- Boost - lists.boost.org

[iostreams] Build Message Typo
by David Abrahams 30 Sep '05

30 Sep '05

It looks like a semicolon was omitted just before "ECHO" Building Boost.Iostreams with bzip2 support disabled. ECHO To enable bzip2, consult the Boost.Iostreams documentation -- Dave Abrahams Boost Consulting www.boost-consulting.com

2 1

[filesystem] GetFileAttributesExA and Windows 95
by Angus Leeming 29 Sep '05

29 Sep '05

Beman, you use the WinAPI call GetFileAttributesExA only twice in Boost.Filesystem, in _is_empty and file_size, but that means that the code won't run on Windows 95: file is linked to missing export KERNEL32.DLL:GetFileAttributesExA The msdn docs http://msdn.microsoft.com/library/en-us/fileio/fs/getfileattributesex.asp suggest that Win95 support requires only that you grab NewAPIs.h from the SDK Update Site. Here's an example of how it should be used: ========================================================================== http://codeproject.com/win32/harddiskspace.asp?df=100&forumid=15107&exp=0&s… #define COMPILE_NEWAPIS_STUBS #define WANT_GETDISKFREESPACEEX_WRAPPER then in the file where you use GetDiskFreeSpaceEx.... #include <newapis.h> ========================================================================== Alternatively, I append a naïve implementation of the two files using GetFileAttributes and GetFileSize. Is any of this useful? Regards, Angus BOOST_FILESYSTEM_DECL bool _is_empty( const path & ph ) { DWORD const attributes = GetFileAttributes( ph.string().c_str() ); if ( attributes & INVALID_FILE_ATTRIBUTES ) boost::throw_exception( filesystem_error( "boost::filesystem::is_empty", ph, fs::detail::system_error_code() ) ); if ( attributes & FILE_ATTRIBUTE_DIRECTORY ) return is_empty_directory( ph ); DWORD file_size_high = DWORD const file_size_low = ::GetFileSize( ph.string().c_str(), &file_size_high ); if ( file_size_low == INVALID_FILE_SIZE ) { if ( GetLastError() = NO_ERROR ) boost::throw_exception( filesystem_error( "boost::filesystem::is_empty", ph, fs::detail::system_error_code() ) ); return ( !file_size_high && !file_size_low ); } BOOST_FILESYSTEM_DECL boost::intmax_t file_size( const path & ph ) { DWORD const attributes = GetFileAttributes( ph.string().c_str() ); if ( attributes & INVALID_FILE_ATTRIBUTES ) boost::throw_exception( filesystem_error( "boost::filesystem::is_empty", ph, fs::detail::system_error_code() ) ); if ( attributes & FILE_ATTRIBUTE_DIRECTORY ) boost::throw_exception( filesystem_error( "boost::filesystem::file_size", ph, "invalid: is a directory", is_directory_error ) ); DWORD file_size_high = DWORD const file_size_low = ::GetFileSize( ph.string().c_str(), &file_size_high ); if ( file_size_low == INVALID_FILE_SIZE ) { if ( GetLastError() = NO_ERROR ) boost::throw_exception( filesystem_error( "boost::filesystem::is_empty", ph, fs::detail::system_error_code() ) ); return (static_cast<boost::intmax_t>( file_size_high ) << (sizeof( file_size_low )*8)) + file_size_low }

4 11

(no subject)
by Paul A Bristow 29 Sep '05

29 Sep '05

I had been using Boost test with 1.32 successfully #include <boost/test/test_tools.hpp> #include <boost/test/floating_point_comparison.hpp> // Extra test tool for FP comparison. but with 1.33 it fails to compile at using boost::unit_test_framework::unit_test_log; using boost::unit_test_framework::log_level; and used this to control the log threshold unit_test_log::instance().set_log_threshold_level_by_name("test_suite"); and to switch the log file unit_test_log::instance().set_log_stream(flog); // Switch to log file.// unit_test_log::instance().set_log_stream(cout); // Switch back to cout, But this has been removed in 1.33 but I am unclear how to replicate this sort of function. Suggestions? Thanks Paul Paul A Bristow Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB +44 1539 561830 +44 7714 330204 mailto: pbristow(a)hetp.u-net.com www.hetp.u-net.com

2 1

[boostbook] BOOST_ROOT vs BOOST_BUILD_PATH
by Martin Wille 29 Sep '05

29 Sep '05

Hi, the installation script for Boostbook contains: elif test -r "$BOOST_ROOT/tools/build/v2/user-config.jam"; then JAM_CONFIG_IN="$BOOST_ROOT/tools/build/v2/user-config.jam"; Shouldn't BOOST_BUILD_PATH be checked, too? Regards, m Send instant messages to your online friends http://au.messenger.yahoo.com

2 5

OT - to Mr Daryle Walker
by Robert Kawulak 29 Sep '05

29 Sep '05

Dear Mr Daryle Walker, Did you get my e-mail about improvements to integer.hpp? I haven't got any response from you, maybe my message was rejected by some automatic filter? Best regards, Robert ---------------------------------------------------------------------- Tak niesamowity CZAT... >>> http://link.interia.pl/f18ba

1 0

Unit test framework with 1.33 - logging
by Paul A Bristow 29 Sep '05

29 Sep '05

Sorry - reposted with subject :-(( | -----Original Message----- | From: boost-bounces(a)lists.boost.org | [mailto:boost-bounces@lists.boost.org] On Behalf Of Paul A Bristow | Sent: 28 September 2005 19:41 | To: boost(a)lists.boost.org | Subject: [boost] (no subject) | | I had been using Boost test with 1.32 successfully | | #include <boost/test/test_tools.hpp> | #include <boost/test/floating_point_comparison.hpp> // Extra | test tool for | FP comparison. | | but with 1.33 it fails to compile at | | using boost::unit_test_framework::unit_test_log; | using boost::unit_test_framework::log_level; | | and used this to control the log threshold | | | unit_test_log::instance().set_log_threshold_level_by_name("tes | t_suite"); | | and to switch the log file | | unit_test_log::instance().set_log_stream(flog); // Switch to log | file.// unit_test_log::instance().set_log_stream(cout); | // Switch back to | cout, | | But this has been removed in 1.33 | | but I am unclear how to replicate this sort of function. | | Suggestions? | | Thanks | | Paul | | Paul A Bristow | Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB | +44 1539 561830 +44 7714 330204 | mailto: pbristow(a)hetp.u-net.com www.hetp.u-net.com | | | | _______________________________________________ | Unsubscribe & other changes: | http://lists.boost.org/mailman/listinfo.cgi/boost |

2 2

Boost regression notification (2005-09-29 [RC_1_33_0])
by Douglas Gregor 29 Sep '05

29 Sep '05

Boost regression test failures ------------------------------ Report time: 2005-09-29T14:30:10Z This report lists all regression test failures on release platforms. Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-RC_1_33_0/developer/i… 77 failures in 7 libraries: foreach (1) graph (4) iostreams (12) random (1) range (3) serialization (55) spirit (1) |foreach| rvalue_const: vc-7_1 |graph| betweenness_centrality_test: gcc-3.4.4-linux dag_longest_paths: gcc-3.4.4-linux dijkstra_cc: gcc-3.4.4-linux relaxed_heap_test: gcc-3.4.4-linux |iostreams| gzip_test: cw-9_5-darwin gcc-2.95.3-stlport-4.6.2-linux gcc-3.2.3-linux gcc-3.3.6-linux gcc-3.4.4-linux gcc-3_3-darwin gcc-4.0.1-linux gcc-4_0-darwin intel-8.1-linux intel-9.0-linux mingw-3_4_2 vc-7_1 |random| random_test: gcc-3.3.6-linux |range| array: vc-6_5 vc-6_5-stlport vc-7_0 |serialization| test_exported_binary_archive: vc-6_5 test_exported_binary_archive_dll: vc-6_5 test_exported_text_archive: vc-6_5 test_exported_text_archive_dll: vc-6_5 test_exported_text_warchive: vc-6_5 test_exported_text_warchive_dll: vc-6_5 test_exported_xml_archive: vc-6_5 test_exported_xml_archive_dll: vc-6_5 test_exported_xml_warchive: vc-6_5 test_exported_xml_warchive_dll: vc-6_5 test_no_rtti_binary_archive: vc-6_5 test_no_rtti_binary_archive_dll: vc-6_5 test_no_rtti_text_archive: vc-6_5 test_no_rtti_text_archive_dll: vc-6_5 test_no_rtti_text_warchive: vc-6_5 test_no_rtti_text_warchive_dll: vc-6_5 test_no_rtti_xml_archive: vc-6_5 test_no_rtti_xml_archive_dll: vc-6_5 test_no_rtti_xml_warchive: vc-6_5 test_no_rtti_xml_warchive_dll: vc-6_5 test_shared_ptr_132_binary_archive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_binary_archive_dll: vc-6_5 test_shared_ptr_132_text_archive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_text_archive_dll: vc-6_5 test_shared_ptr_132_text_warchive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_text_warchive_dll: vc-6_5 test_shared_ptr_132_xml_archive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_xml_archive_dll: vc-6_5 test_shared_ptr_132_xml_warchive: vc-6_5 vc-6_5-stlport test_shared_ptr_132_xml_warchive_dll: vc-6_5 test_shared_ptr_binary_archive: borland-5_6_4 vc-6_5 test_shared_ptr_binary_archive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_text_archive: borland-5_6_4 vc-6_5 test_shared_ptr_text_archive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_text_warchive: borland-5_6_4 vc-6_5 test_shared_ptr_text_warchive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_xml_archive: borland-5_6_4 vc-6_5 test_shared_ptr_xml_archive_dll: borland-5_6_4 vc-6_5 test_shared_ptr_xml_warchive: borland-5_6_4 vc-6_5 test_shared_ptr_xml_warchive_dll: borland-5_6_4 vc-6_5 |spirit| parametric_tests: gcc-3_4_3-sunos

1 0

Boost regression notification (2005-09-29 [HEAD])
by Douglas Gregor 29 Sep '05

29 Sep '05

Boost regression test failures ------------------------------ Report time: 2005-09-29T05:00:18Z This report lists all regression test failures on release platforms. Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-HEAD/developer/issues… 123 failures in 9 libraries: concept_check (12) graph (2) mpl (1) parameter (12) serialization (33) spirit (8) thread (4) tuple (1) typeof (50) |concept_check| stl_concept_covering: cw-9_4 gcc-2.95.3-linux gcc-2.95.3-stlport-4.6.2-linux gcc-3.3.6-linux gcc-3.4.4-linux gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin intel-win32-9_0 intel-win32-9_0 mingw-3_4_2 vc-7_1 |graph| graphviz_test: vc-7_1 layout_test: intel-win32-9_0 |mpl| size_t: gcc-4_0-darwin |parameter| basics: gcc-2.95.3-linux gcc-2.95.3-stlport-4.6.2-linux efficiency: gcc-2.95.3-linux gcc-2.95.3-stlport-4.6.2-linux gcc-4_0-darwin macros: gcc-2.95.3-linux gcc-2.95.3-stlport-4.6.2-linux sfinae: gcc-2.95.3-linux gcc-2.95.3-stlport-4.6.2-linux unnamed: gcc-2.95.3-linux gcc-2.95.3-stlport-4.6.2-linux vc-7_1 |serialization| test_class_info_load_binary_archive: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_binary_archive_dll: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_text_archive: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_text_archive_dll: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_text_warchive: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_text_warchive_dll: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_xml_archive: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_xml_archive_dll: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_xml_warchive: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_class_info_load_xml_warchive_dll: gcc-3_3-darwin gcc-3_4_3-sunos gcc-4_0-darwin test_set_xml_archive: vc-7_1 test_set_xml_warchive: vc-7_1 test_simple_class_ptr_text_warchive_dll: vc-7_1 |spirit| action_tests_debug: gcc-3_4_3-sunos distinct_tests_debug: gcc-3_4_3-sunos escape_char_parser_tests_debug: gcc-3_4_3-sunos grammar_def_test_debug: gcc-3_4_3-sunos switch_problem_debug: gcc-3_4_3-sunos switch_tests_single_debug: gcc-3_4_3-sunos traverse_tests: intel-win32-9_0 traverse_tests_debug: intel-win32-9_0 |thread| test_condition: gcc-3_4_3-sunos test_mutex: gcc-3_4_3-sunos test_tss: gcc-3_4_3-sunos gcc-3_4_3-sunos |tuple| tuple_test_bench: gcc-3_4_3-sunos |typeof| data_member_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 function_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 function_ptr_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 function_ref_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 lvalue_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 member_function_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 modifiers_emulation: cw-9_4 modifiers_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 noncopyable_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 std_emulation: cw-9_4 std_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 template_dependent_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 template_enum_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 template_int_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 template_multiword_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 template_tpl_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 template_type_native: cw-9_4 intel-win32-9_0 intel-win32-9_0 type_native: cw-9_4 intel-win32-9_0 intel-win32-9_0

1 0

[Test] and [bjam] questions re DLLS
by Robert Ramey 29 Sep '05

29 Sep '05

I am making a test program in order to test the operation of the serialization library when serializable type is serialized from within a dynamically loaded DLL. I've made a DLL which serializes to a polymorphic_?archive. I'm going to make a test program which calls the DLL. Here are my questions: [Test] My DLL has included <boost/test/test_tools.hpp> and test tool macros. Things compile fine. When I try to link I get a failure because I've got not test_main - I'm using the Test Execution Monitor. Now that I think about it - its not clear at all how to address this. Suppose the above is resolved. I run my test and I get a failure in the DLL, will the exception propagate across the DLL interface into the main executable? I would hope so buf I don't know for sure. [bjam] Assuming the above issues are resolved I will have two program modules: test_dll_main and test_dll_A (dll to serialize class A). Not each bjam target lands in its own directory of the form bin/boost/libs/serialization/<test_module_name>.test/<compiler_name>/debug/threading-multi/test_module_name.exe I would guess that the dll lands in bin/boost/libs/serialization/<test_dll_name>.test/<compiler_name>/debug/threading-multi/test_dll_name.dll So in my main test module I would need something like dlopen or LoadLibrary ("bin/boost/libs/serialization/<test_module_name>.test/<compiler_name>/debug/threading-multi/test_module_name.exe") which is going to be pretty inconvenient or make sure that the test_dll is placed in the same target directory as test_main - which I don't know how to make bjam do. Is there any information on how to go about building and testing DLLS/Shared Libraries? Robert Ramey

3 3

Re: [boost] GUI
by Greer, Joe 29 Sep '05

29 Sep '05

> Thats a good point, but is that not convention only? Its not required > behaviour. > Suppose I am on an XWindows system and I want MSWindows behaviour. That > point > about locale is a good one. A sequence of user actions must map to a > state-machine that finally maps the sequence of clicks into a particular > function such as Maximise, Minimise, etc. Maybe its possible to load the > map If you want to have your application certified or branded on a particular platform, then you must follow the style guidelines. Most of the guidelines aren't written in stone, but gratuitous differences will prevent you from getting your app branded. For many, this is important. I think that any GUI should do its best to isolate the app logic from the presentation style and allow for OS specific GUI behaviors to be isolated away from the main program logic in order to facilitate easy porting to various OS-s rather than trying too hard to make it seamless, but vanilla. For example, something a simple as prompting for a file, Windows provides an API call for this that puts up the appropriate dialog box to navigate to the file and select it. The box may be different depending upon which version of Windows the app is running on, but the application doesn't have to care and they will know that the boxes are arranged in the appropriate order, have the right tab stops and shortcuts etc. Every ported app I have seen gets this wrong and it is amazing how jarring that is. I imagine that Mac users would feel the same, if they were prompted with a box that doesn't meet the Mac guidelines. This kind of thing is important for commercial apps. I am not sure how to accomplish this separation of app logic vs GUI logic, but it needs done if you want it used on non-unix systems. A GUI that is tailored to allow unix developers to port to Windows and Mac gets a big yawn from me. I would like a system that encourages developers from every platform to write their apps in a way that allows it to work everywhere and to do that, things need to be able to follow different style guidelines. I'm rambling now, so I will quit. :) Joe

1 0