ to begin with.
STEPcode (SC) includes the class libraries, some of the most widely used EXPRESS
schemas, some tools to work with EXPRESS, and support libraries for those tools.
Two of the tools can create schema-specific libraries that are used with the
class libraries. There are also some test files and programs.
WWW: https://github.com/stepcode/stepcode/wiki
PR: 201046
Submitted by: fernando.apesteguia@gmail.com
According to project documentation and source review/runtime tests this port
doesn't using py-xml, so remove it and bump PORTREVISION.
Add NO_ARCH and limit python version to 2.x while here (because of pygtk2).
With hat: python
The LLVM backend builds much faster assuming clang35 and llvm35
are already available. It probably builds on FreeBSD's base clang but
not ports llvm36. In any case, the alternative backend does not
pass the testsuite on FreeBSD (fails at test 825).
The LLVM back is the default backend for DragonFly which allows the
port to build, but it fails the very first test (fails to lock mutex).
There's no need to bump -- for FreeBSD this should package the same
as it did before.
GHDL is the leading VHSIC Hardware Description Language (VHDL) simulator.
Digital and mixed-signal systems such as field-programmable gate arrays
and integrated circuits can be described by VHDL, and VHDL can also be
used as a general purpose parallel programming language. GHDL compiles
VHDL files and creates a binary which simulates the design.
GHDL fully supports IEEE 1076-1987, IEEE 1-76-1993, IEEE 1076-2002
versions of VHDL, and partially IEEE 1076-2008.
- Replace ${MASTER_SITE_FOO} with FOO.
- Merge MASTER_SITE_SUBDIR into MASTER_SITES when possible. (This means 99.9%
of the time.)
- Remove occurrences of MASTER_SITE_LOCAL when no subdirectory was present and
no hint of what it should be was present.
- Fix some logic.
- And generally, make things more simple and easy to understand.
While there, add magic values to the FESTIVAL, GENTOO, GIMP, GNUPG, QT and
SAMBA macros.
Also, replace some EXTRACT_SUFX occurences with USES=tar:*.
Checked by: make fetch-urlall-list
With hat: portmgr
Sponsored by: Absolight
When appropriate:
- Try to use DISTVERSION{SUF,PRE}FIX
- Replace PORTNAME-PORTVERSION by DISTNAME
- Convert MASTER_SITES to use macros
- Other light cleanup
With hat: portmgr
Sponsored by: Absolight
/usr/local/lib/libwebkitgtk-1.0.so.0: undefined reference to `std::chrono::_V2::steady_clock::now()@GLIBCXX_3.4.19'
/usr/local/lib/libjavascriptcoregtk-1.0.so.0: undefined reference to `std::chrono::_V2::system_clock::now()@GLIBCXX_3.4.19'
collect2: error: ld returned 1 exit status
This is standard C++11, but there is a problem with stdlib.
Note: now it could be built with clang, but the produced binaries don't
behave correctly.
I have the idea that ldraw is always released under the same filename.
The port was marked broken due to a checksum mismatch; the file was
fetched but it had changed.
I've expanded DIST_SUBDIR to include the PORTVERSION. I think the scheme
used is just {YEAR}{ITERATION-2-DIGIT} so I picked 201501 as the new
PORTREVISION. This builds under poudriere testport just fine.
Approved by: blanket
