June 2003 - Boost-users - lists.preview.boost.org

RE: [Boost-Users] A newbie question.
by Jim.Hyslop 18 Jun '03

18 Jun '03

Amit Bhatia wrote: > I have just downloaded boost 1.3.0 a couple of days ago. Welcome! > While going > through the documentation I read that it is not necessary to build > the library, completely. [...] > I have this doubt that is it equally good practice to just > include the > .hpp files rather than building the whole libary. For most of the Boost library items, the header *IS* the entire library (that's the power of templates for you :-) The exceptions are the Python, Regex, Signal and Thread libraries - those you need to build. More details are at http://www.boost.org/tools/build/index.html > Also was just wondering if there are seperate disc groups > for each set of > libaries like numerics/math in one place ? There are separate lists for Python and the Boost.build system (details at http://www.boost.org/more/mailing_lists.htm) Other than that, everything goes through here. -- Jim

2 1

A newbie question.
by Amit Bhatia 18 Jun '03

18 Jun '03

Hi, I have just downloaded boost 1.3.0 a couple of days ago. While going thropugh the documentation I read that it is not necessary to build the library, completely. As of now the uBLAS and the random number library seem to be most useful to me. I have this doubt that is it equally good practice to just include the .hpp files rather than building the whole libary. Does it mean a very messy code, and a rather very different way of compilation, etc ? Also was just wondering if there are seperate disc groups for each set of libaries like numerics/math in one place ? Thanks, amit.

1 0

Newbie COMPILATION PROBLEM
by lebrun_l 18 Jun '03

18 Jun '03

Hi All, I'am using Microsoft Visual C++ .NET 7.0 I can't manage to compile a single example from the graph library. I get the following message : c:\BGL\boost_1_25_1\boost\type_traits\cv_traits.hpp(188): fatal error C1001: ERREUR INTERNE DU COMPILATEUR (fichier du compilateur 'msc1.cpp', ligne 2844) Choisissez la commande Support technique du menu ? (Aide) de Visual C++ ou ouvrez le fichier d'aide du Support technique pour plus d'informations Thank's in advance for your help. Louis

2 1

RE: [Boost-Users] Re: Regex in a service..
by Koss, Loren 18 Jun '03

18 Jun '03

Thanks to all for helping. As soon as I changed the account of the service, it worked which led me to believe it just wasn't finding the DLL, which in fact was the problem. All is working now. Thanks for all the great help and ideas! Loren -----Original Message----- From: Jeff Garland [mailto:jeff@crystalclearsoftware.com] Sent: Tuesday, June 17, 2003 3:38 PM To: Boost-Users(a)yahoogroups.com Subject: RE: [Boost-Users] Re: Regex in a service.. On Mon, 16 Jun 2003 09:05:57 -0700, Koss, Loren wrote > The service can start and I can attach and run through it, but if I > uncomment it (which is what I need to do), essentially it > IMMEDIATELY fails with a ERROR 1053: The service did not respond to > the start or control request in a timely fashion. Is it possible > that its not finding the proper libraries or a permissions problem > is happening? Windows services have to respond to the Windows Service Control Manager (SCM) within a certain amount of time (a few seconds as I recall) or the SCM will assume the start has failed. How long is your Regex taking? Perhaps you should put the Regex in another thread so that your SCM responses are timely.... HTH, Jeff Yahoo! Groups Sponsor <http://us.adserver.yahoo.com/l?M=247865.3456232.4744922.1261774/D=egroupmai l/S=:HM/A=1482387/rand=728390785> Info: <http://www.boost.org <http://www.boost.org> > Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> > Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to the Yahoo! <http://docs.yahoo.com/info/terms/> Terms of Service. [Non-text portions of this message have been removed]

1 0

Python and hello.World class example
by David Brownell 18 Jun '03

18 Jun '03

I am trying to run the hello.World class example in the tutorial and am running into problems. I can compile the dll fine, and then copy the dll (and the boost_python.dll) into my python directory, and then start my Python interpreter. The library will import, but when I call either the greet or set methods, I get the following errors from within the Python interpreter: [greet]: Traceback (most recent call last): File "<interactive input>", line 1, in ? TypeError: unbound method Boost.Python.function object must be called with World instance as first argument (got str instance instead) [set]: Traceback (most recent call last): File "<interactive input>", line 1, in ? TypeError: unbound method Boost.Python.function object must be called with World instance as first argument (got nothing instead) I am failrly new to Python and even newer to Boost.Python, so there may be something very basic that I am missing. I am using boost 1.30, bjam to compile, MSVC 7.1, Python 2.2, and Windows XP. Note that I can compile and successfully execute the inital Hello World example (hello.cpp in the tutorial subdir). Thanks in advance for your help! David Brownell

2 2

Link Problem with filesystem lib test program.
by Adalbert Perbandt 18 Jun '03

18 Jun '03

Hi there, I'm trying to get the boost::filesystem library working on Sun SPARC Solaris 7 with the Sun forte 6.2 compiler. ---------------------------------------------------------------------- $ cd boost_1_30_0/libs/filesystem/test $ bjam -sTOOLS=sunpro -sBOOST_ROOT=/opt/boost_1.30.0 [...] sunpro-Link-action ../../../libs/filesystem/test/bin/convenience_test.test/sunpro/debug/runtime-link-dynamic/convenience_test Undefined first referenced symbol in file boost::__RTTI__1nFboostPchecked_deleter4n0AKfilesystemSdirectory_iteratorLdir_itr_imp___ ../../../libs/filesystem/build/bin/libboost_filesystem.a/sunpro/debug/runtime-link-dynamic/libboost_filesystem.a(9dK8jJofKplZFEI9d1QA.o) [Hint: static member boost::__RTTI__1nFboostPchecked_deleter4n0AKfilesystemSdirectory_iteratorLdir_itr_imp___ must be defined in the program] ld: fatal: Symbol referencing errors. No output written to ../../../libs/filesystem/test/bin/convenience_test.test/sunpro/debug/runtime-link-dynamic/convenience_test CC -o "../../../libs/filesystem/test/bin/convenience_test.test/sunpro/debug/runtime-link-dynamic/convenience_test" -xildoff -g "../../../libs/filesystem/test/bin/convenience_test.test/sunpro/debug/runtime-link-dynamic/convenience_test.o" "../../../libs/filesystem/build/bin/libboost_filesystem.a/sunpro/debug/runtime-link-dynamic/libboost_filesystem.a" "../../../libs/filesystem/build/bin/libboost_filesystem.a/sunpro/debug/runtime-link-dynamic/libboost_filesystem.a" ...failed sunpro-Link-action ../../../libs/filesystem/test/bin/convenience_test.test/sunpro/debug/runtime-link-dynamic/convenience_test... ...skipped <libs!filesystem!test/path_test.test/sunpro/debug/runtime-link-dynamic>path_test.run for lack of <libs!filesystem!test/path_test.test/sunpro/debug/runtime-link-dynamic>path_test... ...skipped <libs!filesystem!test/operations_test.test/sunpro/debug/runtime-link-dynamic>operations_test.run for lack of <libs!filesystem!test/operations_test.test/sunpro/debug/runtime-link-dynamic>operations_test... execute-test ../../../libs/filesystem/test/bin/fstream_test.test/sunpro/debug/runtime-link-dynamic/fstream_test.run **passed** ../../../libs/filesystem/test/bin/fstream_test.test/sunpro/debug/runtime-link-dynamic/fstream_test.test ...skipped <libs!filesystem!test/convenience_test.test/sunpro/debug/runtime-link-dynamic>convenience_test.run for lack of <libs!filesystem!test/convenience_test.test/sunpro/debug/runtime-link-dynamic>convenience_test... ...failed updating 3 targets... ...skipped 6 targets... ...updated 33 targets... ---------------------------------------------------------------------- Is my assumption correct that this again is a consequence of a failure in Sun's forte 6 update 2 compiler? Is there any known workaround for this? Regards, Adalbert Perbandt.

1 0

Newbie: BGL: g++ -O2 or -O3 gives wrong numbers
by michael_wangzy 18 Jun '03

18 Jun '03

HI, I start looking at the Graph library and try the quick_tour.cpp on my machine. Here is something wrong I found out > gcc --vesion gcc (GCC) 3.3 (Gentoo Linux 1.4, PVR 3.3) > g++ -o test -O2 quick_tour.cpp > ./test vertices(g) = A B C D E edges(g) = (A,B) (A,D) (C,A) (D,C) (C,E) (B,D) (D,E) vertex: 0 out-edges: (0,134545512) (0,134545536) in-edges: (2,0) adjacent vertices: 134545512 134545536 vertex: 1 out-edges: (1,134545632) in-edges: (0,1) adjacent vertices: 134545632 vertex: 2 out-edges: (2,134545560) (2,134545608) in-edges: (3,2) adjacent vertices: 134545560 134545608 vertex: 3 out-edges: (3,134545584) (3,134545656) in-edges: (0,3) (1,3) adjacent vertices: 134545584 134545656 vertex: 4 out-edges: in-edges: (2,4) (3,4) adjacent vertices: digraph G { graph [ rankdir="LR", ratio="fill", size="3,3"]; node [ shape="circle"]; 0[label="A"]; 1[label="B"]; 2[label="C"]; 3[label="D"]; 4[label="E"]; 0->1 [label="1.2"]; 0->3 [label="4.5"]; 2->0 [label="2.6"]; 3->2 [label="0.4"]; 2->4 [label="5.2"]; 1->3 [label="1.8"]; 3->4 [label="3.3"]; } Note: the large numbers in the out-edges , like 134545512. Expected numbers are 0,1,2,3,4. Seems some fields are not initilized but read. Note: I tried gcc-3.2.3 and got same results Note: Without -O2 it is fine. My newbie questions are: 1) Is it compiler issue or BGL issue? Cannot we use optimization of gcc when using BGL? 2) I suppose gcc3.3 is good so if this is an BGL issue, where could the thing go wrong? How to fix it? Thanks, Michael

3 2

RE: [Boost-Users] Re: Regex in a service..
by Koss, Loren 17 Jun '03

17 Jun '03

Okay, I now know how to attach to a process - cool! Thanks for that.. However, it looks like when I comment out the following: Regex expression(strMask); The service can start and I can attach and run through it, but if I uncomment it (which is what I need to do), essentially it IMMEDIATELY fails with a ERROR 1053: The service did not respond to the start or control request in a timely fashion. Is it possible that its not finding the proper libraries or a permissions problem is happening? Thanks Loren -----Original Message----- From: Mark Sizer [mailto:yg-boost-users@m.gmane.org] Sent: Saturday, June 14, 2003 6:48 PM To: boost-users(a)yahoogroups.com Subject: [Boost-Users] Re: Regex in a service.. Yes, you can debug it. After starting the service, open the Task Manager and get the process ID. Then type "msdev -p <pid>" on a command line. If the error happens too quickly (e.g. on startup), put an infinite loop at the very start and break out of it with the debugger. int i=1; while (i > 0 ); is my favorite. Just change i to zero and everything starts going as normal (just don't forget to take it out!). Good Luck, - Mark P.S. I'm using regex in a service and it's working fine (at least I assume it is, nothing noticably bad has happened). MSVC 6, boost_1_30_0, W2K AS and W2K Pro. Koss, Loren wrote: > John, > > Thanks for the response. How do you deal with breakpoints within a service? > I'm compiling then starting the service from the service control manager. > Will I still be able to debug it? >>I seem to be having a difficult time using the Regex class in a >>service. It works wonderfully in a regular app. Can anyone help me >>out.. Basically, instantiating a regex class raises an exception. >>Any thoughts? >> > > Not unless you can tell me what exception and where - if you're using the vc > ide then set it to break on OS and C++ exceptions. > > John. Yahoo! Groups Sponsor <http://us.adserver.yahoo.com/l?M=247865.3425083.4707139.1261774/D=egroupmai l/S=:HM/A=1482387/rand=217687422> Info: <http://www.boost.org <http://www.boost.org> > Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> > Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to the Yahoo! <http://docs.yahoo.com/info/terms/> Terms of Service. [Non-text portions of this message have been removed]

3 2

RE: [Boost-Users] to_simple_string and Windows?
by Tom Matelich 17 Jun '03

17 Jun '03

Sorry, I didn't read carefully enough. Have you managed to make a test app fail similarly? tom > -----Original Message----- > From: adutoit [mailto:adutoit@stonethree.com] > Sent: Tuesday, June 10, 2003 11:02 PM > To: Boost-Users(a)yahoogroups.com > Subject: Re: [Boost-Users] to_simple_string and Windows? > > > Yes, I know about the UNICODE project option, that is why I > mentioned it in > my previous email... > Anyway, now I am left wondering why opto enabled (with /O3 or > whatever opto > is required) causes my call to "to_simple_string" to fail. > In case anyone > is wondering what I mean by that, it simply aborts without any further > information somewhere in the bowels of said function. > > A. ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

4 5

Maximum flow algorithms for interactive application
by Kai-Peter B�ckman 17 Jun '03

17 Jun '03

Hello fellow Boost users, I have been using the boost.graph library in an interactive application for solving the maximum flow problem using the push_relabel algorith. The application consists of the user building the graph network interactively, node by node and edge by edge. Currently the maximum flow for the whole system is recalculated after each change. In addition to the total maximum flow the individual flows in all edges are used by the application. The problem is that interactivity is lost when the amount of nodes approaches 10k with the number of edges varying between 20k-40k. My question is: Is there an algorithm that would let me quickly calculate the changed maxium flow solution after each change without needing to do the global calculations again? The idea here would be to amortize the running time over each interactive change and thus keep response times acceptable (<200ms). My intuition tells me something like this should be possible, but I don't want to reinvent the wheel. Thank you in advance. If someone replies by email I will summarize to the group. Have a nice summer. -- Kai-Peter Bäckman, kai-peter.backman(a)mistaril.com http://www.mistaril.com - Space Station Manager - PC strategy game

1 0