- Boost-users - lists.boost.org

[Interprocess/Process] Libraries status?
by Peter Novak 17 Jul '07

17 Jul '07

Dear Boost users/developers, first let me introduce myself to the list as I am newcomer here. My name is Peter Novak and I am using Boost since few months in some of my research projects. I need to implement some extensive interprocess communication and as I am already using Boost quite extensively, I looked a bit around and found that there are Boost.Interprocess and Boost.Process libraries hanging somewhere in the air. As I found, Interprocess by Ion Gaztanaga was already accepted for Boost release last in February 2006. Since then, we already saw 1.34 release, which did not include this library(why?), although it is already included in the CVS tree. Finally, there does not seem to be any special review notice, or any other remark about the fate of the Process library by Julio Vidal, although it was written in the frame of the SoC project last year and did not seem to receive negative comments on the developer's list. Is there somebody who could update me on the fate of these libraries? I mean whether they will be included (Interprocess) in the next Boost release, and if so, when is the next version of Boost scheduled to be released? Are there any considerations to submit Process library for review, or was it dismissed without a proper review process? As of now, I am considering to incorporate the Interprocess library into my source tree anyway and one day when the next Boost version will be released, I will migrate my package to the version included with the Boost 1.35(?). Thanks for your comments and help. Best, Peter. -- Peter Novak, pno-boost(a)aronde.net, http://peter.aronde.net/

2 1

[test] cannot wipe out linker errors
by Christian Henning 17 Jul '07

17 Jul '07

Hi there, I'm using boost 1.34 on MSVC 7.1. For whatever reason I cannot eliminate various linker errors. I have defined my BOOST_AUTO_TEST_CASEs in several .cpp files and this seems to create multiple definition of various symbols. The errors look as the following: string_table_test.obj : error LNK2005: "bool boost::test_tools::`anonymous namespace'::dummy_cond" (?dummy_cond@?A0x7aa5c0d5@test_tools@boost@@3_NA) already defined in ORM_test.obj string_table_test.obj : error LNK2005: "class unit_test_log_t::boost::unit_test_log_t & boost::unit_test::`anonymous namespace'::unit_test_log" (?unit_test_log@?A0x7aa5c0d5@unit_test@boost@@3AAVunit_test_log_t@23@A) already defined in ORM_test.obj Did anyone had to deal which such messages before? Thanks ahead, Christian

2 6

Pointer containers and raw arrays of pointers
by Chris Newbold 17 Jul '07

17 Jul '07

Greetings. I've been wrestling with some legacy C code which makes heavy use of raw arrays of pointers-- typically pointers to quasi-OO C structures which have specific "constructor" and "destructor" functions. Here's a contrived example: struct OldStuff { ... }; OldStuff* CreateOldStuff(); void DestroyOldStuff(OldStuff*); Then there are functions which return zero or more of these structures as output via an array of pointers: bool FiddleWithStuff(OldStuff** output, size_t count, ...); It's the callers responsibility to properly dispose of the results returned in 'output' by calling DestroyOldStuff() on each element. What I'd like to be able to do is provide a wrapper for functions like this which provide automatic management for the output values. Here's an example using ptr_vector: struct OldStuff_Destroyer { // What's up with the const pointer to deallocate?!? static void deallocate_clone(const OldStuff* p) { DestroyOldStuff(const_cast<OldStuff*>(p)); } }; typedef boost::ptr_vector< boost::nullable<OldStuff>, OldStuff_Destroyer> StuffVec_t; bool Wrapper(StuffVec_t& output, size_t count, ...); The pointer containers like ptr_vector seem to be a good fit, particularly given the customizable cloning behavior and is a very natural transition from an array of pointers. The problem I'm having is how to couple ptr_vector to the legacy code which expects to use arrays of pointers. Here's what I'd like to do: bool Wrapper(StuffVec_t& output, size_t count, ...) { // Won't work: resize() isn't supported output.resize(count); // Won't work: no way to get to raw storage return FiddleWithStuff(&output[0], count, ...); } As noted, this won't work for a couple of reasons. The pointer containers have an aversion to containing NULL pointers, which is why I imagine resize() isn't supported. And there's no way to get a pointer to the raw storage for the pointers to provide a buffer for the legacy code to write into. I do understand why the pointer containers don't support what I'm trying to do here, and the restrictions are entirely reasonable to provide additional safety. But... Using a container of smart pointers isn't an option, either, for the same reason: the need to have legacy code write into the storage as if it were an array. Does anyone have any suggestions on how to encapsulate a collection of pointers and some associated deletion logic? Obviously I could write my own pointer collection that satisfies my requirements, but I was hoping to use off-the-shelf bits. Thanks for any ideas. -Chris

3 3

Re: [Boost-users] Reply formatting plea
by Ovanes Markarian 17 Jul '07

17 Jul '07

On Mon, July 16, 2007 13:46, David Abrahams wrote: [...] > It gets really hard (for me, at least) to distinguish what you wrote from what the other person wrote. I know this seems nitpicky, but I ask that anyone who can tolerate it leave a blank line around replies, as follows: > > > quote somebody saying something > > about something they said > > Your reply to the original poster's > thesis, or even: Would you tolerate this style? I like when things which belong together are together... > > quote somebody saying > > something about something they > > said I hope this kind of formatting is clear as well. If not I will do my best and write as proposed by Dave. > Thanks, > > -- > Dave Abrahams > Boost Consulting > http://www.boost-consulting.com > > The Astoria Seminar ==> http://www.astoriaseminar.com > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > It is not my intention to start a long discussion, "yes" or "no" whould be great ;) With Kind Regards, Ovanes Markarian With Kind Regards, Ovanes Markarian

4 4

boost libraries and gcc-2.95 on linux
by Dmitry Teslenko 17 Jul '07

17 Jul '07

Hello! Are compiler requirements equal across all boost libraries? I've tried to statically link against boost_regex, boost_signals and boost_filesystem and last one fails to compile saying: /opt/boost_1_34_0/boost/filesystem/path.hpp:615: syntax error before `&' /opt/boost_1_34_0/boost/filesystem/path.hpp:617: `::basic_ostream' undeclared (first use here) /opt/boost_1_34_0/boost/filesystem/path.hpp:617: `Path' is not a class or namespace /opt/boost_1_34_0/boost/filesystem/path.hpp:617: `string_type' is not a class or namespace /opt/boost_1_34_0/boost/filesystem/path.hpp:617: `value_type' is not a class or namespace /opt/boost_1_34_0/boost/filesystem/path.hpp:617: using `typename' outside of template /opt/boost_1_34_0/boost/filesystem/path.hpp:617: parse error before `,' /opt/boost_1_34_0/boost/filesystem/path.hpp:619: ANSI C++ forbids declaration `operator <<' with no type etc ... While signals and regex compile cleanly. My Jamroot looks like: BOOST_PATH = <path-to-boost> ; project : requirements <include>$(BOOST_PATH) <threading>multi lib boost-signals : [ glob $(BOOST_PATH)/libs/signals/src/*.cpp ] : <link>static ; ... exe boost-test : main.cpp boost-signals ... ; Does 2.95 gcc compiler supported? Thanks.

2 3

[BGL] filtered_graph and bundled_properties
by Robin Ziehe 17 Jul '07

17 Jul '07

Trying to create a property map (passing 'get' a member pointer as property) from a bundled property of a filtered graph i get the following error: 'vertex_bundled' : is not a member of 'boost::filtered_graph<Graph,EdgePredicate>' Is this a known problem? I am using boost cvs with Visual Studio 2005.

1 0

Date-time - How to handle mixed instances of floating, UTC and local times
by Mike Higginbottom 17 Jul '07

17 Jul '07

I have an application that needs to deal with floating time, UTC and local time. I can use ptime to represent floating and UTC but there's then no way to tell, for a given ptime instance, whether it represents a floating time or UTC. I could use local_time with a null tz to represent UTC or floating (seems like a hack and still only allows me one or the other representation) and a non-null tz for local times. Given only two 'time' classes (ptime and local_time) what's the best way to represent three different time concepts? Should I inherit from local_time to give me my three representations? -- Mike Higginbottom Paul Smith Computer Services Silk Mill House 196 Huddersfield Road Meltham Holmfirth HD9 4AW United Kingdom http://www.pscs.co.uk +44 (0)1484 855800

1 0

[interprocess] construct<> and find_or_construct<>
by "JOAQUIN LOPEZ MU?Z" 17 Jul '07

17 Jul '07

Are construct<> and find_or_construct<> self-synchronized? For instance, can several process do the following seg.find_construct<type>("name"); (same seg, type and name) without guarding the operation with a mutex? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 2

Re: [Boost-users] [regex] How to compile with desired versionwhenmultiple versions of boost library are installed
by Chandan Nilange 17 Jul '07

17 Jul '07

Also as suggested by David, I had to change LD_LIBRARY_PATH=/home/ChandanNilange/work/boost/lib:$LD_LIBRARY_PATH Thanks David! Rgds, Chandan --- Chandan Nilange <chandan_nilange(a)yahoo.com> wrote: > Hello John & Steven, > > Thanks for ur help. > I was able to resolve the issue. > > As per the link > http://gcc.gnu.org/onlinedocs/cpp/Search-Path.html > > "All the directories named by -I are searched, in > left-to-right order, before the default directories. > The only exception is when dir is already searched > by > default. In this case, the option is ignored and the > search order for system directories remains > unchanged." > > Since we have "boost" directory at default > path:/usr/include/boost, it used to look > /usr/include/boost before > -I<install_path>/boost_1_34_0/boost > > Workaround Done: used -I<install_path>/boost_1_34_0 > Now "boost_1_34_0" is not in default path hence > looked > first > > Though this workaround works, let me know if this is > the right way to go. > > Thanks & Regards, > Chandan > > > > > > ____________________________________________________________________________________ > Boardwalk for $500? In 2007? Ha! Play Monopoly Here > and Now (it's updated for today's economy) at Yahoo! > Games. > http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow > > ____________________________________________________________________________________ Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games. http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow

1 0

Re: [Boost-users] [regex] How to compile with desired version when multiple versions of boost library are installed
by Chandan Nilange 17 Jul '07

17 Jul '07

Hello, I have boost_1_32 version installed at /usr/include/boost Also, I have installed boost_1_34 at another path:/home/myname/work/boost/boost_1_34_0/ Now when I try to run with 1_34 version, it still points to 1_32 version at /usr/include (this is observed when I debug using gdb it refers to /usr/include/boost/regex/v4/match_results.hpp:118) I tried following things: 1] export LD_LIBRARY_PATH=/home/myname/work/boost/boost_1_34_0/boost:$LD_LIBRARY_PATH export PATH=/home/myname/work/boost/boost_1_34_0/boost:$PATH 2] I am using following command for compilation g++ -c -g -I/home/myname/work/boost/boost_1_34_0/boost ${TEST_FILES_PATH}/${TEST}.c g++ ${TEST}.o -L/home/myname/work/boost/lib -lboost_regex 3] I am including boost header in ${TEST}.c as below #include "boost/regex.hpp" Any help is most welcome. Thanks and Regards, Chandan ____________________________________________________________________________________ Pinpoint customers who are looking for what you sell. http://searchmarketing.yahoo.com/

5 7