September 2010 - Boost-users - lists.preview.boost.org

CreateProcess
by gast128 02 Oct '10

02 Oct '10

Hello all, I was asking if there is CreateProcess like functionailty in Boost. In our application we lock files, and spawn a process (thru _spawn). However it seems that m$ implemented in this way that handles are inherited by child process. This means that the child process is also getting the handle of the locked file, preventing the main application to write to this locked file. So I was wondering if there is a portable c++ way to create prcoesses. Otheriwse we have to write it ourselves and spying the sources of m$ it seems that parsing the commandline is not trivial.

5 5

boost serialize - stack overflow for large problems
by "Jörg F. Unger" 01 Oct '10

01 Oct '10

I have implemented the serialization routine for a rather complex structure with many nested classes. The serialization seems to work for small examples, however, for bigger data sets, I get a stack overflow error. If I increase the stack size from 8MB to 20MB, the examples seems to work. However, I'd like to optimize the code the way that I do not need to increase the stack size (especially, since not all users can do that themselves) So here are my questions: 1. Why is the required stack size different for different data sets different. The structure of the datasets is identical, only the number of data sets differs - so why is the recursive level different? The only thing I can imagine is, that I have a class (container) with a vector of objects. Each object stores a pointer to the container class. Is it possible that once serialization in the container class starts, that there is a recursive serialization pattern? This seems to be support by the fact that the objects in the container class are not serialized in the same way as they are stored in the container. class container { std::vector<object*> myObjects; }; class object { const container* myContainer; } 2. How can I try to decrease the required stack size for the serialization routines and is there a way to estimate the required stack size for a specific problem a priori, so that I can throw an exception with an error message manually instead of having a segmentation fault. Thanks for any suggestions JFU

6 9

boost for embedded systems?
by Jeff Graham 01 Oct '10

01 Oct '10

Hello, I have a Texas Instruments OMAP3 EVM board with an ARMv7 processor and a DSP, running Linux with kernel 2.6.32 and cross compiling with CodeSourcery's gnu g++ 4.3.3. I would like to know if any of boost is usable on this platform? I am most interested in the threading, bind, signals, and asio. I finally found the following page, and although it shows the compiler and hints at the platform, it is not obvious about an OMAP/DaVinci processor... http://www.boost.org/development/tests/release/developer/summary.html Any info is appreciated. Thanks, JEG

2 1

[BGL] visualizing large graphs
by Cedric Laczny 01 Oct '10

01 Oct '10

Hi, although this might seem a bit off-topic, I think it is an important topic and might be very well interesting for some of us. The boost graph library provides methods to export graphs e.g. for the use with graphviz. However, this is a rather "one-way" solution. Besides that, it is difficult to handle large graphs with graphviz. Therefore, I was wondering if there have been some recent developments here? I would especially be interested in a solution that allows the direct visualisation of a graph (even previously serialized), possibly integrating ways to filter out some of the data in order to be able to visualize the graph in some way. It is clear to me that visualizing arbitrarily large graphs will be limited by physical limitations mostly (RAM etc.). But I would find it very convenient to at least load the graph, set some filters, so that I don't need to do this seperatly via a C++-program. Actually, I am writing my graph to a format that can be visualized by Cytoscape. Unfortunately, this suffers from large limitations on the size of the graph (number of edges and nodes). I am looking forward to hear your ideas about this and possible suggestions about libraries or programs that might achieve this goal. Thank you. Best regards, Cedric

6 12

[format] print float number with fixed number of decimals *and* integer digits
by alfC 01 Oct '10

01 Oct '10

I always found the rules for printf formatting, which Boost.Format uses, cryptic. I usually find the right combination by trial and error. In this case I want to print decimal numbers with both fixed number of integer digits and decimal digits. How can I do this with Boost.Format. #include<boost/format.hpp> #include<iostream> int main(){ int i = 2; std::cout << boost::format("%04i") % i << std::endl; //prints 0002, ok! double d = 3.1415; std::cout<< boost::format("%04.8f") % d << std::endl; // ^ prints 3.14150000, // want 0003.14150000 //(or even ' 3.14150000') //(or even ' 3.1415 ') return 0; } Thank you, Alfredo

3 3

Re: [Boost-users] [serialization] boost::serialization adds huge amounts of exports to resultant Windows PE file (Robert Ramey)
by Chris Yuen 01 Oct '10

01 Oct '10

Robert Ramey wrote: > Are you sure you're building in release mode? Yes of course. Is that expected to make a difference? Reading from `force_include.hpp`, even Release mode wouldn't escape from this (see below) I dug into the problem a little bit more. I replaced `force_include.hpp` with the following two lines and then actually recompiled boost::serialization #define BOOST_DLLEXPORT #define BOOST_USED The code bloat is gone (as shown by `dumpbin /exports`, yes, actually by 8MB), *and* the resulting program actually compiles and runs fine (sample program at: http://boost.codepad.org/f9LodmfD **). I tested again at MSVC2010 at both Release|x64 and Release|Win32. This seems to be a rather intrusive hack that seems to have become outdated with MSVC2010 (haven't tested with MSVC2005 and MSVC2008 yet. It's installed on another machine). Right now, `force_include.hpp` reads this: #if defined(BOOST_HAS_DECLSPEC) && !defined(__COMO__) // ... # define BOOST_DLLEXPORT __declspec(dllexport) This means we are unconditionally doing this explosive exports for all MS platforms. Even Release mode cannot optimize away these explicitly marked DLL exports. Should we at least have an option to turn off this very intrusive workaround? Right now my only option is to patch `force_include.hpp` manually. There might be some people that actually need the hack, but for some of us who's dug into it and actually confirmed it's not needed, it would be nice to be able to turn that off since the bloat introduced is definitely non-trivial. ** Of course I don't mean this particular sample program generates 8MB of export symbols. This is only to show what I mean by "it compiles and runs." The actual program that resulted from boost::serialization's 8MB bloat contained 100+ "exported" classes serialized through base pointers. The specs is quite close to what I showed in the sample program. ** That doesn't seem to play well with CodePad's boost. Oh well. Best regards, Chris Yuen

2 1

invalid itialization of reference of type boost::thread&& from expression of thread 'boost::thread'
by Carla Strembicke 30 Sep '10

30 Sep '10

Hello I am having a problem compiling with boost thread. I have installed gcc 4.5.1 on my linux platform recently. I am using boost libraries 1.42. I do understand that this library has not been tested with this compiler. My question is, I see that 1.44 version of boost has been tested with gcc 4.5.1 on redhat. Is this issue, that I provide a link to, resolved in 1.44 or will I still see the same problem? http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43915 regards carla CONFIDENTIAL INFORMATION NOTICE: The information contained in this e-mail is privileged, confidential and intended solely for the use of the addressee named above. If the reader of this e-mail is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail is strictly prohibited. If you received this in error, please contact the sender and destroy all copies of this e-mail. Thank you.

2 1

Making boost::any play nice in function overloads
by dakatz＠akamai.com 30 Sep '10

30 Sep '10

Hi - I've got a method which should take a variety of types and with the exception of taking a function pointer it shouldn't care about which types it's getting, so I thought that a boost::any would be a nice way to achieve this. Here's a simplified prototype: ------------------------------------------------------------------------ class Table { public: // Base case void doSomething(const boost::any& input); // Special overload for function pointers for some special reason. // // Has logic to interpret NULL pointer as integer 0 and pass to // "base" boost::any case. void doSomething(void (*f)(void)); }; ------------------------------------------------------------------------ When I tried to call doSomething with a string literal argument (i.e. table.doSomething("Hello")), it of course failed since boost:any can't be initialized with an array. Okay - no problem. Another overload should do the trick, right? So I added an overload like this: ------------------------------------------------------------------------ class Table { public: void doSomething(const boost::any& input); // Supposed to take a const char*, convert it to a string, and // pass off to boost::any case. void doSomething(const std::string& input); void doSomething(void (*f)(void)); }; ------------------------------------------------------------------------ on the theory that a const char* would convert to a std::string. I couldn't add an overload for const char* directly since that would conflict with the overload for the function pointer (i.e. table.doSomething(0) would be ambiguous between the function pointer and the const char*). Unfortunately, this didn't work. When I called this with "Hello.": ------------------------------------------------------------------------ $ g++ -W -Wall -Werror -ansi any_overload_problem.cpp any_overload_problem.cpp: In function 'int main()': any_overload_problem.cpp:32: error: call of overloaded 'doSomething(const char [7])' is ambiguous any_overload_problem.cpp:8: note: candidates are: void Table::doSomething(const boost::any&) any_overload_problem.cpp:19: note: void Table::doSomething(const std::string&) ------------------------------------------------------------------------ I admit that I didn't foresee this, although I suppose I should have. Even though a boost:any can't be initialized with a const char*, that signature is still participating in the overload resolution. So does anyone have any suggestions for how to deal with this sort of thing? Is there a good way to get the boost::any conversions to match "less well" than something like the std::string conversion? Or am I just way off-base in trying to use boost::any here at all? Thanks! Dan Katz

2 2

[asio] promiscuous mode?
by Ryan McConnehey 30 Sep '10

30 Sep '10

If winpcap or some such library is installed on a machine is it possible to have asio read from the Ethernet card in promiscuous mode? Ryan

3 3

[iostreams] tellg not supported on devices / filters that aren't seekable
by Sebastian Karlsson 30 Sep '10

30 Sep '10

I find it inconvenient that tellg isn't supported on filters / devices that isn't seekable, since tellg doesn't really seek but simply reports the current read head position. Here follows some example code that will throw due to this: #include <ostream> #include "boost/iostreams/categories.hpp" #include "boost/iostreams/filtering_stream.hpp" #include "boost/iostreams/filter/zlib.hpp" int main() { boost::iostreams::filtering_stream< boost::iostreams::input > strm( ( boost::iostreams::zlib_decompressor() ) ); char sourceArr[512]; strm.push( boost::iostreams::array_source( sourceArr, sourceArr + sizeof( sourceArr ) ) ); strm.tellg(); } The problem is that, at least on windows, and I assume other platforms due to the somewhat lackluster interface of stream_buf, is implemented in terms of pubseekoff( 0, ios_base::cur, ios_base::in ). This call doesn't really alter the read head and should in this instance be supported, but isn't.

1 0