swith. This allows build with threaded Tk (otherwise configure fails to use
Tk). Also depend on threaded Tk by default in case of THREADS support - with
non-pthread tk library it will not be stable enough
- Add OPTIONS
- Enable builds of non-threaded targets
- Depend on X in case if X is installed, since configure does this by default.
- Rename patchfiles to make portlint happy
- Eliminate the number of reinplace patterns (remove duplicates, optimize)
- Move manpage list in the separate files
- Pass pthread flags to compiler (PTHREAD_CFLAGS/LIBS safety)
- Remove some ugly hacks on setting permissions etc (eliminated by setting
correct permission in COPY_TREE macros)
- Eliminate duplicate entries in PLIST in case of unstalling optimized binaries,
also optimize this part
- Remove unused PLIST_SUB variable
- Miscelanios cleanups in Makefile
- Pass maintainership to submitter
The following files was added:
- files/man1
- files/manlinks
- files/mann
- files/patch-camlp4-man_Makefile
- files/patch-camlp4_lib_Makefile
- files/patch-ocamldoc_Makefile
- files/patch-stdlib_Makefile
I removed the following files:
- files/patch-camlp4-man::Makefile
- files/patch-camlp4::lib::Makefile
- files/patch-ocamldoc::Makefile
- files/patch-stdlib::Makefile
Also add bsd.ocaml.mk until it will be included in Mk/
PR: ports/101122
Submitted by: Stanislav Sedov
This module allows you to kill off Apache httpd processes if they grow
too large. You can make the decision to kill a process based on its
overall size, by setting a minimum limit on shared memory, or a maximum
on unshared memory.
You can set limits for each of these sizes, and if any limit is exceeded,
the process will be killed.
You can also limit the frequency that these sizes are checked so that
this module only checks every N requests.
WWW: http://search.cpan.org/dist/Apache-SizeLimit/
Justification: socialtext dependency
description:
TRE is a lightweight, robust, and efficient POSIX compliant regexp
matching library supporting:
- approximate (fuzzy) matching,
- strict standards conformance,
- predicable and modest memory consumption,
- wide-character and multibyte character support,
- binary pattern and data support,
- thread-safe implementation.
At the core of TRE is a new algorithm for regular expression matching
with submatch addressing. The algorithm uses linear worst-case time
in the length of the text being searched, and quadratic worst-case
time in the length of the used regular expression.
This module provides a simple functional "named parameters" style interface
for creating URIs. Underneath the hood it uses URI.pm, though because of
the simplified interface it may not support all possible options for all
types of URIs.
It was created for the common case where you simply want to have a simple
interface for creating syntactically correct URIs from known components
(like a path and query string). Doing this using the native URI.pm
interface is rather tedious, requiring a number of method calls, which is
particularly ugly when done inside a templating system such as Mason or
TT2.
WWW: http://search.cpan.org/dist/URI-FromHash/
Justification: socialtext dependency
- Add -D/--debug and adjust the verbosity level.
- Show before and after file sizes when -v is specified.
- Introduce a configuration file in which command aliases can be
defined.
- Add -i/--preserve-inode so inode numbers are preserved even when -b
or -s is specified.
This module provides functions that deals with formatting data with
Content-Type 'text/plain; format=flowed' as described in RFC2646
(http://www.rfc-editor.org/rfc/rfc2646.txt). In a nutshell,
format=flowed text solves the problem in plain text files where it
is not known which lines can be considered a logical paragraph,
enabling lines to be automatically flowed (wrapped and/or joined)
as appropriate when displaying.
In format=flowed, a soft newline is expressed as " \n", while hard
newlines are expressed as "\n". Soft newlines can be automatically
deleted or inserted as appropriate when the text is reformatted.
WWW: http://search.cpan.org/dist/Text-Flowed/
Justification: socialtext dependency
This module allows a Class::AlzaboWrapper::Cursor object to be used as
a TT2 iterator.
For a cursor which returns one object at a time, the iterator simply
returns one object per iteration. When the cursor returns multiple
objects, the iterator returns a hash reference where the keys are the
table name of the object's class in lower-case, with camel-casing
turned into underscores. The values of the hash are the objects.
So if the cursor returns Foo::User and Foo::Page objects, the keys are
"user" and "page".
WWW: http://search.cpan.org/dist/Template-Iterator-AlzaboWrapperCursor/
Justification: socialtext dependency
implements a genetic algorithm to find the "best" options for
compiling programs with the GNU Compiler Collection (GCC) C and C++
compilers. "Best", in this context, is defined as those options
that produce the fastest executable program from a given source
code. Acovea is a C++ framework that can be extended to test other
programming languages and non-GCC compilers.
WWW: http://www.coyotegulch.com/products/acovea/index.html
PR: ports/101211
Submitted by: trasz <trasz at pin.if.uz.zgora.pl>
The Readonly module (q.v.) is an effective way to create non-modifiable
variables. However, it's relatively slow.
The reason it's slow is that is implements the read-only-ness of variables
via tied objects. This mechanism is inherently slow. Perl simply has to do
a lot of work under the hood to make tied variables work.
This module corrects the speed problem, at least with respect to scalar
variables. When Readonly::XS is installed, Readonly uses it to access the
internals of scalar variables. Instead of creating a scalar variable object
and tying it, Readonly simply flips the SvREADONLY bit in the scalar's
FLAGS structure.
Readonly arrays and hashes are not sped up by this, since the SvREADONLY
flag only works for scalars. Arrays and hashes always use the tie interface.
Why implement this as a separate module? Because not everyone can use XS.
Not everyone has a C compiler. Also, installations with a statically-linked
perl may not want to recompile their perl binary just for this module.
Rather than render Readonly.pm useless for these people, the XS portion was
put into a separate module.
WWW: http://search.cpan.org/dist/Readonly-XS/
Justification: socialtext dependency
This provides a simple interface to Plucene. Plucene is large and multi-
featured, and it expected that users will subclass it, and tie all the
pieces together to suit their own needs. Plucene::Simple is, therefore,
just one way to use Plucene. It's not expected that it will do exactly
what *you* want, but you can always use it as an example of how to
build your own interface.
WWW: http://search.cpan.org/dist/PluceneSimple/
Justification: socialtext dependency
Quirks: 1/6 test fails
in many of my programs, but which aren't "big enough" to warrant
an individual library.
Key features of Coyotl include:
-- A polymorphic collection of the best psuedorandom number generators,
including the Mersenne Twister and Marsaglia's favorites.
-- Utilities for floating-point numbers, including additional functions
for trigonometry, least common multiple, greatest common denominator,
rounding, and other purposes.
-- A simple cross-platform command-line parser.
-- A framework for generating random rectangular mazes.
-- A template for fixed-point math based on different integer sizes
and decimal point locations.
-- Templatized sorting utilities (designed before Std. C++'s <algorithms>,
but still useful)
-- Validation tools for "Design by Contract" programming.
WWW: http://www.coyotegulch.com/products/libcoyotl/index.html
PR: ports/101209
Submitted by: trasz <trasz at pin.if.uz.zgora.pl>
But it doesn't work unless libiconv is linked.
- Link with libiconv.
- Correct charset examples
- bump PORTREVISION
PR: ports/101223
Submitted by: TAOKA Fumiyoshi <fmysh_AT_iijmio-mail dot jp>
