This port installs qmake, which is only needed to build Qt -- not
to use it. It should also help futute work on the Qt port, by not
requiring the developer to recompile qmake as frequently as at
present. This port uses Qt-3.2.1, which is not committed yet.
Discussed with: kde@
The Isotemplate API is somewhat tricky for a beginner
although it is the best one you can build. template::parse()
[phplib template = Isotemplate] requests you to name a
source and a target where the current block gets parsed
into. Source and target can be block names or even handler
names.
PR: ports/56304
Submitted by: Alex Miller <asm@asm.kiev.ua>
All future updates for msp430 target will be provided only for 3.2.x and 3.4
versions of gcc.
So, I've downgraded gcc port to 3.2.3 + last msp430 patches.
Obtained from: Target's maintainers
to build console style tables.
A PEAR class that makes it easy to build console style tables.
This package is used by the new Horde's translation module.
PR: ports/56240
Submitted by: Thierry Thomas <thierry@pompo.net>
This patch allows the admin of the machine to choose either
print/ghostscript-gnu or print/ghostscript-afp1
PR: ports/54924
Submitted by: Jens Rehsack <rehsack@liwing.de>
Poslib is a portable C++ DNS library, a part of Posadis
project. It consists of two parts: a client library and a
server library.
Using the client library, you can simply develop applications
that use the Domain Name System (DNS). It includes many
functions for resolving, domain-name manipulation and
Resource Record (RR) creation.
The server library, based on the client core, can be used
to develop DNS servers. By implementing a query entry-point
function using the Poslib library of functions, you can
easily create DNS servers, without worrying about low-level
details such as DNS message compilation, domain-name
compression and UDP/TCP transmission.
Author: Meilof Veeningen <meilof@users.sourceforge.net>
WWW: http://posadis.sourceforge.net/projects/poslib.php
PR: ports/55195
Submitted by: Sergei Kolobov <sergei@kolobov.com>
Tclxml (ports/devel/tclxml) depends on ports/devel/tcllib,
but this is not present in RUN_DEPENDS variable (see
Makefile).
PR: ports/54584
Submitted by: Nicola Vitale <nivit@libero.it>
XXL is a library for C and C++ that provides exception handling
and asset management. Asset management is integrated with the
exception handling mechanism such that assets may be
automatically cleaned up if an exception is thrown, which
allows for much simplified program structure with respect to
error handling.
By allowing XXL to track assets and using its exception
handling features, the programmer no longer has to check error
conditions on every function call and cleanup the assets on
failure because XXL does the work.
PR: 56128
Submitted by: Robert Schlotterbeck <rws@suki.rs.tarrant.tx.us>
A small, but useful library of data structures. Has AVL
tree, binary heap, hash table, a queue, a stack, variable
length array.
PR: ports/44565
Submitted by: Peter Bozarov <peter@bozz.demon.nl>
Module::CoreList contains the hash of hashes %Module::CoreList::version,
this is keyed on perl version as indicated in $]. The second level hash
is module => version pairs.
It also contains %Module::CoreList::released hash, which has ISO formatted
versions of the release dates, as gleaned from the perlhist manpage.
Submitted by: Autrijus Tang <autrijus@autrijus.org>
This module scans potential modules used by perl programs, using
line-by-line analysis and elaborate heuristics.
Submitted by: Autrijus Tang <autrijus@autrijus.org>
This module is a simple wrapper around Locale::Maketext::Lexicon,
designed to alleviate the need of creating Language Classes for module
authors.
If Locale::Maketext::Lexicon is not present, it implements a minimal
localization function, so the program can function normally.
Submitted by: Autrijus Tang <autrijus@autrijus.org>
The PAR Toolkit is a cross between Java's JAR and Perl2EXE; It makes
cross-platform packaging and deployment a breeze for Perl programmers.
Notable features include:
* Turn your Perl programs into ready-to-run executables
* Pack scripts and requered libraries with a binary loader
* Put PAR files into @INC to avoid version conflicts
* Works with remote URL as well as local files
* Supports XS modules and DATA sections
* Turns CPAN module distributions into PAR distributions
* Install, uninstall, signs and verifies PAR distributions
* Runs scripts inside PAR files, generated by "pp -p"
Submitted by: Autrijus Tang <autrijus@autrijus.org>
This module implements overloaded version objects for all versions
of Perl, including all of the features of version objects which will
be part of Perl 5.10.0 except automatic v-string handling.
Submitted by: Autrijus Tang <autrijus@autrijus.org>
This module creates and manipulates PAR distributions. They are
architecture-specific PAR files, containing everything under blib/
of CPAN distributions after their "make" or "Build" stage, a META.yml
describing metadata of the original CPAN distribution, and a MANIFEST
detailing all files within it. Digitally signed PAR distributions
will also contain a SIGNATURE file.
The naming convention for such distributions is:
$NAME-$VERSION-$ARCH-$PERL_VERSION.par
For example, "PAR-Dist-0.01-i386-freebsd-5.8.0.par" corresponds to the
0.01 release of "PAR-Dist" on CPAN, built for perl 5.8.0 running on
"i386-freebsd".
Submitted by: autrijus@autrijus.org
Add the virtual category "scheme" to the maintained ports
(except sxm). I guess no maintainer would deny this trivial
change.
PR: ports/56052
Submitted by: Kimura Fuyuki <fuyuki@nigredo.org>
Reformat and update pkg-descr
Release 1.16 is associated with the following changes:
- Declarations can be intermixed with statements (C99).
- __typeof can have as its argument an expression
and not only an identifier.
- Support for C99 variable number of arguments preprocessor macros.
- Allow case expression ranges (gcc extension).
- Recognise __atribute__(__unused__) for determining which
identifiers should not be reported as unused (gcc extension).
- Command-line option to generate a wrongly scoped identifier and unused
include file and identifier warning report.
- Separate identifier attribute for enumeration constants.
This allows us stop incorrectly categorizing them as having global
(compilation unit) visibility.
- Error reporting format is now compatible with gcc.
- Dereferencing a function yields a function (common extension).
- Command-line option to process the file and exit.
- Document processing of the FreeBSD kernel.
- Correct typing of assembly-annotated declarators.
- Fixed assertion failure that could be caused when parts of concatenated
identifiers were no longer available (e.g. when processing files
with the -m T option.)
- Correct handling of macro parameters that match other macros and
are followed by a concatenation operator (they were erroneously replaced).
- Add workaround for gcc __builtin_va_copy in the provided
definition files.
- Corrected the handling of main() in the example definition
files.
Approved by: netchild
* The variant-to-string typecast problem that caused compilation errors
with GCC 3.3 was fixed.
* A problem with template lists was fixed.
* Documentation improved.
PR: 55958
Submitted by: maintainer
When you build a PECL module, NO_BUILD is defined => the
build target does not build anything, and make all is delayed
until `make install'.
PR: ports/55168
Submitted by: Thierry Thomas <thierry@pompo.net>
Helps keep track of modes/permissions/ownership of files in cvs
Change in distinfo is only because it isn't gzipped
PR: ports/55071
Submitted by: R.I.Pienaar <rip@devco.net>
This patch solves two problems of the actual ViewCVs port:
1. it is forbidden as it is CSS-vulnerable, ViewCVS's CVS
contains a patch but a new release was still not created
by the authors 2. it overwrites the configuration files on
installation
To solve problem 1 I "back-ported" the patch 1.117 to
lib/viewcvs.py
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/viewcvs/viewcvs/lib/viewcvs.py#rev1.117
as the author itself says, it solves the problem:
http://mailman.lyra.org/pipermail/viewcvs-dev/2002-July/000776.html
To solve problem 2 I changed the install script to install
viewcvs.conf.dist directly instead of renaming it to
viewcvs.conf, leaving it up to the user and specifying it
in the pkg-message.
PR: ports/51464
Submitted by: Lapo Luchini <lapo@m4d.sm>
- Utilize variables: PYTHON_SITELIBDIR, DOCSDIR, EXAMPLESDIR
- INSTALLS_SHLIB=yes because we do
- make portlint happy
PR: 55906
Submitted by: maintainer
- Bug fixes.
- Rudimentary interceptor support.
- Support for Python 2.3.
- Per thread and per object reference timeouts.
- Generation of stubs for all COS IDL.
- New omniORB.any module to help convert to and from Anys.
Assign port maintainership to Sergey Matveychuk <sem@ciam.ru>,
he is omniORB port maintainer and will be able to take care of
this port better.
PR: 55849
Submitted by: Sergey Matveychuk <sem@ciam.ru> (new maintainer)
- Bug fixes.
- Use of standard iostreams in omniNames etc..
- Performance improvements.
- Use DOCSDIR
PR: 55849
Submitted by: Sergey Matveychuk <sem@ciam.ru> (maintainer)
Add inc.build.mk which should be used by modules using Module::Build to build
so that they respect PREFIX
Grab maintainership
PR: 55115
Approved by: demon (mentor)
Shuffle is a perl module which performs a one pass, fair shuffle on a
list. If the list is passed as a reference to an array, the shuffle
is done in situ.
The running time of the algorithm is linear in the size of the list.
For an in situ shuffle, the memory overhead is constant; otherwise,
linear extra memory is used.
The algorithm used is discussed by Knuth [3]. It was first published
by Fisher and Yates [2], and later by Durstenfeld [1].
PR: 55574
Submitted by: andrew@scoop.co.nz
- added DT::Set count()
- fixed DT::SpanSet duration()
- methods that accepted a DT object, now accept a list of DT objects.
Suggested by Ben Bennet, I think.
- added _callback_next - "previous-only recurrences"
- optimizes intersection of recurrence with span
- internal 'S::I' recurrence class
- small fix in DT::Set clone()
- DateTime::Set uses DateTime infinity constants; max and min now return
DateTime::Infinite::Future/Past objects.
- DateTime::Span uses DateTime infinity constants; max and min now return
DateTime::Infinite::Future/Past objects.
Docs update to explain 'max()' value when a span is built using 'before'.
- added method set_time_zone()
- clone() is documented
Approved by: demon (mentor)
This module provides functions that deal the date formats used by the HTTP
protocol (and then some more).
PR: 54187
Submitted by: mat
Approved by: demon (mentor)
* Correct package name for Quick.
* Augment license conditions.
* Fixed timezones in fall.t and quick.t tests.
* Fallthrough example and test added.
* Quick parser added to simplify fallthrough stuff.
* Rejigged internals to allow for on_fail argument to multi-parsers.
Approved by: demon (mentor)
* the 'locale' release
* Applied patches from Joshua Hoblitt to move the the brand new
DateTime::Locale rather than the old ::Language modules
* Implemented %x, %X and %c for locale formats
* Fixed a bug on the two digit year determinator
* Added a test for locales known as 004locale.t
* This is a major change and you should install this release if you are using
DateTime >= 0.14 (the first release with Locale)
Bump PORTEPOCH because version went from 1.0302 to 1.04
Approved by: demon (mentor)
* It is now possible to turn validation on and off at runtime. To make this
easier, it can be set via the PERL_NO_VALIDATION environment variable, or the
$Params::Validate::NO_VALIDATION global variable. Go ahead, shoot yourself
in the foot with it!
Approved by: demon (mentor), maintainer
Arch is a really nifty revision control system. It's "whole-tree
changeset based" which means, roughly, that it can handle (with atomic
commits) file and directory adds, deletes, and renames cleanly, and
that it does branching simply and easily. Arch is also "distributed"
which means, for example that you can make arch branches of your own
from remote projects, even if you don't have write access to the
revision control archives for those projects.
This looks to be as close to an open source p4 replacement as one could
hope without being p4. I'll go so far as to suggest that if this SCM
was employed by the BSD crowd, merging changes between dragonfly (post
source repo reorog), NetBSD, and OpenBSD would be radically less painful.
It is very possible that the dragonfly fork may not have happened under
the arch SCM development methodology, but if it did, at the very least it
would be possible to incorporate dillion's reorg work in a single patch
set, no cvs admin repo surgery needed.
WWW: http://arch.fifthvision.net/bin/view
* Add support for the Polish locale
* Try to add more intelligence to the direct command use code
* Fix up the wording about numbered info files [1]
Suggested by: gerald [1]
* Clean up some more Perl warnings
* Make sure all the MAN macros are printed in all uppercase
* Add perl, ruby, python, and the auto* tools to the list of direct
commands [1]
PR: 55529 [1]
Submitted by: Jens Rehsack <rehsack@liwing.de> [1]
This is a minor bugfix release that fixes a problem with the javah task on JDK
1.4.2 and a couple of bugs in the Visual Age for Java intergration tasks. If
you don't use javah or VAJ, there is no reason to upgrade.
See:
http://ant.apache.org/antnews.html
