- Boost - lists.boost.org

Re: Points vs Vectors / GUI Library Proposal for a Proposal)
by George van den Driessche 08 Nov '04

08 Nov '04

Noah Stein wrote: > >> This makes them almost the same thing. You can treat vector as a >> point, and point as a vector. Besides, we are talking about R^2, >> which is a vector space. > > Sure, there's an isomorphism between points and vectors; however, that > doesn't make them the same. There's an isomorphism between the > functions > f(x)=x and f(x)=2x; however, that does not make those two equations the > same. This could be explained by analogy with C++, which more readers here might be familiar with: the difference between points and vectors is the same as the difference between C++ pointers and C++ ints. You can add a pointer to an int, or an int to a pointer, but you can't add two pointers. However, you can subtract two pointers to get an int. Likewise, it's not really meaningful to talk about the sum of two positions, but the difference between them is the vector from one to the other. Back on the subject of a GUI library: I understand the motivation - that's where wxWindows arose. And I realise that nowadays a GUI library could be done in a much more modern C++ style. But my question is: is C++ the right language? I don't want this to descend into a religious war - C++ is my favourite language, but it isn't the best language for every task. Would a higher-level, more dynamic language be more suitable for GUI work? In a perfect world, I'd do all my back-end stuff in C++ and all the front-end development in, say, Python. By doing all the GUI work in C++ instead of an interpreted language, you: a) spend a lot more time recompiling. b) can less easily tweak a running program to try out different ideas. c) have to spend a lot of effort to construct a domain-specific language (c.f Spirit; see (a)). d) probably end up with a bigger codebase. So what do you stand to gain, other than being able to write GUIs in your favourite language, as opposed to your second-favourite language? The only thing I can think of is that you don't have to worry about any bridge between languages. George

3 2

OOPSLA Workshop...
by Tommy 08 Nov '04

08 Nov '04

Hi, a couple of months ago, I noticed an announcement in this list for Boost’s Oopsla Workshop. The topics and goals of that workshop (review position papers / libraries, discuss new library proposals, their relative merits, benefits and weaknesses, and provide feedback on initial designs) looked very interesting. So, after that event took place (Oct 25), I’ve looked out for material describing the results of that workshop - but I have not been able to find anything!? Has the results of the workshop been put together? If so, is the outcome publically available somewhere? Regards, Curious Tommy

2 1

boost::ref simple question
by Neal D. Becker 08 Nov '04

08 Nov '04

This must be a common question. I want to use std::transform with a function object that is expensive to copy, so I want to pass by reference. I tried to use boost::ref for this, but this simple test doesn't work: struct Op { int operator() (int x) { return 1 + x;} Op() {} private: Op (const Op&); }; int main() { vector<int> x (4); vector<int> y (4); Op o; std::transform (x.begin(), x.end(), y.begin(), boost::ref(o)); } Any suggestion?

1 0

[regex] any interest in finite automata-based regex engine?
by Vladimir Pozdyayev 08 Nov '04

08 Nov '04

Hello. Are there any plans to add finite automata-based subengine to the regex library? If there are, I can submit the algorithm/implementation of a modified nondeterministic FA which takes care of * greedy and lazy quantifiers, * eagerness of the alternation operator, * non-longest matches (`best' in the Perl-ish sense with respect to greediness/laziness/eagerness), * substring capturing. All these features being included, the time cost of a match is O(N(E+C)L), where N is the number of states, E is the average epsilon-closure size per state, C is the number of capturing parentheses, and L is the input length. As far as I know that's the lowest order of time complexity in this area (corrections are welcome)---though, of course, deterministic FA would be even faster. I suppose there's no point in adding another regex library, but being able to guarantee linear time cost in an existing one would be nice. -- Best regards, Vladimir Pozdyayev.

6 22

Program options auto-link bug
by John Maddock 08 Nov '04

08 Nov '04

The program options library currently doesn't select the dll-import library when auto-linking to the dll build of the library - it always tries to link to the static lib. The following patch fixes the issue: $ cvs diff -u boost/program_options/config.hpp johnmaddock(a)cvs.sourceforge.net's password: Index: boost/program_options/config.hpp =================================================================== RCS file: /cvsroot/boost/boost/boost/program_options/config.hpp,v retrieving revision 1.2 diff -u -r1.2 config.hpp --- boost/program_options/config.hpp 13 Jul 2004 15:12:24 -0000 1.2 +++ boost/program_options/config.hpp 6 Nov 2004 12:37:09 -0000 @@ -20,6 +20,10 @@ // Set the name of our library, this will get undef'ed by auto_link.hpp // once it's done with it: #define BOOST_LIB_NAME boost_program_options +// tell the auto-link code to select a dll when required: +#if defined(BOOST_ALL_DYN_LINK) || defined(BOOST_PROGRAM_OPTIONS_DYN_LINK) +# define BOOST_DYN_LINK +#endif // And include the header that does the work: #include <boost/config/auto_link.hpp> OK to commit? John.

3 5

[multi_index] regression in gcc-3.1-linux --seeminlgy MPL-related
by JOAQUIN LOPEZ MU?Z 08 Nov '04

08 Nov '04

Currently, one of the tests of Boost.MultiIndex is failing for GCC 3.1 under Linux: http://tinyurl.com/638oh The ICE happens inside boost/mpl/aux_/at_impl.hpp. This test used to work OK against the RC branch a couple of days ago. Any idea? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 2

Auto-linking and getting started docs.
by John Maddock 08 Nov '04

08 Nov '04

I've added a section on auto-linking with MS-Windows compilers in the getting started docs, in the main branch only (No URL yet, cvs hasn't cycled it through yet, but it'll be at http://cvs.sourceforge.net/viewcvs.py/*checkout*/boost/boost/more/getting_s… eventually). If it looks OK, I'll commit to the release branch as well. John.

5 6

Re: [boost] [multi_index] regression in gcc-3.1-linux --seeminlgyMPL-related
by JOAQUIN LOPEZ MU?Z 08 Nov '04

08 Nov '04

----- Mensaje original ----- De: Martin Wille <mw8329(a)yahoo.com.au> Fecha: Domingo, Noviembre 7, 2004 7:15 pm Asunto: Re: [boost] [multi_index] regression in gcc-3.1-linux -- seeminlgyMPL-related > JOAQUIN LOPEZ MU?Z wrote: > > Currently, one of the tests of Boost.MultiIndex > > is failing for GCC 3.1 under Linux: > > > > http://tinyurl.com/638oh > > > > The ICE happens inside boost/mpl/aux_/at_impl.hpp. > > > > This test used to work OK against the RC branch > > a couple of days ago. Any idea? > > Neither CVS head, nor the RC branch have been tested > against gcc 3.1 under Linux. I (re)added this compiler > because I had spare CPU time. > > I don't think gcc 3.1 is something one has to worry > about too much. I doubt it is much in use. (Either > people still stick to the even more broken gcc 3.1 > or they use gcc 3.3 or even 3.4) You're absolutely right, even my docs state that GCC 3.1 is not supported by Boost.MultiIndex. Sorry for the noise. Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

1 0

RE: [boost] Re: GUI Library Proposal for a Proposal
by Reece Dunn 08 Nov '04

08 Nov '04

Russell Hind wrote: >Edward Diener wrote: >> >>No !!!!! WxWidgets has almost nothing to do with modern C++. Let's no go >>there, or have anything to do with supporting any other GUI >>implementation. > >Also, wxWindows is a huge library. If we were to introduce something in >boost, I'd very much like it to be as small and light weight (as win32gui >is supposed to be), not a collection of multi-megabyte dlls required, just >to wrap win32 or what ever the underlying library is. I am in the process of writing a platform-independant GUI (available at: http://uk.geocities.com/msclrhd/gui/gui.zip) that aims to be small and lightweight. This code is stll in the early stages and currently targets Win32/64 with minimal Cocoa/PalmOS support. >If people want more powerful components, they can be added afterwards, but >the underlying library around core OS controls should be as small as >possible. In my library, I am working on: * gui::event_handler supporting lightweight "windowless" controls * gui::window supporting windows like the main frame, MDI child frames, etc. * gui::form (i.e. a frame with controls/layout set by an external designer referenced by ID) that includes dialog boxes in Win32. I do not have an implementation of gui::form yet. * gui::control supporting controls/widgets like buttons within a window or form These 4 classes should then make it possible to derive your own classes to support edit controls, property sheets, MDI windows, etc. I do not explicitly implement any of these extra UI components because they are platform dependant and will add to the complexity of the framework. This is especially true for things like "splitter" windows and floating/docking toolbars. Also, other components can be added at a later stage if necessary. As for the event processing mechanism, I have decided to use a std::map< event_id, void ( gui::event_handler::* )( gui::event * ) > that allows: class main_frame: public gui::window { public: void ev_destroy( gui::event * ); void EvDraw( gui::event * ); main_frame( gui::string name ): gui::window( ... ) { set_handler( WM_DESTROY, &main_frame::ev_destroy ); set_handler( WM_PAINT, &main_frame::EvDraw ); } }; At the moment, this does not support chaining of handlers (something on my todo list). The main event loop is controlled via gui::event_loop that does support filtering of events (e.g. processing keyboard shortcuts). There are also platform-independant versions of position (NSPoint, POINT, etc.), size (NSSize, PointType, etc.) and area (NSRect, RECT, etc.) that make use of properties allowing window positions to be handled in a platform-independant manner, e.g.: void main_frame::resizing( gui::event * ) { area a = get_client_area(); a.deflate( gui::size( 5, 5 )); pane -> move( a ); } At the moment, you need to register Win32/64 window classes using: gui::win::window_type frame_class( TEXT( "CxxGUIFrame" ), COLOR_BTNFACE ); but I am looking to abstract this. I am all for getting a group of people together from the boost mailing list interested in GUI and developing a general GUI framework. It may be useful to have at least one person from the C++ standards committee to get input from the standards people. Regards, Reece _________________________________________________________________ Want to block unwanted pop-ups? Download the free MSN Toolbar now! http://toolbar.msn.co.uk/

10 26

how to compile regression tests
by Roland Schwarz 07 Nov '04

07 Nov '04

Sorry for this possibly naive question. But I can't figure out how to compile the regression tests. So far I found cd'ing into a specific libs/test directory and invoking bjam on it compiles the test for a specific library. But how can I compile all? I cannot find a rule like stage or install in the root Jamfile. Isn't it possible to recurse through all directories? Then I cannot manage to get a results page using the description in http://www.boost-consulting.com/boost/more/regression.html Instead I get: > C:\Programme\boost>SET PYTHON_ROOT=C:\Programme\Python23 > > C:\Programme\boost>SET BOOST_ROOT=C:\Programme\boost > > C:\Programme\boost>SET TOOLS=vc-7_1 > > C:\Programme\boost>ECHO ON > > C:\Programme\boost>cd C:\Programme\boost\status > > C:\Programme\boost\status>bjam --dump-tests test 1>bjam.log 2>&1 > > C:\Programme\boost\status>REM start notepad bjam.log > > C:\Programme\boost\status>process_jam_log 0<bjam.log > Usage: bjam [bjam-args] | process_jam_log [--echo] [locate-root] > locate-root is the same as the bjam ALL_LOCATE_TARGET > parameter, if any. Default is boost-root. > boost_root: C:/Programme/boost/status/.. > locate_root: C:/Programme/boost/status/.. > no errors detected > > C:\Programme\boost\status>compiler_status C:\Programme\boost cs-win32.html > > **** exception(205): std::exception: > boost::filesystem::directory_iterator constructor: > "C:\Programme\boost\status\bin": > The system cannot find the path specified. > ******** errors detected; see standard output for details ******** > > C:\Programme\boost\status>rem Specify links file, even though it will > be overwritten, so report html includes links > > C:\Programme\boost\status>rem This works because the generated > bookmark names are the same regardless of other settings > > > C:\Programme\boost\status>compiler_status --ignore-pass --no-warn > C:\Programme\boost cs-win32-fail.html cs-win32-links.h > tml > > **** exception(205): std::exception: > boost::filesystem::directory_iterator constructor: > "C:\Programme\boost\status\bin": > The system cannot find the path specified. > ******** errors detected; see standard output for details ******** > > C:\Programme\boost\status>compiler_status --ignore-pass > C:\Programme\boost cs-win32-warn-or-fail.html cs-win32-links.htm > l > > **** exception(205): std::exception: > boost::filesystem::directory_iterator constructor: > "C:\Programme\boost\status\bin": > The system cannot find the path specified. > ******** errors detected; see standard output for details ******** > > C:\Programme\boost\status>compiler_status C:\Programme\boost > cs-win32-full.html cs-win32-links.html > > **** exception(205): std::exception: > boost::filesystem::directory_iterator constructor: > "C:\Programme\boost\status\bin": > The system cannot find the path specified. > ******** errors detected; see standard output for details ******** > > C:\Programme\boost\status> I am using vc7.1 on W2 with the latest CVS MAIN. I currently assume this should work under windows, doesn't it? Thank you for help, Roland

1 0