- Boost - lists.boost.org

Re: [boost] Re: RPC library
by Iain Hanson 09 Mar '05

09 Mar '05

On Thu, 2005-03-03 at 22:28 +0000, Jarl Lindrud wrote: > Iain Hanson <Iain.Hanson <at> videonetworks.com> writes: > > [snip] > You're making my point for me. For something quick and dirty, who is going to > want to type all that out? And before I do, I have to learn about IDL, _var > types, safe downcasts, factories, CORBA::string's, etc, while in the back of my > mind there's the nagging suspicion that I've missed some subtle detail > somewhere. I hardly find 5 lines of code onerous. And it is easy to wrap up part of it in a template. Quick and dirty may be something you do to learn how to do something. But AFAIK, not something we create boost libraries for. IDL is what gives you type safety, something we usually appreciate in C++. > And whoever will be looking at my code needs to learn these things as > well. This is something that never fails to amaze me. Building distributed systems is hard to do correctly. It is very similar to the difficulty with threaded systems but the issues are subtly different and I think harder in distributed systems. If you can't learn a few conventions in half a day, then you probably have no business trying to build such systems. I probably spend half of my working life fixing systems that have been quickly hacked and then put into production. CORBA /EJB both have marketed themselves as the easy and safe way to write distributed systems and they are not. They are easy to write bad systems in. What you are proposing is an easier way to write even worse systems. > I also have to modify the build process to handle the IDL files, and then > find an ORB to build and link to. None of this is difficult, but I personally > find it rather cumbersome, and above all, for my situation it's just plain > unnecessary. All I wanted to do was to make a couple of remote calls between two > C++ programs! Something along these lines seems to me much more proportional to > the task at hand: You already have to link to the serialization library that you have had to build. I don't understand the difference. And your user needs to understand the trade-offs with each archive format. > > // server > > RCF_BEGIN(I_Echo, "I_Echo") > RCF_METHOD_R1(std::string, echo, std::string); > RCF_END(I_X); > > struct Echo > { > void std::string echo(std::string s) { return s; } > }; > > int main() > { > RCF::RcfServer server(50001); > server.bind<I_Echo,Echo>(); > server.start(); > return 0; > } > > // client > > RCF_BEGIN(I_Echo, "I_Echo") > RCF_METHOD_R1(std::string, echo, std::string); > RCF_END(I_Echo); > > int main() > { > std::cout << RcfClient<I_Echo>("localhost", 50001).echo("ho hum"); > return 0; > } > Your client has 4 lines of code to my 5 and you have macros in your interface. Not normally considered a good thing. And you have hard coded the server location. Always considered a bad thing in anything but toy code. [snip] > I guess the argument boils down to this: does anyone write network applications > for which CORBA-style scaleability is not a requirement? I would think so, but > maybe I'm wrong. > There are low level systems that need to work with sockets. The IETF has made a pretty good job of making the internet protocols scale. And there are hard and soft real time projects that need to use sockets. There are probably a lot of applications that have been built using sockets that should have used CORBA. Cost has historically been a big issue with CORBA because commercial implementations have typically ( in the U.K. ) cost of the order of £3,500 per developer a year and £2,500 per user a year run time. > RPC's don't scale too well, we both agree on that. But why do you think all > applications need to scale? If I'm prototyping something, then the last thing in > the world I need is scaleability, and I certainly wouldn't want the development > cluttered and slowed because someone has decreed that we must all use CORBA > because it scales so well. > If you are just playing around then I don't have a problem. But I don't think boost is in the habit of adopting toys. The problem with over simplified solutions is that a prototype is built, the boss asks for a few extra features, and before you know it, its rolled out to production and know one will pay for it to be re-written properly :-(. > Your position seems to be that programmers should not be afforded the option of > choosing; they should just bite the bullet and always use CORBA. If they don't > like that, then they should just avoid networking altogether. Personally I tend to prefer not letting programmers shoot themselves in the foot, if I can avoid it. Unfortunately, they ( and I sometimes ) are incredibly inventive at finding new ways to do so ;-). /ikh

7 11

N-arity operators in C++, also reducing temporaries. [Repost]
by Pavel Chikulaev 09 Mar '05

09 Mar '05

This text not as silly as my "Herb Sutter's operator = can be optimized", so please read it. I'm currently working on library "Lazy" which eases creation of classes which needs lazy evalution and n-arity operators to improve speed. The library's sample usage placed below. Please let me know what do you think about it. I'll appreciate any feedback on my library. Library description: Library eases creation of classes which needs lazy evalution and n-arity operators. Library features are: * n-arity operators support; * operator aliasing, such as a + -b will be computed as a - b; * custom operators; * no unnecessary temporaries during evalution and assigning. Library's sample usage: //Matrix.h #include "lazy.hpp" class transpose_tag {}; //Tag class class Matrix { LAZY_CLASS(Matrix) static lazy::type<Matrix> Matrix_; //Similar to Andrei Alexandrescu's Type2Type template public: Matrix(); Matrix(const Matrix &); Matrix & operator = (const Matrix &); void swap(void Matrix &); //Libray won't work without swap function. //If name of your swap function isn't swap then use //LAZY_CLASS_EX(Matrix, YourSwapFunctionName) //lazy constuctors which enables lazy operators Matrix(LAZY_OP(Matix_ + Matrix_)); //Matrix operator +(Matrix, Matrix) Matrix(LAZY_OP(Matrix_ * Matrix_)); //... Matrix(LAZY_OP(-Matrix_); Matrix(LAZY_OP(Matrix_ + Matrix_ * Matrix_)); //ternary operator //Now enable operators LAZY_ENABLE_OP(Matrix_ = Matrix_ + Matrix_); LAZY_ENABLE_OP(Matrix_ = Matrix_ * Matrix_); LAZY_ENABLE_OP(Matrix_ = -Matrix_); LAZY_ENABLE_OP(Matrix_ = Matrix_ + Matrix_ * Matrix_); //operator aliasing LAZY_ALIAS(Matrix + -Matrix_ = Matrix_ - Matrix_); //Custom operation Matrix(LAZY_CUSTOM_OP(transpose_tag, (Matrix_))); LAZY_ENABLE_CUSTOM_OP(transpose, Matrix_ = lazy::custom_op<transpose_tag>(Matrix_)); //same as Matrix transpose (const Matrix &) but there //won't be any temporary objects at all }; //Matrix.cpp Matrix::Matrix(LAZY_OP(Matix_ + Matrix_) op) { //Construct matrix object //lhs matrix = op._1 //rhs matrix = op._2 } Matrix::Matrix(LAZY_CUSTOM_OP(transpose_tag, (Matrix_) op) { //Construct matrix object tranposed to op._1 } //etc //Now the use of Matrix class Matrix a, b, c, d; //... a = b + c; //No tempories at all //Using normal practice it will take two temporary objects (no NRVO or RVO ) a = b + -c; //Will be evaluates as a = b - c; No temporaries again //Using normal practice it will take three temporary objects (no NRVO or RVO) a = b + c * d; //Will use ternary operator. No temporaries again a = transpose(b); //Custom lazy operation. No tempories again Maybe the library is very similar to uBLAS's matrix_expression and vector_expression. But with my approach there is no need to code matrix_expression and vector_expression at all and it is easy to create new classes which needs lazy evaluation and n-arity operators. I believe that my library should be very useful with big_int. So what do you think? Is there any similar approaches available? Is there any need of such library? Thanks in advance. -- Pavel Chikulaev

7 47

Herb Sutter's Assignment operator can be optimized + Proposal for very small boost utility library
by Pavel Chikulaev 09 Mar '05

09 Mar '05

Hi, Do you know that well-known Herb Sutter's assignment operator in most cases does unnecessary copying? No? Then read it to learn how to elminate it. class Matrix { //.... Matrix& operator = (const Matrix& that) { Matrixtemp(that); swap(temp); return *this; } //... }; Consider first example: Matrixa, b; //.... a = b; Here Herb Sutter's operator = works just fine. But consider second example: Matrix func (); Matrix a; //... Matrix = func(); Result of func() is copied to temporary objectand then operator = copies it again. So assigning to result of function creates one useless object. Elminating of temporary int object will cost nothing, but elminating of temporary std::vector of 10000 elements will cost a lot. But how can we define that copying is necessary? - Every function returns object not of type T but temporary object of type T. So let's code something: template<typename T> struct temporary : T { temporary() {} temporary(const T & t) : T(t) {} //some operators and constructors to create base object in-place not listed here }; Now let's change function consider operator +: Matrix operator + (const Matrix & a, const Matrix & b) { Matrix temp(a); temp += b; return temp; } To optimize it we do the following: temporary<Matrix> operator + (const Matrix & a, const Matrix & b) { Matrix temp(a); temp += b; return temp; } Maybe it can be hardly optimized using NRVO, so let's help the complier rewriting as following: temporary<Matrix> operator + (const Matrix & a, const Matrix & b) { temporary<Matrix> temp(a); temp += b; return temp; } By now, even returning temporary<Matrix> instead of Matrix, no optimization is done. To enable the optimization we need to change the class Matrix by adding one more assignment operator : Matrix & operator = (const temporary<Matrix> & that) { swap (const_cast<Matrix &>(static_cast<const Matrix &>(that))); return *this; } or using "deprecated" casts Matrix & operator = (const temporary<Matrix> & that) { swap ((Matrix &) that); return *this; } By now we've elminated unnecassary copying. Is this worth being called Pavel Chikulaev's temporary assignment operator? Is this worth being included in Boost? By including temporary.hpp, adding one operator in class, and replacing each function result to explicitly marked temporary you get really a lot. My temporary assignment operator as safe as Herb Sutter's. Maybe returing temporary<T> will make some obstacles to templated programming, so i'm going to add temporary_traits: is_temporary, add_temporary, remove_temporary. Some changes of BOOST_TYPEOF will be also needed. So, what do you think? Copyright 2005 Pavel Chikulaev. Use, modification, and distribution are subject to the Boost Software License, Version 1.0

12 29

Sockets
by Boris 09 Mar '05

09 Mar '05

There hasn't been much talk about a socket library recently. The Wiki stuff about sockets is quite old so I don't know if Hugo is still working on it? It's some time ago I added some proposals for a multiplexing library to the Wiki. So I went through most of the Wiki pages of Boost.Socket and Multiplexing and tried to gather and sort the ideas a bit. As I don't want to go even more into details by adding more C++ code I tried a top-down-approach using UML diagrams. Without sorting the ideas and getting a big picture we are stuck in the details and don't proceed. The network API isn't small, there are many different goals to reach and the network library might overlap with other libraries - quite a lot to do. :-) So here is the next small step: http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl?BoostNet As usual any comments are appreciated, Boris

7 13

Boost 1.32.0 and VS.NET 2005 under AMD64
by p.klotz＠icoserve.com 09 Mar '05

09 Mar '05

Hi I have sent a mail about this topic to boost-users a few weeks ago but got no response. There exist a few problems when trying to compile Boost 1.32.0 under AMD64 with VS.NET 2005: * No toolset exists to compile Boost 1.32.0 native on the AMD64 platform (no cross compile). The attached file vc-8_0-amd64-tools.jam fixes that. * InterlockedExchange() in line 44 of lwm_win32.hpp collides with a macro of the same name in WinBase.h One finds in WinBase.h: #elif defined(_M_AMD64) && !defined(RC_INVOKED) ... #define InterlockedExchange _InterlockedExchange I changed the name of Boost's function to InterlockedExchangeBoost() to avoid this problem. * No type long_type exists when compiling with the Microsoft Compiler. The 64Bit part in lwm_win32.hpp seems to have been written for the Intel Compiler. An additional #ifdef now checks if the 64Bit compiler is from Intel or Microsoft. The InterlockedExchange() and long_type issues are fixed by boost_1_32_0-AMD64.patch. Maybe these changes can be integrated in some form in the next version of Boost. Best regards, Peter.

2 2

Re: [boost] [1.33.0]: Some Moderation Please.
by Dave Gomboc 09 Mar '05

09 Mar '05

> Date: Tue, 08 Mar 2005 13:50:54 -0700 > From: "Victor A. Wagner Jr." <vawjr(a)rudbek.com> > > And I _hope_ what I put in the <hostile>...</hostile> was offensive, > it was intended to be. I'd hate to think that I'm accidently offensive > but can't be when I want to. Victor, please, quit while you're behind. Dave

1 0

[testing] Possible changes to testing..
by Rene Rivera 09 Mar '05

09 Mar '05

All, I've been reading all the testing related post and not having time to respond to them. So I decided to talk about what might be done to improve things in one big posting. For along time now one of my objectives has been to use BuildBot (http://buildbot.sf.net/) to improve the management of running the regression tests. For those who don't feel like reading about that software here's a quick summary: It is a client-server based configuration where the test server controls the test clients telling them what to do for test. In this arrangement the client do not directly control what they are testing, nor how. All the clients provide is an execution environment for the testing. The server is configured to control the various clients as to what they test, including how to get the source, and when they test. The testing is change driven instead of time driven as we currently have. This means that each time a change occurs on the source (CVS in our case) the server decides to tell the clients to run tests. Because of this direct communication to the clients the server is able to give direct feedback as to what the testing is doing. This includes a live display of test progress, down to a dynamic view of the test log. Some of the issues and how this step might help. And other changes I've thought about might help.. *Reproducibility* BuildBot controls what the checkouts are and what triggered the build. You can see this by selecting one of the yellow "Build #" links on the BuildBot display. If there's ever a question about what code produced an error one can get the specific version of the tree to attempt reproduction. The history of what tests have run is kept. Which means that we can finally answer the dreaded "When did that start breaking?" question. *Scalability* One big change I'd like to see is the breakup of running tests to make it possible for testers to run tests on subsets of libraries, or even individual libraries. For example there might be some testers which have a special interest in a particular library, Boost.Python comes to mind. It would be ideal to make it possible for them to only run those tests, and to go through the extra steps of doing the Python setup. Also for some popular platforms, it becomes possible to get much faster response rates from testing if, for example, we partition the library testing space throughout those platforms and they would test the libraries in parallel. For this to happen some significant organizational changes need to happen to the tests. As it currently stands such a division is not possible because of the way tests are defined and organized. We have a single test point, status/Jamfile, which a) points to the rest of the test points, libs/*/test/Jamfile, and b) defines it's own tests. The problem is that there is a conflict, for some libraries, between the tests defined in status/Jamfile, and the tests defined in libs/*/test/Jamfile. For example Boost.Config has a reduced set of test in status/Jamfile. This situation is likely a reaction to reduce test times to something manageable. And I understand that library authors would need to have a place to run their own set of comprehensive tests. My proposal is to create a set of canonical tests that comprise the regression testing suite independent of the library author's tests. This set of tests would be structured so that it's possible to run each libraries tests independent of others. It would be adding this type of structure: boost-root/tests/<library>/Jamfile boost-root/tests/<library>/<sub-library>/Jamfile boost-root/tests/<library>/<some-other-grouping>/Jamfile So for example a tester could say she only wants to test python/*, or numeric/ublas/*, etc. *Fragility* Simply put because BuildBot exposes the complete testing procedure it becomes much easier to see where problems occur in the testing. Also, because there is a central server managing the procedures it's more likely that problems can be solved at that one location instead of opaquely trying to fix the clients. With the, hopefully, additional distributed testers it becomes harder for testing to break down completely. Another aspect we need to address is fragility of the server. Ideally we would have multiple BuildBot servers to add redundancy. To make such a multi-server, multi-client setup resource efficient we would need to manage the distribution of testing between them. *Resources* The only predictable way to address the resource usage, is to distribute the testing so we can create more capacity. Breaking up the tests is the only way I can see to move there. It was already suggested that only slicing the testing to single toolsets would help. But that doesn't really address the problems. For example it would still be prohibitive for me to run tests on my G3 Mac for CW-8.3 because it's a slow machine and it would take days to run just one cycle of tests making the results useless. But it would be possible for me to run a minimal set of tests, for example Boost.Config and other basic tests. Restructuring also brings up the possibility of moderating the set of tests that make it into the regression suite. Right now we are in a position of asking library authors to define what the tests for the library. We can't moderate what gets into testing, or what needs go out. We need some procedure for reviewing tests, and some form of approval to get tests into the regression system. I'm not saying that authors would not be able to do additional testing, just that we need to define what the standard set of tests are so that we can concentrate our testing efforts. It would still be possible to set up additional resources to run "experimental" tests. *Response* The gain from segmentation and distribution of testing is hopefully obvious ;-) But another advantage of using BuildBot is that we are not tied to waiting for the XSLT processing to see results. Sure the results are not going to be as incredibly well organized as the Meta-Comm results but they are immediately available. So if there is a significant delay in the processing, because of load or breakage, we can still continue working. *Releases* Managing the testing for a release was brought up a many times. And it's clear that requiring testers to do manual changes is just not working. For BuildBot having the control of what is tested on the server means that at any point one, or some small number, person(s) can make the switch to have testing resources devoted to release testing. One possibility that the finner grain testing allows for is to limit the testing for a release to the required set of toolsets and platforms. OK, that's enough rambling... I know I haven't mentioned many other items raised, particularly about bjam and Boost.Build. I'll leave that for another post. I just need to get back to setting up the BuildBot server now ;-) -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com - 102708583/icq

2 1

RE: [boost] thread safety in boost date time
by Yitzhak Sapir 09 Mar '05

09 Mar '05

> From: Jeff Garland > > On Tue, 8 Mar 2005 11:34:15 +0200, Yitzhak Sapir wrote > > Boost date_time apparently has a thread safety problem in linux in > > that it calls localtime() in boost/date_time/time_clock.hpp and > > related functions rather than localtime_r() and related functions. > > This is even though BOOST_HAS_PTHREADS is defined by configure in > > user.hpp. > > That's true -- date_time doesn't do anything to ensure thread > safety. I'm a little reluctant to protect these functions since the > user can determine better if they need to take the associated > performance hit. Of course this needs to be documented, which > it isn't now. Thoughts? Ok, having investigated the problem a little more, I get the following.. The problem comes down to the fact that localtime/gmtime call tzset(), which in turn calls getenv("TZ"). This call isn't done in the _r versions which use pretty much the same code. getenv() appears to be not thread safe, but I'm not sure why. This does indicate, however, that probably the _r versions are a little quicker (because they don't call getenv()). Running localtime (no _r) in loops in four threads on a multi cpu machine resulted in no segmentation fault. Maybe something else is interfering with getenv() in unforseeable ways? boost is not the only one making these calls. log4cxx is another third party we use and it also makes these calls. It also calls wcsftime, a version of strftime, which appears to always call tzset()/getenv(). So, wcsftime/strftime appear to be not threadsafe either. For the record, we use libgcc 3.3.3 and glibc 3.2.3. The libgcc is a by product of compiling an equivalent gcc on all machines for distcc purposes, while the glibc is the version that was installed with the operating system. I am wondering if this version mismatch could be the cause of the problems, but it seems the getenv.c hasn't changed in glibc for 2 years. A link to an example posting about a problem that seems very similar: http://lists.gnu.org/archive/html/bug-glibc/2002-11/msg00092.html If it is to be documented and not fixed, it should probably be documented in a central place for all boost. (This is a good idea regardless). I think it should be fixed, given that the main difference is using a static buffer vs a stack allocated buffer and that the current version does more work (damage?) than the reentrant version. Yitzhak Sapir

2 1

copy_if, composition, and proper predicate
by Michael Goldshteyn 08 Mar '05

08 Mar '05

Given: // Definitions class Element { public: const std::string &GetTag() const; }; typedef std::vector<Element> ElementContainer; // Now here is what is required ElementContainer elements; ... ElementContainer matchingElements; std::string matchingTag("abc"); copy_if(elements.begin(),elements.end(),back_inserter(matchingElements)),pred); -- What is pred? Assuming that iElements represents the iterator inside copy_if, pred should be something like: iElements->GetTag()==matchingTag I could go about this by creating a new function object class/struct, which takes matchingTag as its sole constructor argument and does the comparison inside operator(). But, I want to use in place composition using binders, etc... to create the pred, so as not to have to create a function object class/struct. The question is, "What piece of code should pred be?" // Almost does the right thing, but it's only *iElements==matchingTag, I need iElements->GetTag()==matchingTag bind_2nd(equal_to,matchingTag) This seems like a very common operation. How can I compose the predicate, in place to do this? If there are multiple solutions, I would like to hear them as well! Thanks, Michael

2 1

Archive stuck
by Gennadiy Rozental 08 Mar '05

08 Mar '05

Sorry I know I was given direct address. But I couldn't locate it for some reason. Gennadiy

1 1