- Boost - lists.boost.org

Re: [boost] IOStreams formal review start (memory-mapped files)
by Jonathan Graehl 01 Sep '04

01 Sep '04

Generally: four stars, two thumbs up, and a partridge in a pear tree. I'm interested in directly using the Windows/POSIX mmap portability wrapper; might we promote it out of the detail namespace? That is, I may want to use mmap to perform some explicit memory swapping/caching. It looks like I could first create a file of the desired size, then acquire a mapped_file_resource - but then I don't get access to the actual pointer to memory (it's hidden in the private _pimpl). This is fine; what I really want is a portability wrapper for mmap :) Thanks, Jonathan

2 1

Problems with Boost.Random docs
by Daryle Walker 01 Sep '04

01 Sep '04

* BOOST_ROOT/libs/random/random-misc.html Mentions "generator_iterator" at the top, but there's nothing else about it. Does it have to be a secret that "random_number_generator" can now be accessed in its own header ("random_number_generator.hpp")? The web page still places it in "random.hpp" (which does include the previous header). I just noticed that "random.hpp" doesn't mention anything about a generator-iterator either. * BOOST_ROOT/libs/random/random-distributions.html Mentions that users that want an iterator interface should just put an adaptor over the existing generator interface. That sentence gives a link to the Iterator Adaptor library, instead of its own generator-iterator (which I guess is good in a way since it's MIA). There's a problem with this arrangement: * BOOST_ROOT/libs/iterator/doc/index.html Of its list of specialized adaptors, there's is NO generator adaptor! (There is an adaptor that goes the other way, i.e. output iterator to unary function, but we need generator to input iterator.) The sum up, it looks like the Random guys dumped their private generator adaptor for iterators, expecting the Iterator guys to replace it, but the latter guys never did. Unless the generator adaptor is hidden somewhere else, can someone add it (back). BTW, a generator would have to support a Boolean conversion where TRUE means it's still usable and FALSE means it isn't. (This is a new semantic for generators.) This would allow a reasonable way to implement the end-input-iterator. -- Daryle Walker Mac, Internet, and Video Game Junkie darylew AT hotmail DOT com

1 0

[type_traits] function_pointer_traits extension
by Tobias Schwinger 31 Aug '04

31 Aug '04

I'll continue this discussion (previously in private email) on the boost mailinglist, in order to allow others to join in. Any critiques (i.e. having a look at the (comparatively small) submission and dropping me a couple of lines) are still very welcome, in fact ! Alexander Nasonov wrote: >>>[... char array in struct as sizer for type classification via overload resultion ...] >>> >>Good point. This will be changed. >>By the way: the same hack is in the current boost release >>[boost/type_tratis/function_traits.hpp:177]. >> >> >> This has been changed [ref. 5.3.3]. >>> // tester returns a reference to an array >>> template< /* ... */ > >>> char (& tester( /* ... */ ) ) [size]; >>> >>> This would have been my personal preference. Unfortunately this causes bcc to fail with an 'Internal compiler error'. >>>or making it a meta-function >>> >>> template<std::size_t Size> >>> struct sized_type >>> { >>> typedef char (&type)[Size]; >>> }; >>> >>>and using typename sized_type<Size>::type in place of sized_type<Size>. >>> >>> >>> This is the current, but not necessarily the best implementation. Comments ? I believe this could be a common issue - for that matter: What about a type synthesizer like 'type_with_size<size_t>::type' or a 'BOOST_TYPE_WITH_SIZE(n)' macro ? Or even: BOOST_SIZED_RESULT_TYPE_OVERLOAD_PROTO(template_args,function_args,size_expression). Is there something like this in boost somewhere, already ? >>>The example compiles fine on g++ 3.2.2, g++ 3.4.0 and Intel 8.0 under linux. >>>I didn't check that BOOST_STATIC_ASSERT covers all cases but non-empty >>>output looks suspicious for metaprogramming library ;-) >>> >>> >>> The static assertion just aids keeping the errors where they belong. If it doesn't work there is a more obscure error message and one has to read the manual for proper usage. The documentation has been updated (Intel 8.0 is listed as tested and your work has been acknowledgeded). >>>I appreciate your efforts in boostifying function traits. Keep doing! >>> >>> >>I sure will ! >> >>Thanks a lot for your work, ideas and the very nice mail - despite the >>fact of having no time, that is ! >> >> > >Have you considered wrapping function signature into mpl sequence? > > Currently function_signature<T>::type is an mpl::vector and therefore a model of sequence. I believe I don't understand you correctly. In general, I aim to keep it as 'pure' as possible, so any conveniene feature should outweigh its cost in terms of code complexity , compilation time and maintainability. Tobias

3 6

Review Wizard Status Report
by tom brinkman 31 Aug '04

31 Aug '04

Review Wizard Status Report 1) Summary 2) Review Managers Needed 3) Review Schedule 4) Libraries under development //////////////////////////////////////////////////////////////////// 1) Summary //////////////////////////////////////////////////////////////////// Daryle Walker's library "More IO" is currently under review, starting August 21, 2004. Download the library at: http://boost-sandbox.sourceforge.net/more_io_4.zip. Please post a review of this library if you have not already done so. For information about submitting a Formal Review, see http://www.boost.org/more/formal_review_process.htm. The next library to be reviewed is Jonathan Turkanis "IOStreams" library. The "Iostreams" library allows the easy creation of standard C++ streams and stream buffer classes for new data sources and sinks. Download the library at: http://home.comcast.net/~jturkanis/iostreams. Be prepared to review this library in the next few days. If there are no objections, starting with the "More IO" library, I will attach an ID number to each reviewed library. Please reference this ID number in any postings to the group. This will help in organizing the reviews and assist in searching for related postings. I've also added a permanent section to this status report, called "Libraries Under Development". The purpose of this section is to prevent duplication of effort and support collaberation between library authors. Please contact me to add libraries to this section. Note: If you have any suggestions about how I could improve the Review Wizard's status report, please email me at reportbase(a)yahoo.com. //////////////////////////////////////////////////////////////////// 2) Review Managers Needed //////////////////////////////////////////////////////////////////// If you would like to volunteer to be a review manager, please contact me at reportbase(a)yahoo.com. The following libraries still require review manager volunteers: 10006 - Named Parms 10007 - Policy Pointer 10008 - Wave 10009 - Finite State Machine (FSM) 10010 - Fixed Strings //////////////////////////////////////////////////////////////////// 3) Review Schedule //////////////////////////////////////////////////////////////////// August 21, 2004 10001 - More IO (Currently Under Review) August 28, 2004 10002 - Iostreams September 7, 2004 10003 - Output Formatters September 15, 2004 10004 - Mini-Review wiki algorithims September 17, 2004 10005 - Smart Containers September 27, 2004 10006 - Named Params October 6, 2004 10007 - Policy Pointer October 16, 2004 10008 - Wave October 26, 2004 10009 - Finite State Machine (FSM) November 4, 2004 10010 - Fixed Strings 10001 - August 21, 2004 More IO (Currently Under Review) Author - Daryle Walker - darylew(a)hotmail.com Review Manager - Tom Brinkman - reportbase(a)yahoo.com Download: http://groups.yahoo.com/group/boost/files/more_io_4.zip [or] http://boost-sandbox.sourceforge.net/more_io_4.zip Description: Provides a multitude of I/O-oriented extensions. Some are extra manipulators; variants of "std::endl", a resetting manipulator, and a configurable manipulator. The library includes a set of base class templates for wrapping a new stream-buffer type around a stream class. Changes from the last interim version (called "more_io_3b.zip"): * split the manipulator header into several headers * refined the license statements in some already-affected files Changes from the first formal review (called "more_io.zip"): * refined the interface and implementation of array-streams and stream-buffer wrappers (including a wrapping example) * added the null-stream, value-stream, and pointer-stream headers * added generalizations to the new-line and skip-line manipulators * added the non-oriented 'reset' and 'form' manipulators * split the manipulator header into several headers * changed the license statements 10002 - August 28, 2004 Iostreams Author - Jonathan Turkanis - turkanis(a)kangaroologic.com Review Manager - Jeff Garland - jeff(a)crystalclearsoftware.com Web-Page: http://home.comcast.net/~jturkanis/iostreams Download: http://home.comcast.net/~jturkanis/iostreams Description: The Iostreams Library serves two main purposes: To allow the easy creation of standard C++ stream and stream buffer classes for new data sources and sinks. To provide a convenient interface for defining i/o filters and attaching them to standard streams and stream buffers. The library focuses on freeing users from writing boiler plate code and allowing them instead to create highly reusable components. In addition to providing an abstract framework the library provides a number of concrete filters, sources and sinks which serve as example applications of the library but are also useful in their own right. These include components for accessing memory-mapped files (based on work of Craig Henderson), for file access via operating system file descriptors, for code conversion, for text filtering with regular expressions, for line-ending conversion and for compression and decompression in the zlib, gzip and bzip2 formats. 10003 - September 7, 2004 Output Formatters Author - Reece Dunn - msclrhd(a)hotmail.com Review Manager - John Torjo - john(a)torjo.com Download: http://cvs.sourceforge.net/viewcvs.py/ boost-sandbox/boost-sandbox/boost/outfmt/ [or] http://cvs.sourceforge.net/viewcvs.py/ boost-sandbox/boost-sandbox/libs/outfmt/ Description: The I/O formatters library provides an easy and customizable way to input/output arrays, STL arrays, STL collections, sub-arrays from/to STL streams. It allows you to configure how each element in such an array will be written to the stream or read from the stream, while providing reasonable defaults. 10004 - September 15, 2004 Mini-Review wiki algorithims Author - Various Review Manager - Tom Brinkman - reportbase(a)yahoo.com Download: to be determined Description: not available 10005 - September 17, 2004 Smart Containers Author - Thorsten Ottosen - nesotto(a)cs.auc.dk Review Manager - Pavol Droba - droba(a)topmail.sk Download: the boost sandbox: ptr_container Description: The Smart Container Library consists of several container classes that are capable of storing heap-allocated objects in an exception-safe manner. The interface of the container classes effectively hides the underlying pointers so one can work on objects directly; this indirection is done to make the classes safer and easier to use. The library can also be customized to allow, for example, that a container does not rake ownership of the stored objects. 10006 - September 27, 2004 Named Params Co-Authors - David Abrahams - dave(a)boost-consulting.com) & Daniel Wallin (dalwan01(a)student.umu.se) Review Manager - to be determined Download: Boost sandbox boost/named_params.hpp [and] libs/utility/doc/named_params.html Description: This library is an attempt to address the problems of positional parameters by associating each parameter with a keyword identifier. Using this library, users can identify parameters by name instead of just argument position, and pass them in any order: window* w = new_window("alert2", movable = movability, opacity = 50); 10007 - October 6, 2004 Policy Pointer Author - David Held - dheld(a)codelogicconsulting.com Review Manager - to be determined Download: available via boost CVS Description: Smart pointers are used to automate memory management by handling the deletion of dynamically allcoated objects (and other resources). They assist in ensuring program correctness, exception safety, and memory integrity. Policy Pointer is a policy-based smart pointer framework designed to accomodate the large number of smart pointer designs. Through the use of policy classes, virtually any smart pointer type can be constructed within this framework. This library is a Boostification of the original Loki::SmartPtr type with significant modifications. 10008 - October 16, 2004 Wave Author - Hartmut Kaiser - hartmutkaiser(a)t-online.de Review Manager - to be determined Download Boost-sandbox Description: The Wave C++ preprocessor library is a Standards conformant implementation of the mandated C99/C++ preprocessor functionality packed behind a simple to use interface, which integrates well with the well known idioms of the Standard Template Library (STL). It uses the Spirit[4] parser construction library to implement a C++ lexer with SO/ANSI Standards conformant preprocessing capabilities. It exposes an iterator interface, which returns the current preprocessed token from the input stream. This preprocessed token is generated on the fly while iterating over the preprocessor iterator sequence (in the terminology of the STL these iterators are forward iterators). 10009 - October 26, 2004 Finite State Machine (FSM) Author - Andreas Huber - ah2003(a)gmx.net Review Manager - to be determined Download: http://boost-sandbox.sf.net/fsm.zip Documentation: http://boost-sandbox.sf.net/libs/fsm Description: The boost::fsm library is a framework that allows you to quickly transform a UML state chart into executable C++ code, without needing to use a code generator. Thanks to support for almost all UML features the transformation is straight-forward and the resulting C++ code is a nearly redundancy-free textual description of the state chart. 10010 - November 4, 2004 Fixed Strings Author - Reece Dunn - msclrhd(a)hotmail.com Review Manager - to be determined Download: boosts-sandbox/boost/fixed_string [or] boost-sandbox/libs/fixed_string Description: The fixed string library provides buffer overrun protection for static sized strings (char s[ n ]). It provides a C-style string interface for compatibility with C code (for example, porting a C program to C++). There is also a std::string-style interface using a class based on flex_string by Andre Alexandrescu with a few limitations due to the non-resizable nature of the class. //////////////////////////////////////////////////////////////////// 4) Libraries under development //////////////////////////////////////////////////////////////////// 10011 boost::xpressive - Under Development Author: Eric Niebler (eric(a)boost-consulting.com) Download: Boost sandbox at sourceforge.net Description: Advanced regular expression template library that allows regexes to be written as expression templates (statically bound) or as strings (dynamically bound) and allows regexes to nest and call each other recursively (like perl6) and bind to semantic actions (like Spirit). Development status: Under active development. Basic functionality in place. 10012 boost::geometry2d - Under Development Author(s): Aleksey Gurtovoy Description: simple cartesian geometry classes (point<>, rectangle<>, etc.) Download: yahoogroups files section, geometry2d/ Development status: basic functionality working Tom Brinkman Boost Review Wizard reportbase(a)yahoo.com

7 12

RE: [boost] Re: Re: Review of Daryle Walker's "More IO" librarybeginstoday, August 21, 2004.
by Reece Dunn 31 Aug '04

31 Aug '04

Jonathan Turkaniswrote: >"Rob Stewart" <stewart(a)sig.com> wrote in message >news:200408262114.i7QLEOP14442@lawrencewelk.systems.susq.com... > > From: "Jonathan Turkanis" <technews(a)kangaroologic.com> > > > - multi_skipper: I'm not sure I see why this is needed; how > > > often do you need to skip a sequence of repeated characters? Id rather > > > > When parsing, it is often useful to ensure that there were > > exactly 10 spaces between one token and the next, for example. > >I'll take your word for it, but an example would help. I am used to just >worrying about whitespace. Haskell uses whitespace to associate code on the right of the whitespace with the code on the left. e.g. [haskell.hs] factorial x = fimpl x where fimpl 0 = 1 fimpl x = fimpl ( x - 1 ) * x power (x:xs) = power xs * x power [] = 1 [end] where [] is an empty list and (x:xs) is a list with x as the head and xs as the tail. In this example, fimpl is associated with factorial and is only visible in its scope. factorial and power (both patterns) are in global scope. When writing a Haskell lexer, you need to be aware of the number of space characters after a given new-line. Therefore, something like multi_skipper would be useful, but it would need to: [1] reset an internal nspace variable on recieving a new-line character, allowing read access to it by the lexer; [2] incrementing by n (where n=3,4 or 8) upon recieving a tab character; [3] incrementing by 1 when recieving a space character I am not sure how the above could be implemented using multi_skipper, but this is an example where the number of spaces is important. Another example would be in padding. For example, the TAR file format ensures that each file fills a number of blocks of 256 bytes, so you need to skip the remaining bytes at the end of each file. Regards, Reece _________________________________________________________________ Want to block unwanted pop-ups? Download the free MSN Toolbar now! http://toolbar.msn.co.uk/

3 3

RE: [boost] Re: [Boost.Test] New testing procedure?
by Rozental, Gennadiy 31 Aug '04

31 Aug '04

> FYI, if there are no files in that directory of the > repository, cvs will generally not create the directory on > update/checkout. Not sure what's happening here, but you > might just have to add a dummy file to ensure it gets created. This directory have some files already. > That said, Boost tests should be able to be run from a > read-only CVS tree, so if you're creating files you probably > ought to do it elsewhere (like /tmp/...). Could you show how to do this portably? Gennadiy.

2 1

toolset proliferation
by Robert Ramey 31 Aug '04

31 Aug '04

I would like to see a little more order/organization in the creation of new toolsets. For example in the current CVS we have for Codewarrior Cw Cw-9.2 Cwpro-9.2 Cw-8.3 Cwpro-8.3 This is only an example. The same problem is occurring in gcc and mingw It creates a couple of problems a) we're not all testing with the same Jamfile. This leads to a lot of confusion. b) If I have a jam file which needs special treatment for a particular compiler, I have to included a line for each toolset - and that can change in the future creating another obscure bug to track down. c) The documentation on the "getting started" page is way out of sync with the toolsets in existences. A somewhat related problem - or maybe the cause of the above is the fact that the Jamfile doesn't have access to the compiler version. I'm not sure what to do about this. But, here's a suggestion for consideration to start the discussion. Right now we pass a variable to bjam with the toolset name. e.g bjam -sTOOLS=cwpro-9.2 How about the idea Bjam -sTOOLS=cw -sCOMPILER_VERSION=9.2 -sSTD_LIBRARY=stlport The "base" toolset would fill in variables COMPILER_VERSION and STL_LIBRARY_VERSION with defaults - probably the currently shipping version. Or a "base" toolset could require one or the other of the above switches. Also the "base" toolset could include a "config_check" target what would double check that compiler version invoked is the same as the one that is specified by the switch. This would catch the case where one has one compiler in his path and tries to build for another. This would have the following advantages: a) a smaller number of toolsets b) a greater likelihood that we're all testing the same thing c) documentation more likely to be insync with the actual toolsets available The recently re-organized stlport.jam "add-in" with all stlport stuff factored out so it can be "added-in" to any library is a good idea. Unfortunately it has a bug but that's fixable. This would allow us to specify usage of STLPort (or any other library) without creating a new toolset. I guess we sometimes use: bjam "-sTOOLS=cwpro-9.2 msvc" to handle multiple compilers in one shot. If its important to keep this feature then we could use Bjam "-sTOOLS=cw msvc" "-sCOMPILER_VERSION=cw-9.2 msvc-6.0" -sSTD_LIBRARY=cw-stlport or something similar. Just a small suggestion to keep the pot boiling. Robert Ramey

2 1

[multi_index] Question about 'modify'
by Vladimir Prus 31 Aug '04

31 Aug '04

I have a question about the 'modify' method of ordered indexes in the multi_index library. I have an iterator obtained from an ordered index and would like to modify the pointed element. I'm pretty sure that the modification won't change the order, won't create collisions and so on. The documentation suggests to use either 'replace', or more efficient 'modify' method. Why can't I do the following: typedef variables_t::nth_index<1>::type index_t; index_t& index = variable_declarations.get<1>(); index_t::iterator existing = index.find(vd); const_cast<Variable_declaration&>(*existing) .make_external(); ? TIA, Volodya

2 3

user defined facets in date_time library
by Markus Schöpflin 31 Aug '04

31 Aug '04

While trying to make date_time tests compilable on tru64cxx65, I have come accross an issue with user defined facets (namely date_names_put and all_date_names_put) that leaves me scratching my head. Is a user defined facet required to be default constructible? Or does it need a constructor of the form "foo(size_t refs = 0) : locale::facet(refs) {}"? [lib.facets.examples] doesn't say so but most examples I could find either have a default constructor or a constructor taking a reference count. Markus

2 3

copyright question (bjam examples)
by John Maddock 31 Aug '04

31 Aug '04

Some of the examples in bjam v2 are actually taken verbatim from the QT libraries examples, and currently have no copyright or license declaration on them at all. Vladimir asks: "Will /* This file is part of Qt and is available under the Q Public License, available from http://doc.trolltech.com/3.3/license.html */ be enough to make inspect tool happy. If we're really going to be extra-picky, I don't think I can add any "Copyright" string to the file I don't own." Is adding such a string acceptable? Clearly we need to do something, but what? Thanks, John.

4 4