This document collects status reports for GNU packages.
This is a revival of the GNU Status Reports from the historical GNU's Bulletins, http://www.gnu.org/bulletins. The goal is to provide GNU-wide news and information from time to time, from as many packages as possible.
http://www.gnu.org/manual lists all GNU packages, with links to
online manuals and home pages. All GNU packages can be accessed on
the web via <http://www.gnu.org/software/pkgname>,
as shown in the headings here. This report includes items for only a
few of the hundreds of GNU packages; we hope more will be represented
in future installments.
See also http://planet.gnu.org, which aggregates the individual GNU news items posted on Savannah (http://savannah.gnu.org).
See also the [email protected] mailing list, where announcements of new GNU releases are posted. Subscribe or view the archives at http://lists.gnu.org/mailman/listinfo/info-gnu.
The aim of the present report is to be somewhat higher-level and more general than the others, although there is inevitably some overlap.
Questions, comments, and suggestions about this document in general are welcome; please email [email protected]. Issues for specific packages should of course be addressed via their usual routes.
From Nick Clifton: The GNU Binary Utilities project (binutils
to its friends) is continuing active development. Version 2.21 has
just been released containing many bug fixes and some new features:
== id for imports and exports. This
allows aliases of the import/export table names to written in PE
image.
objcopy program now supports a
--interleave-width option to allow copying a range of bytes
from the input to the output with the --interleave option.
.section directive flags for COFF targets that allows the
section's alignment to be specified.
MEMORY regions in
linker scripts.
Meanwhile work continues on ld to add support for plugins and
compressed debug sections. gold is improving and soon will be
able to replace ld for linking important projects like the
Linux kernel.
The next release of the binutils is expected in the middle of 2011.
From Jim Meyering and Pádraig Brady: We've just released coreutils-8.9, which fixes a number of relatively important bugs (http://savannah.gnu.org/forum/forum.php?forum_id=6662 for 8.8, http://savannah.gnu.org/forum/forum.php?forum_id=6679 for 8.9). Most were in sort and were introduced via the parallelization support that was added in coreutils-8.6.
The only significant non-bug-fix change was to add a useful set of features to split that lets you split input into n roughly-equal pieces, with options to split on line boundaries or not, and, when honoring line boundaries, to distribute lines in a round-robin fashion or not. See the manual via ‘info split’ for details and examples, or online at http://www.gnu.org/software/coreutils/manual/html_node/split-invocation.html.
Some of the hardest work on coreutils is knowing what to reject and providing appropriate justification to the contributors. The contributions below all came in recently, and while good ideas, they were not included for various reasons detailed on the mailing list.
test -edtouch -R`find . -exec touch -am {} +` is more general.
df autoscaledf -h was thought good enough.
df/ls --blocksize={decimal,binary}sha1sum --raw | base64.`openssl dgst -sha1 -binary $file | openssl enc -base64`
was deemed available enough.
--examplesOn the upcoming front, we should soon resume work on getting the FIEMAP (efficient sparse copying) code into the master development branch http://git.savannah.gnu.org/cgit/coreutils.git/log/?h=fiemap-copy.
From Stuart Rubin: The Electric VLSI Design System is a complete CAD package for integrated-circuit design. It lets you draw schematics, draw IC layout, and even use textual hardware-description-languages. It has many tools for analysis and synthesis, and can interface to many more.
Electric has been free software from the Free Software Foundation for 12 years now and continues to be developed and distributed. Currently, the Electric development team is employed by Oracle, and they have just released version 9.00.
From Steve White: In GNU FreeFont release 20100919, several language scripts were added:
In addition, Sinhala was replaced by a version taken from the TeX
font sinh by Yannis Haralambous.
Several of the symbol ranges were expanded, and a few glyphs were corrected.
Several technical improvements were made to font tables, some of which should improve display.
For more information, see https://savannah.gnu.org/forum/forum.php?forum_id=6518.
From Arnold Robbins: Gawk has been seeing lots of progress recently. A major new release is in the works and should be ready within a few more months. New stuff includes:
BEGINFILE and ENDFILE patterns
FPAT variable for content-based field splitting
From Gerald Pfeifer: The last couple of months have seen several GCC (GNU Compiler Collection) releases, first GCC 4.4.5 in October, then GCC 4.5.2 in December and we are working full steam to brush up the first release in the 4.6 series (http://gcc.gnu.org/gcc-4.6/changes.html).
GCC 4.6 will bring support for the Xilinx MicroBlaze softcore processor on the backend and a new frontend (and run-time library) for the Go language, plus many new Objective-C features, such as declared and synthesized properties, dot syntax, fast enumeration, optional protocol methods, method/protocol/class attributes, class extensions, and a new GNU Objective-C runtime API.
GCC 4.6 is also adding a new optimization level -Ofast that is
-O3 with options that can affect standards compliance but
result in better optimized code. The link-time optimizations (LTO) we
introduced with GCC 4.5 have been improved, stabilized and reduced in
terms of resource consumption which allows GCC to build itself or
Mozilla Firefox with LTO. As a user you can control how many
compilations to execute via -flto=N. Inlining heuristics and
interprocedural optimizations have been tuned heavily, too.
Back to frontends, stack checking for Ada has been tightened and now
detects all overflows on several platforms; experimental support for
the upcoming C1X standard has been added, new function attributes
leaf and callee_pop_aggregate, and a new warning
-Wdouble-promote. Support for the upcoming C++0x standard has
seen a lot of love and care and C++ code generation can be tuned more
finely (cf. new options -fstrict-enums and
-fnothrow-opt).
On systems supporting the new libquadmath library, GNU Fortran now
also supports a quad-precision, kind=16 floating-point data
type (REAL(16), COMPLEX(16)). The -fwhole-file
command-line option is now enabled by default which improves code
generation and diagnostics, and both Fortran 2003 and Fortran 2008
support have seen their fair dose of enhancements.
In terms of platform enhancements, -march=core2 (and
-mtune=core2) provides support for Intel Core 2,
-march=corei7 for Intel i3/i5/i7 and -march=btver1 gives
you support for AMD Bobcat (family 14). Darwin, FreeBSD, MinGW and
Cygwin now all support __float128 on 32-bit x86 targets. MIPS
now supports the Loongson 3A processor and s390x added full support
for the zEnterprise z196 processor with a number of new instructions
facilities.
As far as operating systems go, GCC provides a convenient way of
building native libraries and applications for the Android platform
and on Windows #pragma push_macro & #pragma pop_macro,
decimal floating point and more have been added.
Should you find any issue testing this forthcoming release, or any other, please make use of our updated Bugzilla at http://gcc.gnu.org/bugzilla.
From David Sugar: GNUCOMM is the umbrella name for the GNU Telephony system. It includes several packages: Bayonne, Sipwitch, and supporting libraries ccAudio, ccRTP, ccScript, and CommonCPP.
The ZRTP communication protocol is spreading as a free (as in freedom)
means to communicate privately with GNU GPL licensed software. We are
close to releasing ZRTP for Android (libcppzrtp already builds
on the Android NDK), along with a complete Android client built off of
CSipSimple for Android.
We of course continue to maintain full compatibility and interoperability with the emerging IETF ZRTP draft standards, and with ZRTP for Android will now cover even more users. Previously we saw GNU ZRTP4J, which is used in SIP Communicator (a cross-platform Java client for GNU/Linux, OS/X, Windows, and others), as well as Twinkle (a QT ZRTP agent), so ZRTP for Android will extend this reach further.
In GNU uCommon, there is also greater emphasis on secure communication and crytographic practices in application development, using a new secure library which was introduced earlier this year. It also acts as a generic wrapper for either GNUTLS or OpenSSL (with a configure preference for the former). This is still evolving. It fits into our overall effort in developing secure communications and promoting privacy in application development by default by making it easier to do so, an overall goal for GNU Telephony.
From Jason Self: I'm working to rewrite GNUtrition in C with SQLite instead of depending on an external SQL server, a modern UI, and some extra features. I plan to unveil the new version at LibrePlanet in March.
The food database is being upgraded. The current version of GNUtrition uses a very old version (from 1999) so it's time to upgrade. This means more information on more foods—2,000 or so new food items.
From Jim Meyering: Grep saw several maintenance releases in 2010.
There have been a few minor bug fixes since the last release, and
there's at least one pending, unfixed bug that is triggered when
performing a case-insensitive search that finds a match on a line
containing a Turkish upper-case I.
Once that is fixed, I expect to make another release.
From Simon Josefsson: GSASL is the GNU Simple Authentication and Security Layer library. During December 2010 a new stable release (1.6.0) was published, which includes support for the two most recent SASL mechanisms: SCRAM and GS2. These are the new generation of SASL mechanisms, intended to replace (over time) the older CRAM-MD5, DIGEST-MD5 and GSS-API mechanisms.
The development series for the next stable branch has been started already, and plans include improved GS2 support for non-Kerberos mechanisms, support for the two SAML mechanisms that are discussed in the IETF, and more. Now is a good time to join the mailing list and contribute to the project!
From Spencer Buckner: GSEGrafix is a GNOME application which uses an anti-aliased GNOME canvas for creating scientific and engineering plots. The program is written in C and reads ASCII parameter files and data files. The parameter files contain keywords and corresponding arguments for specifying plot parameters (such as data file names, data file formats, plot type, plot style, axis type, axis labels, and more). Eleven example plots, corresponding examples of Octave code or C code for creating the data files, and corresponding parameter files are included. The program can be run from a terminal window or from a graphical user interface.
The current version is gsegrafix-1.0.5. Version 1.0.5
added data files for Examples 1 and 7, for use in becoming
familiar with the program. There are links to each of these files in
the corresponding example descriptions. The program includes an HTML
help file, and an XML help file was added in this version.
Version 1.0.6 is under development and includes two additional
keywords: background_color and background_image. The
former enables the background color of the plot window to be specified
as either white, the default, or black. If black is chosen, the plot
box, tick marks, axis labels, title, and text are white. The keyword
background_image enables a background image, such as a map, to
be displayed in the plot box. The image can be scaled in four
different ways by specifying one of the parameter values:
center, fill, scale, or zoom.
From Brian Gough: The GNU Scientific Library is a numerical library written in C. It provides a large collection of mathematical routines such as random number generators, special functions, fourier transforms and least-squares fitting. There are over 1000 routines in total with an extensive test suite.
The role of the project is to provide an alternative to the well-known proprietary numerical libraries, which hinder scientific cooperation by preventing the sharing of scientific software.
The project has been in stable release for 10 years, and has maintained API stability throughout that time. Ongoing work continues on bug-fixing, particularly improving the stability of special functions, and a 1.14 maintenance release was made earlier this year.
Currently the project does not have sufficient contributors to tackle major developments, and is limited to bug fixing. A good knowledge of numerical methods is needed to work on the project—anyone interested in helping should contact [email protected] for more information.
From Brian Gough: The GNU Source Release Collection (GSRC) is a build system for GNU packages. The goal is to make it easy to install the latest GNU releases from source in your home directory on an existing distribution.
GSRC simplifies this process by automatically downloading and building
packages and their dependencies. You only need a single ./configure
&& make to get everything, as usual.
For those familiar with BSD systems, this is similar to BSD Ports. GSRC uses GAR, a build system written in GNU Make by Nick Moffit, and GARstow by Adam Sampson. GSRC is also similar to the GARNOME system for GNOME.
There are currently 136 GNU packages in GSRC which build and install successfully, including GCC, Guile, GnuPG, Emacs, coreutils, GnuTLS and many others. Packages are updated several times a week, as new releases come up.
The whole collection is also built regularly on the GCC compile farm and a weekly report posted on http://planet.gnu.org with the current build status of all GNU packages.
Goals for the project are to support the remaining GNU packages—particularly those with complex dependencies, such as Gnash and GnuCash, and to also make the latest alpha releases available as well (this has already been done for a few projects). Additional help from people familiar with those packages is welcome.
From Ludovic Courtès: GNU Guile is an implementation of the Scheme programming language that can be embedded in applications to allow users to extend them. Extensibility empowers users to leverage software freedom by making it easier to start making changes. Just as Emacs Lisp allowed unanticipated applications to be written within the Emacs environment, Guile should do the same for other GNU applications. It should spread the Emacs nature to the rest of GNU.
Guile has undergone major changes over the last couple of years, resulting in 15 pre-releases—labeled 1.9.x. This will lead to the 2.0 stable series, scheduled for release in the first quarter of 2011. The 1.8 stable series was released in 2006. It offered many improvements such as support for exact rational numbers, multi-threading, and a better C programming interface.
Guile 2.0 is a breakthrough in Guile's history. First and foremost, it is based on a compiler and a virtual machine, and comes with a powerful read-eval-print loop (REPL) and debugger. The compiler compiles Scheme code to bytecode, applying well-known optimizations. As a result, Scheme code runs noticeably faster with Guile 2.0. Compilation can occur transparently: when the compiled form of a module is not found in cache, it is automatically compiled before being run.
In addition, Guile 2.0 has a wealth of exciting new features. Hygienic macros are supported natively and well integrated with the module system—they are what makes Scheme itself extensible. It supports Unicode, as well as most of R6RS, the latest revision of the Scheme programming language.
A dynamic foreign function interface (FFI) is provided, allowing bindings to C libraries to be written entirely in Scheme. Also new in 2.0 is support for composable continuations (a.k.a. “prompts”), a novel control flow mechanism. New modules include web tools for HTTP and URIs, XML/SXML, pattern matching, LALR parsing, purely functional hash lists, and code coverage, to name a few.
While Guile 2.0 remains primarily an implementation of Scheme, its compiler comes with a front-end for ECMAScript—one more choice for users seeking to extend their applications. An Emacs Lisp front-end is also available. It is the result of work towards the unification of Guile and Emacs, the flagships of extensibility in GNU.
2011 promises to be another exciting year for Guile. We believe Guile 2.0 is a great tool to push extensibility further and enhance software freedom!
From Karl Berry: GNU Hello remains a complete example package, attempting to exemplify some best practices for GNU packages. For example, it uses Automake for configuration, Gettext for internationalization, Help2man to avoid maintaining separate man pages, and Gnulib to share common sources.
For examples in other programming languages, please see Gettext. The Automake manual also contains several examples worked from scratch.
From Olaf Buddenhagen, Arne Babenhauserheide, Thomas Schwinge: Yeah, that's quite right: this project is still alive!
According to our mission statement, the goal is creating a general-purpose kernel suitable for the GNU operating system, which is viable for everyday use, and gives users and programs as much control over their computing environment as possible. It has a unique multi-server microkernel-based architecture—bringing advanced operating system research to the mainstream. More concretely, it's a collection of user-space server processes that run on the GNU Mach microkernel.
The Hurd doesn't fully deliver on the everyday usability goal yet, but it is seeing continuous improvement—and 2010 has been no exception. Let's take a look at the progress throughout the year.
Jérémie Koenig ported the modern Debian Installer to Debian GNU/Hurd. Installation images using the new installer are replacing the previous CD images, which were using an installer based on the old Debian boot floppies (and running under the Linux kernel)—Philip Charles has been maintaining these single-handedly for almost ten years! The new installer images are available from http://people.debian.org/~sthibault/hurd-i386/installer/cdimage/.
Emilio Pozuelo Monfort was investigating specific compatibility problems exposed by the extensive test suites coming with some software packages. Emilio's analysis uncovered a number of programming errors in the Hurd code, and he fixed several of them. As these typically affected other programs too, this improved stability and compatibility in general.
Some other translators (gopherfs, netio, tarfs) which have been created by external contributors in the past have been fixed up by Manuel Menal, and packaged in Debian. Thus, some of the results of Hurd's extensible architecture are now easier to access, and these updated translators can serve as examples for other developers to implement their own ideas.
This is a very short digest of what happened in the last year. You can read our regular “Month of the Hurd” at http://www.gnu.org/software/hurd/news.html, or by subscribing to our RSS feed at http://www.gnu.org/software/hurd/index.rss.
If you are interested, for example, in doing a university project on a
multi-server microkernel-based operating system, or if you are
interested in contributing to Hurd development in general, please see
http://www.gnu.org/software/hurd/contributing.html. Or just
talk to us at [email protected] or the #hurd IRC
channel on freenode.
From Karl Berry: We hope the next major release of Texinfo will be made in 2011, after several years of development. Thanks to my colleague Patrice Dumas, the Makeinfo program has been completely rewritten in Perl while retaining essentially complete compatibility. The new program is based on and is as compatible as possible with texi2html. It will provide vastly better support for HTML customization, Unicode, additional back-ends, and so on. If you want to try it now, it's in the development sources.
texinfo.tex, as a separable file, has been updated separately on an ongoing basis. The current version is available from http://ftpmirror.gnu.org/texinfo/texinfo.tex (and from Gnulib), and we encourage you to always get the current version when making a release.
From Arun Persaud: XBoard is a graphical user interface to chess in all its major forms (and many others). Over the last year XBoard development has seen three new releases fixing lots of bugs and including many new features with another major release just around the corner. We now support many more variations of chess (xiangqi, shogi, makruk, as well as better support for variations like crazyhouse, bughouse and many more). Also, the support for computer chess has been improved.
An ongoing effort is to merge the code of XBoard and its so-called “Winboard” companion back into one code base and adapt it to use GTK for the GUI front end. If you are interested in this and are willing to help, feel free to contact us at [email protected]. We can always use more help! You can of course also contact us to discuss other issues/ideas too.