- Boost - lists.boost.org

[interprocess] managed_memory_impl.hpp
by Markus Schöpflin 23 Oct '07

23 Oct '07

This file contains at line 246: return instream.read(detail::char_ptr_cast(this->get_address()), (std::streamsize)this->get_size).good(); This most probably should read this->get_size(), but as no compiler besides cxx caught this error, I'm wondering if this code is ever used at all? Regards, Markus

2 1

For inclusion to boost::function
by Preston A. Elder 23 Oct '07

23 Oct '07

(repost, I accidentally posted this to boost::build list ... ugh). I have written code to solve this problem: I have class foo, like so: class foo { boost::function<void (foo *)> func_ptr; void call() { func_ptr(this); } }; I want func_ptr to be either a free-standing function (or class static), OR a class member function of a derived class, so: void bar(foo *f) { ... } OR: class baz : public foo { public: baz() { func_ptr = &baz::doit; } void doit() { } }; The obvious problem with the latter is that it needs to be downcast implicitly, and some kind of exception thrown if the wrong type is passed. I have created a boost::function proxy class to handle exactly this problem. It uses a lot of boost::function's own tricks with the pre-processor and such. I also had to use another suggestion I saw on this list to work around the fact boost::function does not accept member function pointer syntax (argh! That took a lot to figure out). My solution is this: http://www.neuromancy.net/fisheye/browse/mantra/trunk/Mantra-I/mantra/utils… Which is included by a boost preprocessor iteration here: http://www.neuromancy.net/fisheye/browse/mantra/trunk/Mantra-I/mantra/utils… As I mention, I use a lot of boost::function's own tricks. I think this would be a lot cleaner if done as part of boost::function, rather than me re-implementing the same macros and such. I have made sure my implementation works, and more importantly, has all the tests you would expect (ie. that the return value arguments except the class are the same, that the class is derived from the original function typedef's first argument, and so on). Here is my test case: http://www.neuromancy.net/fisheye/browse/mantra/trunk/Mantra-I/test/schedul… Any way, feedback is as always appreciated, but if someone could look at this for an addition to boost::function it would really be great PreZ

3 2

[system] Possible breakage: naming changes to conform to C++ std
by Beman Dawes 23 Oct '07

23 Oct '07

At the Kona meeting two weeks ago, the C++ committee accepted most of the Diagnostics changes proposed in N2415. See http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2415.html Most of Boost.System doesn't change since it already reflects N2415. The committee did, however, change the name of the nested "posix" namespace to "posix_error". For consistency, that means the "Linux" namespace changes to "linux_error" and the "windows" namespace changes to "window_error". As a consequence, the "linux_error" enum changes to "linux_errno". To minimize code breakage, I've added namespace aliases to error_code.hpp: # ifndef BOOST_SYSTEM_NO_DEPRECATED namespace posix = posix_error; # endif # ifndef BOOST_SYSTEM_NO_DEPRECATED namespace Linux = linux_error; # endif # ifndef BOOST_SYSTEM_NO_DEPRECATED namespace windows = windows_error; # endif That was sufficient to prevent any code breakage in the Boost.System and Boost.Filesystem test code, but any user code that actually mentions the linux_errno enum (rather than its constants) will have to be changed. --Beman

1 0

License in boostbook generated HTML pages
by John Maddock 23 Oct '07

23 Oct '07

I note we have a requirement - enforced by the inspect tool - for all HTML doc pages to display the boost license somewhere. However, quickbook/boostbook generated docs only place the license info on the first page. Is this something that needs fixing, or are our requirements too strict? While we're at it, it would be nice if the revision information was in the footer of each page as well :-) John.

1 0

[math] typo in special_functions/fpclassify.hpp rev. 40321
by Boris Gubenko 23 Oct '07

23 Oct '07

math library tests failed on HP-UX because of typo in math/special_functions/fpclassify.hpp rev. 40321. I've fixed it in rev. 40323. Thanks, Boris

2 1

[sunpro][patch] execution_monitor.ipp
by Simon Atanasyan 23 Oct '07

23 Oct '07

Sun C++ cannot compile execution_monitor.cpp on Solaris (Linux is ok). % CC -library=stlport4 -fast -KPIC -DBOOST_ALL_NO_LIB=1 -DBOOST_TEST_DYN_LINK=1 -DNDEBUG -I. -c libs/test/src/execution_monitor.cpp "./boost/test/impl/execution_monitor.ipp", line 177: Error: va_list is not defined. "./boost/test/impl/execution_monitor.ipp", line 178: Error: Cannot assign void* to int. "./boost/test/impl/execution_monitor.ipp", line 180: Error: The function "vsnprintf" must have a prototype. 3 Error(s) detected. There are two problems: 1. va_list is in the std namespace and do not introduced into the global namespace by default. 2. On Solaris vsnprintf is declared in the stdio.h only. Patch fixes that bug is attached to this mail. -- Simon Atanasyan

1 0

[test] --output_format parameter doesn't work?
by David Abrahams 23 Oct '07

23 Oct '07

setting BOOST_TEST_OUTPUT_FORMAT=XML gives me XML output. passing --output_format=XML on the command line doesn't. Gennadiy, can you explain? Thanks, -- Dave Abrahams Boost Consulting http://www.boost-consulting.com

2 2

[Test] Build failures on Mac OS X
by Doug Gregor 23 Oct '07

23 Oct '07

The Test library is failing to build on the Subversion trunk with the following error: darwin.compile.c++ ../../../../bin.v2/libs/test/build/darwin/debug/ link-static/execution_monitor.o ../../../../boost/test/impl/execution_monitor.ipp: In member function 'void boost::detail::system_signal_exception::report() const': ../../../../boost/test/impl/execution_monitor.ipp:426: error: 'SIGPOLL' was not declared in this scope ../../../../boost/test/impl/execution_monitor.ipp: In constructor 'boost::detail::signal_action::signal_action(int, bool, bool, char*)': ../../../../boost/test/impl/execution_monitor.ipp:527: error: expected id-expression before '(' token ../../../../boost/test/impl/execution_monitor.ipp: In constructor 'boost::detail::signal_handler::signal_handler(bool, int, bool, char*)': ../../../../boost/test/impl/execution_monitor.ipp:608: error: 'SIGPOLL' was not declared in this scope The definition of SIGPOLL in Mac OS X's signal.h is conditional: #if defined(_POSIX_C_SOURCE) #define SIGPOLL 7 /* pollable event ([XSR] generated, not supported) */ #else /* !_POSIX_C_SOURCE */ #define SIGIOT SIGABRT /* compatibility */ #define SIGEMT 7 /* EMT instruction */ #endif /* !_POSIX_C_SOURCE */ I'll be happy to test fixes, if that would help, but I'm too confused about the handling of SIGPOLL in execution_monitor.ipp to come up with my own patch. In one place, the use of SIGPOLL is guarded by ! defined(__CYGWIN__), in other places there doesn't seem to be such a guard. A little Googling notes that SIGPOLL is also missing on Cygwin, so I assume the Test library is broken on that platform, too. - Doug

2 2

borland failure
by Gennadiy Rozental 23 Oct '07

23 Oct '07

Does any one can help me to find workaround for borland, so this unit test http://tinyurl.com/2kxobo will work? Genandiy

1 0

Boost Logging Lib v2 - added documentation
by John Torjo 22 Oct '07

22 Oct '07

Hi all, I've added documentation for the Logging Lib v2 : http://torjo.com/log2/ (Logging homepage) http://torjo.com/log2/doc/html/ (documentation) Just to make sure, the Boost Log Library has these features: * A simple and clear separation of concepts o concepts are also easily separated into namespaces * A very flexible interface * You don't pay for what you don't use. * Fits a lot of scenarios: from very simple (dumping all to one log) to very complex (multiple logs, some enabled/some not, levels, etc). * Allows you to choose how you use logs in your code (by defining your own LOG_ macros, suiting your application) * Allows you to use Log levels (debug, error, fatal, etc). However this is an orthogonal concept - the library will work whether you use levels, categories or whatever , or not. * Efficient filtering of log messages - that is, if a log is turned off, the message is not processed at all * Thread-safe - the library allows you several degrees of thread-safety, as you'll see * Allows for formatters and destinations o formatters format the message (like, prepending extra information - an index, the time, thread is, etc) o destinations specify where the message is to be written o Formatters and Destinations are orthogonal to the rest of the library - if you want you can use them, otherwise you can define your own writing mechanism * Easy manipulation of the logs (turning on/off, setting formatters, destinations, etc) Feedback is most welcome - I'm sure there's lot to improve on the docs ;) Best, John -- http://John.Torjo.com -- C++ expert ... call me only if you want things done right

4 6