|
"Examples of designs that meet most of the criteria for "goodness" (easy to understand, flexible, efficient) are a recursive descent parser, which is traditional procedural code. Another example is the STL, which is a generic library of containers and algorithms depending crucially on both traditional procedural code and on parametric polymorphism." --Bjarne Stroustrup |
Welcome to SpiritSpirit is an object oriented recursive descent parser framework implemented using template meta-programming [1] techniques. Expression templates [2] allow us to approximate the syntax of Extended Backus Normal Form [3] (EBNF) completely in C++. Parser objects are composed through operator overloading and the result is a backtracking, top down parser that is capable of parsing rather ambiguous grammars. The Spirit framework enables a target grammar to be written exclusively in C++. Inline EBNF grammar specifications can mix freely with other C++ code and, thanks to the generative power of C++ templates, are immediately executable. Spirit is part of Boost Libraries, a peer-reviewed, open collaborative development effort. |
News
August 17, 2008.
Spirit2 and Boost 1.36.0
Boost 1.36.0 has been released. Along with it, comes Spirit 1.8.7, integrated with Spirit V2, "The New Generation".
Even if the directory structure has changed (the Spirit Classic headers are now moved to the '$BOOST_ROOT/boost/spirit/home/classic' directory), we created forwarding headers allowing to compile existing applications without any change. These forwarding headers are deprecated, though, which will result in corresponding warnings generated for each of the headers. The forwarding headers are expected to be removed in the future. The recommended way of using Spirit Classic is now to include header files from the directory '$BOOST_ROOT/boost/spirit/include'. All files of Spirit Classic have now a 'classic_' prefixed to their name. For example the include
#include <boost/spirit/core/core.hpp>
now should be written as:
#include <boost/spirit/include/classic_core.hpp>
To avoid namespace conflicts with the new Spirit V2 library we moved Spirit Classic into the namespace boost::spirit::classic . This change will be automatically deactivated whenever the deprecated include files are being used. This ensures full backwards compatibility for existing applications. For more details about this change please consult the documentation.
Spirit2 is in beta. For the adventurous only. The interface is stable (and has been for almost 2 years now). The documentation is still sparse, but there are tons of examples in there. Also, don't complain about compile times. We haven't optimized on compile times yet. There's lots of room for improvement in that area still.
April 18, 2008.
Spirit2 development
FYI, Spirit2 development has been moved to Boost SVN trunk. We successfully merged with Spirit 1.8.x which is called "classic" Spirit now. Users of "classic" Spirit should not have to do anything; it's fully backward compatible. However, there are changes in the include structure to accommodate different versions under one spirit directory at the same time. While the old includes will still work, they are deprecated with a warning message:
"This header is deprecated. Please use: xxx.hpp"
The new include scheme is easy to use. We have an "include" directory where we place all forwarding headers to the actual files. For classic spirit, the files are prepended with "classic_". For example, what used to be:
"boost/spirit/core/parser.hpp"
is now:
"boost/spirit/include/classic_parser.hpp"
November 28, 2007.
Post-skips (Important for Boost 1.34 and above users)
In boost 1.34, if there is a trailing white space characters in an input sequence, parse returns with info.full set to false, even though parsing was (essentially) full. The skip parser is ignored for the last sequence. In 1.33 parsing succeeds as expected.
The change in behavior was intentional. It was made on the grounds of consistency (parsers shouldn't consume any more output than what they have matched), as well as to avoid a hang when directly parsing stdin: namely, parse() would hang waiting for a non-whitespace character at the end of a match.
You may workaround this change by adding eps_p or end_p after your grammar, as in:
my_grammar g ;
my_skipper skipper ;
parse ( str , g >> eps_p , skipper ); // 1
parse ( str , g >> end_p , skipper ); // 2
The 2nd option requires a full match, while the first one gets you the previous behaviour without making a full match obligatory.
Here's where the change was first announced, for context:
http://thread.gmane.org/gmane.comp.parsers.spirit.general/9839
September 02, 2007.
Spirit 1.8.5 and 1.6.4 released
It is my pleasure to announce the release of Spirit 1.8.5 and 1.6.4 .
Spirit 1.8.5 is a standalone variant of the version that is included in Boost 1.34.1; it doesn't provide new features, but contains all the fixes that were applied since Spirit 1.8.4 . You can download Spirit 1.8.5 here.
Spirit 1.6.4 is the latest in a series of releases that are based on an earlier implementation that is friendlier towards older, less standard compliant compilers. This is a bug fix only release and is likely to be the final 1.6.x version. Spirit 1.6.4 is available here.
Both Spirit 1.8.5 and 1.6.4 are available in the following variants:
- Library only (the spirit-1.x.y.zip / spirit-1.x.y.tar.gz files) - These contain only Spirit itself and require an existing Boost installation. Both were tested only with Boost 1.34.0 and 1.34.1, but should work with other reasonably recent releases.
- Miniboost (the spirit-1.x.y-miniboost.zip / spirit-1.x.y-miniboost.tar.gz files) - These contain also the minimal subset of Boost that is needed to use Spirit. This is a new version of the Miniboost that is based on Boost 1.34.1 and has the same directory structure as Boost itself.
Enjoy!
-- Nicola Musatti
April 05, 2007.
BoostCon '07
Have you heard? Boost will be having its inaugural conference. It will be in Aspen, Colorado, May 14-18 2007.
Spirit2 will debut at the Boost conference. It will be a complete parsing and output generation system that attempts to cover the whole spectrum from lexing to output generation.
Spirit2 will be ready before the Boost conference where Hartmut and I will give a talk titled "A cookbook approach to parsing and output generation with Spirit2". Dan Marsden and I will also have a talk on Boost.Fusion. Oh yeah, Eric Niebler will also give a talk on Text processing with Boost in general, and the much anticipated Proto library with which Spirit2 is built on.
I invite you all to come:
See the sessions here. You'll see people like Kevlin Henney, Scott Meyers, Sean Parent (Adobe ASL), Dave Abrahams, etc -- all these amazing people. It will be an amazing event -- the first ever Boost Conference! Register now to secure your place at BoostCon 2007. Space is limited.
It will be fun to meet up with "Spiritistas" :-) as one of the pioneers, JCAB, puts it. Let's have lods of fun and lotsa pizzas! And of course, not to forget beer!
Spirit 1.8.4 Released
October 29, 2006.
We just released Spirit v1.8.4. The files are available here:
https://sourceforge.net/project/showfiles.php?group_id=28447&release_id=459393
This is mainly a bugfix release.
Spirit 1.8.4 Release candidate + Roadmap
October 1, 2006.
Spirit 1.8.4 will be released in the next few days. You can get a beta snapshot here:
http://spirit.sourceforge.net/dl_more/spirit-1.8.4.beta1.zip
http://spirit.sourceforge.net/dl_more/spirit-1.8.4.beta1.tar.gz
These betas are provided to make sure that all is ok. We appreciate feedback and reporting of problems encountered. We'd like to make sure that the packages are in perfect shape before we do the final release.
Roadmap:
- Spirit 1.8.4 release
- Soon after the release of Spirit 1.8.4, Spirit 1.8.5 will follow. Spirit 1.8.5 is the bridge to Spirit-2. Spirit 1.8.5 integrates Fusion2 and Phoenix2. It is already in the final stages of development. Spirit 1.8.5 is not 100% backward compatible to Spirit 1.8.4. We shall provide some documents to make the transition as smooth as possible. The difference is minimal. Hopefully, 1.8.5 (and subsequent releases) will ease the transition to 2.0.
- Development of Spirit-2 shall proceed parallel to Spirit 1.8.5++.
We welcome comments and feedback.
Fusion
September 16, 2006. Fusion2 was recently accepted into Boost. See this link.
It's been a year. Most of the time was spent on infrastructure development for the planned Spirit2 development. This includes Phoenix2 and Fusion2 plus a documentation tool called Quickbook. These are major developments. Fusion2 will be the core infrastructure on which Spirit2 will be built. Phoenix2 is already built on top of Fusion2.
What is it? Fusion is a library for working with heterogenous collections of data, commonly referred to as tuples. Collectively the containers and views are referred to as sequences, and Fusion has a suite of algorithms that operate upon the various sequence types, using an iterator concept that binds everything together.
Special thanks to Ronald Garcia, and the review manager and to all the people in the boost community who participated in the review: Andreas Pokorny, Andreas Huber, Jeff Flinn, David Abrahams, Pedro Lamarao, Larry Evans, Ryan Gallagher, Andy Little, Gennadiy Rozental, Tobias Schwinger, Joao Abecasis, Eric Niebler, Oleg Abrosimov, Gary Powell, Eric Friedman, Darren Cook, Martin Bonner and Douglas Gregor.
New Web Page
August 26, 2005. What you are looking at now is a completely redesigned web page. Now that Boost has a brand new web page design, so should Spirit.
Spirit 1.8.3 and Spirit 1.6.3 Released
August 20, 2005. Spirit 1.8.3 and 1.6.3 released. You can download the files here and view the change log here. As far as 1.8.3 is concerned, this release contains exactly the same files as the Boost 1.33.0 release.
Boost 1.33.0
August 11, 2005. Boost 1.33.0 has been released. See Boost's Site. Boost 1.33.0 includes Spirit 1.8.3. Go get it. As usual, a stand-alone Spirit package with a Mini-Boost 1.33 (just enough to get Spirit going) will be released here.
Boost.Wave
February 23, 2005. Wave has been accepted into Boost!
With special thanks to Tom Brinkman, who volunteered to be the review manager.
With thanks to:
David Abrahams, Beman Dewes, Reece Dunn, Larry Evans, Doug Gregor, Joel de Guzman, Baptiste Lepilleur, Andy Little, Paul Mensonides, Dan Nuffer, Andreas Pokorny, Vladimir Prus, Gennadiy Rozental, Michiel Salters, Jonathan Turkanis, Chris Uzdavinis, Pavel Vozenilek, Michael Walter.
for bug reports, fixes and hints during the review process.