September 2008 - Boost-users - lists.preview.boost.org

[xpressive] Using placeholder in semantic actions
by Symons, Greg - SSD 04 Sep '08

04 Sep '08

Hi! I'm trying to use xpressive with semantic actions that use a placeholder, but I can't quite see the right way to do it. Here's my code: namespace grammar { using namespace boost::xpressive; struct ParsedUri { std::string hostname; std::string modelPath; std::string diagramPath; }; static placeholder<ParsedUri> result; static const mark_tag model(1), diagram(2), host(1); static const char pathSep = '/'; static sregex unreserved = set [ alpha | digit | '-' | '.' | '_' | '~' | ' ' | ':' | '$' ]; static sregex pctEncoded = as_xpr('%') >> xdigit >> xdigit; static sregex pathElem = *(unreserved | pctEncoded); static sregex path = pathElem | *(pathElem >> pathSep >> by_ref(path)); static sregex drive = alpha >> ':'; static sregex quad = repeat<1,3>(digit) [ check(as<int>(_) >= 1 && as<int>(_) <= 255) ]; static sregex ipv4 = repeat<3>(quad >> '.') >> quad; static sregex hostname = !(ipv4 | pathElem); static sregex pathNoHost = pathSep >> drive >> pathSep >> path; static sregex pathHost = (host= hostname) >> pathSep >> path [ result.hostname = host ]; static sregex modelPath = pathNoHost | pathHost static sregex uri = "rosert://" >> (model= modelPath) >> '?' >> (diagram= path) [ result.modelPath = model, result.diagramPath = diagram ]; } What happens is that when I compile the code, I get a compile error for the lines accessing the placeholder saying that the various members of the structure are not members of placeholder<T>, which looking at regex_actions.hpp seems to make sense to me. But in the example using placeholder<T> in the docs, it looks like you can use it in the semantic action as if it were a T. Somewhere I'm misunderstanding something, and I don't know what. Can anyone help me here? -- Gregory Symons Software Engineer ITT Space Systems Division greg.symons(a)itt.com (260) 451-6402 This e-mail and any files transmitted with it may be proprietary and are intended solely for the use of the individual or entity to whom they are addressed. If you have received this e-mail in error please notify the sender. Please note that any views or opinions presented in this e-mail are solely those of the author and do not necessarily represent those of ITT Corporation. The recipient should check this e-mail and any attachments for the presence of viruses. ITT accepts no liability for any damage caused by any virus transmitted by this e-mail.

2 2

MS VC++ wchar_t type build Q
by Scott Morgan 04 Sep '08

04 Sep '08

Is there an option to build the Boost libs with or without the wchar_t treated as built in (that whole /Zc:wchar_t mess MS dropped on us) Scott

2 3

Re: [Boost-users] Subsets of boost under version control (Angel Tsankov)
by Rodolfo Federico Gamarra 04 Sep '08

04 Sep '08

Hi Angel. I read your question and tought a little about it. I have some doubts about what exactly you're trying to do. >therefore it would be a good idea not to version control all of boost together with such projects. Do you want to version-control Boost's sources within the Boost client project's repository? Since, for that project which uses Boost, Boost it's an external dependency (and you're not touching Boost's sources) I believe that the general recommendation is not to version control those sources. But, it is sometimes helpful to have in the same repository the dependencies so as to check-put the whole thing easily. What do you mean by repository? The whole one or just particular SVN project within a given repository. You may create independent subsets as different SVN projects, within a same repository. The future additions would be new, independent, projects. Then, each one checks out the project that he/she needs. >Can someone recommend a way to achieve this with subversion's branches-tags-trunk policy? Trying to use tags/branchs/trunk, it seems that the concept of tag is closer to what you may need. In general, a tag is used to take a snapshot of the trunk and give a name to it. But it could be used in more ways. Take a look at this http://svnbook.red-bean.com/nightly/en/svn-book.html#svn.branchmerge.tags.m… With this, you can make-up in your local workspace whatever a subset you desire, and then upload it to the repository with any name you want to use to identify it. I hope you find these comments useful. Rodolfo. -- Rodolfo Federico Gamarra

1 0

How does boost.serialization do with BOM in text/xml files
by Tan, Tom (Shanghai) 04 Sep '08

04 Sep '08

I am think of using boost.serialization to replace CMarkup(http://www.firstobject.com/) as handle my XML config files. By playing with the examples coming with Boost, I found boost.serialization - does not handle BOM of UTF-8 files. - does not ignore <!-- ... > comments. Is there any workaround to this? Thanks.

3 2

[bjam] Presence of Mac OS X 10.6 SDK prevents boost 1.36 build
by Allen Cronce 04 Sep '08

04 Sep '08

Hi all, I have the Mac OS X 10.6 (Snow Leopard) SDK on my dev system. The presence of this SDK prevents boost 1.36 from building: --snip-- ./tools/jam/src/bin.macosxx86/bjam debug release --toolset=darwin -- architecture=combined --layout=system threading=multi address-model=32 link=shared,static macosx-version=10.4 --prefix=../1_36_0_install -- exec-prefix=../1_36_0_install install boost_1_36_0/tools/build/v2/build/feature.jam:466: in feature.validate- value-string from module feature error: "10.6" is not a known value of feature <macosx-version> error: legal values: "10.5" "10.4" "10.3" "10.2" "10.1" "iphone-2.0" "iphone-1.x" "iphonesim-2.0" boost_1_36_0/tools/build/v2/build/property.jam:276: in validate1 from module property boost_1_36_0/tools/build/v2/build/property.jam:301: in validate from module property boost_1_36_0/tools/build/v2/build/property.jam:310: in property.validate-property-sets from module property boost_1_36_0/tools/build/v2/build/toolset.jam:151: in flags from module toolset boost_1_36_0/tools/build/v2/tools/darwin.jam:227: in init-available- sdk-versions from module darwin boost_1_36_0/tools/build/v2/tools/darwin.jam:178: in darwin.init from module darwin boost_1_36_0/tools/build/v2/build/toolset.jam:38: in toolset.using from module toolset boost_1_36_0/tools/build/v2/build-system.jam:434: in process-explicit- toolset-requests from module build-system boost_1_36_0/tools/build/v2/build-system.jam:513: in load from module build-system boost_1_36_0/tools/build/v2/kernel/modules.jam:281: in import from module modules boost_1_36_0/tools/build/v2/kernel/bootstrap.jam:128: in boost-build from module boost_1_36_0/boost-build.jam:16: in module scope from module --snip-- The workaround was for me to hide the SDK, run configure again, then build. I still can't build successfully, due to an unrelated issue. But I'll post that separately in order to try and keep things clear. Snow Leopard will be here before you know it, so at some point the boost build system should be updated to know about 10.6. For that matter, maybe a new SDK would just generate a warning, not a hard error. Best, -- Allen Cronce

2 2

Subsets of boost under version control
by Angel Tsankov 04 Sep '08

04 Sep '08

Hello! It should come as no surprise that some projects use only small subsets of boost and therefore it would be a good idea not to version control all of boost together with such projects. Can someone recommend a way to achieve this with subversion's branches-tags-trunk policy? The recommendation should allow for future additions of new subsets of boost to the same repository. Regards, Angel Tsankov

1 0

Boost and the .net framework
by Richard Miller 04 Sep '08

04 Sep '08

Can anyone tell me, are there issues using Boost with the .NET framework and managed C++? How would shared_ptr etc work? Are there issues with the collection classes?

3 2

Re: [Boost-users] Serializing a char *
by Basilevs 04 Sep '08

04 Sep '08

On Sat, Aug 30, 2008 at 11:12 AM, Robert Ramey <ramey(a)rrsd.com> wrote: > I'm suspecting that there is a confusion about how to seriailzation a > pointer to > a null terminated string. It's natural to want to do something like > > char *p; > ... > p = "jsdkfjads" > ... > ar & p; > > because the library syntax encourages one to think that way. > > But in this case, one would want > > save(... > int i = strlen(p); > ar << i; > ar << binary_object(i, p); > } > > load(... > int i; > ar >> i; > ar >> binary_object(i,p); > } > How do we deal with the trailing null? In this example trailing null is not saved, and it is a good thing, but shouldn't we add it after loading?

3 5

Deployment of boost DLLs
by Tan, Tom (Shanghai) 04 Sep '08

04 Sep '08

- there are two versions of ms vc runtime, one with sp1 and the other without. Please make sure you are using the same version 1)to build boost, 2) to build your project 3)most importantly, to be distributed to the target machine. Please check the build number to check it. You may want to check the manifest file to see the runtime version your project is dependent on. - the best way to create a setup project using visual studio. It'll detect and wrap the correct runtime for you. -----Original Message----- Subject: [Boost-users] Deployment of boost DLLs Hi, I haven't been able to find much information on this, I hope someone can help me: I've only recently started using boost. I finally finished writing an application on my development machine. - Windows XP with Service pack 2 - Application developed in native C++ using MS VS2005 - Boost version 1.35, multithreaded, dynamically linked, installed boost using the installer from boost computing I then tried to deploy a release version of the application and the ms vc runtime redistributables to a test machine running a clean XP install and my application failed to run. The test machine also has XP Service Pack 2 installed. Strangely only the signal and the system DLL give the following error message: Error: The Side-by-Side configuration information for "c:\rtvc\bin\BOOST_SYSTEM-VC80-MT-1_35.DLL" contains errors. This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem (14001). The thread, date time, etc dlls are fine. However the signal and system dll are fine on my development machine. Has anyone encountered similar problems? Thanks, Ralf *********************

1 0

[review] Dataflow Library
by David Abrahams 04 Sep '08

04 Sep '08

I was reading other review commentary when it occurred to me that Boost.Accumulators is also implementing a dataflow system. I would like to know if Stjepan considered Boost.Accumulators and whether his library could build on some of that work. Thanks, -- Dave Abrahams BoostPro Computing http://www.boostpro.com

1 0