- Boost - lists.preview.boost.org

[regression][testing][config] Regression results not displayed.
by Adam Wulkiewicz 11 May '15

11 May '15

Hi, Since 2015/04/26 the tests started to dissapear from Geometry summary pages for develop branch. And now the pages aren't even displayed properly, I see only: Error extracting file: No matching files were found. for: http://www.boost.org/development/tests/develop/developer/geometry.html http://www.boost.org/development/tests/develop/developer/geometry-extension… and on: http://www.boost.org/development/tests/develop/developer/geometry-index.html there are only 2 tests. Furthermore "geometry" and "geometry-extensions" can't be found in the menu. I can see the results for Geometry in the paskages sent to the server. I can also see them in the sanitizer reports sent by Ben. So I'm guessing that the error must be on a later stage of processing, probably in the Regression. Just before the time I saw the issue the first time I added new tests to "geometry-extensions", though it seems wierd that it could affect "geometry" and "geometry-index": https://github.com/boostorg/geometry/commit/9bc8baaa3b3fffc96fbb0abd10a8cb0… https://github.com/boostorg/geometry/commit/07c8b24b89e9fec75c6b7efc7e94032… Geometry is probably the library with the most complex tests structure. We use nested directories and build-project directive in Jamfiles. May this be a part of the problem? AFAIR at that time there were also some changes in the Regression, and the last time something like this happened Config was changed. Could some changes in those modules cause the issue I'm writing about? What may be the reason of such behavior and most importantly do you know how to fix it? Regards, Adam

2 6

What variants should be build on Windows by default?
by Olaf van der Spek 11 May '15

11 May '15

Hi, What library variants should be build on Windows by default? Currently it builds: --build-type=<type> Build the specified pre-defined set of variations of the libraries. Note, that which variants get built depends on what each library supports. -- minimal -- (default) Builds a minimal set of variants. On Windows, these are static multithreaded libraries in debug and release modes, using shared runtime. On Linux, these are static and shared multithreaded libraries in release mode. Personally I use a static-release runtime often (for simple deployments). What variants do others use and what should be build by default? Greetings, -- Olaf

4 3

[array] Support for multidimensional arrays?
by Peter Dimov 10 May '15

10 May '15

Currently, creating a multidimensional boost/std::array is a bit unwieldy: array<array<array<int, 5>, 4>, 3> a; // int a[3][4][5] a[2][3][4] = 5; As a first step, we should consider enabling the element type T of array<T, N> to be an array type. array<int[4][5], 3> a; // int a[3][4][5] a[2][3][4] = 5; This almost works already, except for the array interface functions specified and implemented in terms of, say, std::equal or std::fill. It is an open question whether these standard algorithms should not be fixed themselves to work on array types, as swap_ranges has been, but for now, we can implement the array functions in a way that works. Opinions on that part? If we agree that this is a good idea, the next logical step would be to eliminate the special handling of the first extent and simply allow array<int[3][4][5]> a; // int a[3][4][5] a[2][3][4] = 5;

4 4

[boost.org website] Incorrect links on Boost website
by Michael Levine 10 May '15

10 May '15

I was browsing through the Boost website this morning and the links on the right sidebar caught my attention. It appears that there is a duplicate entry for "coding guidelines for constant integral expressions", under Development / Requirements and Guidelines. The first of these links points to the correct page. The second links to a page regarding library metadata. I'm not sure who, exactly, to contact for this, so I figured this list would be the most appropriate means to have this addressed. Best regards, Shmuel

2 1

[algorithm] Removal of preference to std implementation
by Rob Conde 10 May '15

10 May '15

Marshall/Others-in-the-know, In boost algorithm change set 4dac507b773cf4a82727703ddffa2b36c42ab814, the code which uses std version of algorithms when available was removed. I'm curious about the motivation for this change. I can guess at a few potential reasons, but I figure you can just share the real reason. This change potentially explains a performance regression in my code so I'm just trying to understand the motivation before I make any code changes. Thanks! Rob Conde

1 0

[testing] Add a tester with hidden visibility
by Andrey Semashev 09 May '15

09 May '15

Hi, Is it possible to add a tester that enables hidden visibility (- fvisibility=hidden -fvisibility-inlines-hidden) when building and testing Boost? I believe, a single tester on Linux+GCC would be enough to discover most issues related to symbol visibility. The reason I'm asking is that I believe Boost should support building with hidden visibility by default. This mode allows more optimizations and does not export unintended symbols from libraries. I've set hidden visibility for Boost.Log (in develop) and, as I understand, Robert is doing a similar change in Boost.Serialization. However, when building 1.58 with hidden visibility for my project I found out that not all libraries work well with that setting. What do you think about it?

4 20

Re: [boost] Windows XP support survey
by Thomas Trummer 07 May '15

07 May '15

The sad reality in a lot of industries is that we still have to support Windows XP for the foreseeable future. Though, I understand the desire to drop support for it. Believe me, everyone would be more than happy to just require Windows 7 as default and be done with it. However, it is hard to convince costumers to make the switch when there is only risk and no tangible benefit attached to it. In addition to this we do have the problem that we control complex (and expensive) hardware that consists of components that don’t have drivers for Windows 7 because the manufacturer discontinued them. For these people the end of support from Microsoft isn’t really an argument since the computers are not connected to the Internet and continue to work fine, not to mention that you can still buy extended support from Microsoft. The huge outcry from the C++ community when Microsoft wanted to drop support for XP in Visual Studio should be a clear indicator that a lot of people livelihoods still depend on it. Unfortunately since then the install base hasn’t really shrunken too much. So even Visual C++ 2015 will continue to support it. So, as long as a Visual C++ version supported by Boost can build for Windows XP and there isn’t a significant cost attached to it, Boost should also continue to support Windows XP.

3 2

Re: [boost] Windows XP support survey
by Andrzej Horoszczak - Billon 07 May '15

07 May '15

1 0

Re: [boost] Interest in a container which can hold multiple data types?
by James Armstrong 07 May '15

07 May '15

Sorry about the README file, I pretty much just uploaded what I had to github so people can grab the source. Admittedly, things there are lacking. I'll work on filling out the readme with some instructions hopefully today. The main difference between this and a container of boost:any is that the container of boost::any still is a container of one data type, just being that one data type is a very flexible one. This container natively supports insertion of different data types. So, you could do, int my_int = 5; double my_double0 = 9342.132; double my_double1 = 987.654; std::string my_string("This is a string."); std::vector<double> my_vec; my_vector.push_back(123.456); omni my_container; my_container.push_back(my_int); my_container.push_back(my_double); my_container.push_back(my_string); my_container.push_back(my_vector); and the container happily stores the data. You can then access the data through iterators //access all doubles for (auto itr = my_container.begin<double>(); itr != my_containert.end<double>(); ++itr) { std::cout << *itr << std::endl; } or through specifying the data type and index //get the 0th element string std::cout << my_container.at<string>(0) << std::endl; -- James Armstrong

9 27

[lambda] Addressing pull requests
by Edward Diener 07 May '15

07 May '15

The maintainer for lambda in the libs/maintainer.txt list is Jaako Jarvi, who is the original author. But I see others who have made changes to the library. So I am wondering if anyone is still actively maintaining the library. There are a number of pull requests for lambda, related to updating the code for type_traits Version2, by John Maddock and myself. Is someone handling these ?

5 6