- Boost - lists.boost.org

Re: [boost] Switch library overloads of call operator
by dan marsden 06 Jan '08

06 Jan '08

Steven Watanabe wrote: >dan marsden <danmarsden <at> yahoo.co.uk> writes: >> The documentation is not specific about which specific type >> (e.g. mpl::int_<1> for case 1 matching in the >> underlying switch statement) is going to be used. >The type that is passed to f is whatever is in the list of cases. Ok cool. >> It looks to me like >> it would be convenient to overload >> operator() for each case I want to service, but this will be >> awkward to do without knowing the exact type >> that will be incoming. >> >> <snip> > >If multiple cases are not grouped together, I don't >see how it's an advantage to use my library vs. using a >plain old hand-written switch. Would you mind elaborating? I think you're probably right, my original suggestion that multiple overloads of operator() would be likely is probably wrong. I can only suggest handling special cases such as divide by zero for example, but I think my original suggestion may be a red herring. Cheers Dan ___________________________________________________________ Support the World Aids Awareness campaign this month with Yahoo! For Good http://uk.promotions.yahoo.com/forgood/

2 1

Switch library overloads of call operator
by dan marsden 06 Jan '08

06 Jan '08

Hi Another question about the design / docs for the switch library under review. From what I understand switch takes a runtime integral value, uses a PP generated switch statement, and calls out using some sort of MPL integral constant to specify the switch case to be served. This is pretty cool, as we seem to be travelling back from runtime into compile time world. The documentation is not specific about which specific type (e.g. mpl::int_<1> for case 1 matching in the underlying switch statement) is going to be used. It looks to me like it would be convenient to overload operator() for each case I want to service, but this will be awkward to do without knowing the exact type that will be incoming. Otherwise I'm going to have to use enable_if or something to separate out the cases, which seems unnecessarily awkward. The example provided is fortunate that it behaves uniformly for each incoming value, so only 1 overload of operator() is required. IMO this will not be the common case, an example with differing behaviour would be helpful. I think the documentation could do with a bit of clarification in this area. Cheers Dan ___________________________________________________________ Support the World Aids Awareness campaign this month with Yahoo! For Good http://uk.promotions.yahoo.com/forgood/

3 2

Re: [boost] Iostreams and Network [asio]
by Jonathan Turkanis 05 Jan '08

05 Jan '08

Hi JD, JD wrote: > And also, is there any attempt to write a Source/Sink to read and write > from shared memory? Currently there is http://www.boost.org/libs/iostreams/doc/classes/mapped_file.html. While I have been away from Boost, the Interprocess library has been accepted, providing sophisticated support for shared memory and memory mapped files. I am planning to write devices to access the functionality of the Boost.Interprocess for 1.36. > JD >> Hi, >> >> Iostreams documentation states: "For example, Boost.Iostreams can be >> used to create streams to access TCP connections". Is there any such >> attempt to create a TCP/IP communicating Source/Sink couple? >> I know asio is doing this, but I was looking for an pure Iostreams >> solution. Currently there are no devices for TCP/IP communications. I am planning to implement blocking TCP/IO devices using asio for 1.36, but I also hope at some point to achieve a better degree of integration. Standard iostreams are not a very good way to access network connections, since they are designed for blocking i/o; however, Boost.Iostreams is also intended as a framework for filtering, and I was careful to design filter concepts that would work with asynchronous and non-blocking devices. When Boost.Asio was proposed, I was looking forward to the possibility of introducing asynchronous and non-blocking device concepts to make the libraries work together, but unfortunately I became so busy that I was not able to participate in the review process or collaborate with Chris Kohlhoff in the subsequent development of the library. Now that I have more time I am committed to trying to make the two libraries work together as much as possible. >> Regards, >> >> JD -- Jonathan Turkanis CodeRage http://www.coderage.com

1 0

[iostreams] understanding the models
by Benjamin Collins 05 Jan '08

05 Jan '08

I'm trying to write a device for iostreams, but I'm having a hard time understanding how the current sources and the models explained in the documentation are in agreement. Maybe I'm shooting myself in the foot by never being able to work on this for more than 30 minutes at a time and having that 30 minutes be late at night, but... Anyway, it seems odd to me - for example, the boost::iostreams::mapped_file_source class has the following category of source_tag + direct_tag + closable_tag, and has member functions open(), size(), data(), begin(), and end(). None of these are particularly surprising. After all - a mapped_file has to be open()ed at some point. But under what model as defined in the docs? In other words, how and when is that function called, and how does a stream_buffer know he can call open() on a particular device? Also, the documentation indicates that a source device needs to support the read() function - but mapped_file_source has no read() (maybe because it is convertible to 'direct'?). -- Benjamin A. Collins <ben.collins(a)acm.org> http://bloggoergosum.us

2 1

[iostreams] Patch for latest CodeGear (Borland) compiler
by Nicola Musatti 05 Jan '08

05 Jan '08

Hallo, The patch in http://svn.boost.org/trac/boost/ticket/1294 improves support for the latest version of bcc32, 5.9.2 . If I don't receive objections in two or three days I'll apply it. Cheers, Nicola -- Nicola.Musatti <at> gmail <dot> com Home: http://nicola.musatti.googlepages.com/home Blog: http://wthwdik.wordpress.com/

2 1

build error during integration of serialization lib
by Arunava Saha 05 Jan '08

05 Jan '08

Dear Robert/experts, I am getting build errors after the integration of versioning in serialization lib. I have included the version.hpp. The following is the error msgs I am getting. Can someone please point me to the issue? FYI, everything is working fine except this versioning stuff. error: 'version' is not a template error: explicit specialization of non-template error: 'mpl' has not been declared error: ISO C++ forbids declaration of 'int_' with no type error: expected ';' before '<' token error: 'mpl' has not been declared error: 'integral_c_tag' does not name a type Thanks, Arunava.

2 1

[1.35.0] Intel compiler - does anyone care?
by Beman Dawes 05 Jan '08

05 Jan '08

The great bulk of remaining 1.35.0 release test issues are with the Intel compiler. See http://beta.boost.org/development/tests/release/developer/issues.html Does anyone really care about this compiler? Should we drop it from the high priority list? If there are folks who do care, perhaps they could suggest a plan to eliminate the failures, or at least shed light on why so many tests are failing. --Beman

9 20

Switch library review begins today and ends on Jan 9th
by Stjepan Rajko 05 Jan '08

05 Jan '08

The review of the Switch library by Steven Watanabe begins today. Description: The built in C/C++ switch statement is very efficient. Unfortunately, unlike a chained if/else construct there is no easy way to use it when the number of cases depends on a template parameter. The Switch library addresses this issue. You can download the library from the Boost Vault: http://tinyurl.com/23qn7w For convenience, the docs have been uploaded here: http://tinyurl.com/2h6ut9 ... and the source file here: http://tinyurl.com/2cbfyk What to include in Review Comments Your comments may be brief or lengthy, but basically the Review Manager needs your evaluation of the library. If you identify problems along the way, please note if they are minor, serious, or showstoppers. Here are some questions you might want to answer in your review: * What is your evaluation of the design? * What is your evaluation of the implementation? * What is your evaluation of the documentation? * What is your evaluation of the potential usefulness of the library? * Did you try to use the library? With what compiler? Did you have any problems? * How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? * Are you knowledgeable about the problem domain? And finally, every review should answer this question: * Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. Thanks to Dan Marsden, Larry Evans, Tobias Schwinger, and Alexander Nasonov for already opening up the discussion in the following threads: http://tinyurl.com/2u6fjr [nabble/boost-user] http://tinyurl.com/267uj3 [nabble/boost-dev] Thank you for taking the time to look into this library, and I hope that some of you will be able to provide a full review for it. Regards, Stjepan

1 0

[compressed_pair] using compressed_pair to substitute a possibly empty member
by Tobias Schwinger 05 Jan '08

05 Jan '08

Hi, I'm using compressed_pair to substitute a possibly empty member whose type is specified by a template argument and experience difficulties: template< typename T > class A { T x; public: // [ctor, ...] void foo() const { // if T is a reference type, non-const access to the // object referenced by x is possible, here } }; // but: template< typename T > class changed_A : boost::compressed_pair<T,boost::blank> { public: // [ctor, ...] void foo() const { // first() always returns a reference to const // even if T is a reference to non-const } }; So: Did I find a subtle design flaw or bug in compressed_pair or is it just the use case I'm throwing at it? In case of the latter we should probably introduce a compressed_member (or so) and have it implement const semantics consistent with members. Regards, Tobias

1 0

[Interval] Conversion from integer types?
by John Maddock 04 Jan '08

04 Jan '08

Can someone explain what the issue is in converting an integer type to a boost::interval, and in performing mixed integer/floating point arithmetic and comparisons? Is it simply that the integer types may have more bits in their representation than floating point ones, and we may therefore loose precision? If so, can we not convert integers to intervals implicitly, complete with calculation of the error? That would auto-magically make mixed integer/floating point arithmetic work as well. And yes, I'm prepared to invest some time to make this work and do the right thing, if time is the only stumbling block :-) Thanks, John.

3 6