- Boost - lists.preview.boost.org

Re: [boost] GSOC 2010 Boost-Graph
by Chinmaya Poswalia 25 Mar '10

25 Mar '10

On Thu, Mar 25, 2010 at 3:21 AM, Chinmaya Poswalia < chinmaya.poswalia(a)gmail.com> wrote: > Hi Andrew, > I am finding all the graph topic(my favorite topic to work on) very > interesting that's why i asked a lot of questions regarding all in one mail. > > > - i want to work on semantic graph > > > - how about maintaining an adjacency list, denote V as nodes and E as > edges, so the space complexity will be O(V+E). > > > - Also we can employ pointers of type void* to point to edges/vertices > of different types so as to maintain the graph of different edge and vertex > types. > > > - rather than employing a vector in each node and edge either we should > allocate space during insertion or we maintain the size of vector array by > ourself. > > > - Through generic programming, all the operations being implemented on > graph up till now will be made applicable on these new members of the > library. > - various functions that can be induced on these graphs are > > > 1. Distribution of node degree > 2. finding clustering co-efficient of semantic graph > 3. relevance of a link and node > 4. scale on semantic graph > 5. disparity of connected types/dispersion co-efficient. > > > - Semantic graph are generating huge level of importance in networking > (network semantics on web), video screening, statistical analysis of > data,etc so it will be a great thing to work on it. > > Can you suggest some more links to read on for these > > > > I am working on my proposal about the graph and your help is crucial for > me. > With regards > Chinmaya >

1 0

GSOC Proposal for BIGINT - Long Mail
by swagat konchada 25 Mar '10

25 Mar '10

I had a look around in the Boost Sandbox and I found a BigInt implementation done by zeux during GSoC 07. I had a look at the code and I wanted to know its deficiencies. I'm interested in proposing some improvements in bigint and new bigdecimal implementation that is building on this code. There are 2 questions that I wanted to ask the list before moving further on this track: 1. Is there sufficient interest for a BigDecimal proposal on the lines of the Java implementation of BigDecimal (or has there been some prior work that can be used as a foundation?) 2. What are the deficiencies of bigint that have prevented its inclusion into the Boost distribution so as to address the same as part of my GSoC proposal. I did some homework and contrasted the bigint api in Boost and Java and found some interesting things that could be done. Boost.bigint was written by zeux and he used GMP bigint's features writing the Boost.bigint code around it. I think I could learn and analyse more after having a look at GMP bigint. After all, I think Boost.bigint should be more exhaustive. -------------------------------------------------------------------------------------------------------------------------------------------------------------- Feature Boost bigint Java BigInt -------------------------------------------------------------------------------------------------------------------------------------------------------------- Constructors from | int | byte[] #byte array with 2's complement representation of the BigInt. | unsigned int | int bitlen, int certainity #random BigInt with given bit length | int64_t | | uint64_t | | char *, int base | String | string& str, int base | String, radix -------------------------------------------------------------------------------------------------------------------------------------------------------------- *Conversion from string of digits to bigint is present. However conversion from/to byte-array(2's complement notation) to/from bigint isn't. Also the generation of a random arbitrary precision bigint should be done. int bitCount() #return the no of bits in 2's complement of the bigint; int bitLength() #no of bits excluding the sign bit double doubleValue() #casting bigint to double must be done. #even casts to int, uint and int64_t, char * must be done.* -------------------------------------------------------------------------------------------------------------------------------------------------------------- Arithmetic Operators += In Java, add, subtract, multiply, etc., | -= | *= | /= | %= | |= | &= | ^= #xor | <<= | >>= | ++ #postfix & prefix | -- #postfix & prefix -------------------------------------------------------------------------------------------------------------------------------------------------------------- Unary Operators + - BigInteger[] negate() ~ BigInteger[] not() ! #is zero < int compareTo(BigInteger val) <= BigInteger max,min(BigInteger val) > >= == != -------------------------------------------------------------------------------------------------------------------------------------------------------------- *These are some bit manipulators BigInteger clearBit(int n) #not in Boost.bigint #even setBit & toggleBit could be done, with a feature of changing the precision of the bigint. BigInteger[] testBit(int n) #also bit testers l* -------------------------------------------------------------------------------------------------------------------------------------------------------------- Binary Operators + BigInteger add(BigInteger val) - * BigInteger multiply(BigInteger val) {bigint lhs, int rhs} {int lhs, bigint rhs} / BigInteger divide(BigInteger val) % BigInteger mod(BigInteger val) , BigInteger modPower(BigInteger val) | BigInteger or(BigInteger val) & BigInteger and(BigInteger val) ^ BigInteger andNot(BigInteger val) << BigInteger shiftLeft(int n) >> BigInteger shiftRight(int n) -------------------------------------------------------------------------------------------------------------------------------------------------------------- Math function overloads abs(bigint_base) BigInteger abs(int exponent) pow(bigint_base, uint64_t) BigInteger pow(int exponent) div(bigint_base lhs, bigint_base rhs, bigint_base& remainder) BigInteger[] divideAndRemainder(BigInteger val) sqrt(bigint_base) ----------No Java version for BigInterger------------ -------------------------------------------------------------------------------------------------------------------------------------------------------------- *BigInteger gcd(BigInteger val) #GCD would be one of the most important requirements for bigint #Also the Java function of bool ProbablyPrime() is very interesting and would be challenging to implement.* -------------------------------------------------------------------------------------------------------------------------------------------------------------- -- Problem is a phenomenon indicating lack of thought. Swagat Konchada

3 2

Debugging Visualizers
by Alex Miller 25 Mar '10

25 Mar '10

quoted from the ongoing "5 Observations" thread: On Tue, Mar 23, 2010 at 4:36 PM, Steven Watanabe <watanabesj(a)gmail.com>wrote: > > Perhaps you would find this useful: > https://svn.boost.org/trac/boost/wiki/DebuggerVisualizers > > It is mentioned there that the same thing could be done for GDB,etc. However, I can't seem to find anything of the sort by googling (although it's very possible I'm just searching poorly). Could anyone point me towards where docs on how to do this could be found at?

2 2

filesystem::create_directories error
by Zachary Turner 24 Mar '10

24 Mar '10

I'm trying to use boost::filesystem::create_directories() with a path of the form \\?\Volume{Guid}\foo on windows. This seems to break the filesystem library because it doesn't understand the path syntax. when recursing up the parents it removes the slash at each level, so it ends up calling boost::filesystem::exists(L"\\?\Volume{Guid}") which always returns false, I think because GetFileAttributes() requires that paths in this syntax end with a trailing \, whereas the trailing \ has been removed. I'm tempted to fix this, but I'm sure this algorithm is fragile and any change I apply is likely to break another case. Does anyone have any suggestions on how to fix this? Thanks

3 4

[log] review part 1
by Steven Watanabe 24 Mar '10

24 Mar '10

AMDG I mostly have documentation notes at this point. Andrey, since I have so many, and they're mostly fairly minor, would it be better if I made a patch for the documentation source? I think newbies should probably be replace globally with new users, since I kind of feel that the term newbie is somewhat derogatory. http://boost-log.sourceforge.net/libs/log/doc/html/log/moti.html: ================================================================================ Mismatched number: Most of the time ***applications*** ... to monitor ***its*** execution... The beginning of the second paragraph could be improved: "This is where logging may help. The application stores all essential information of its run time to log, and once something goes wrong this information can be used to analyze program behavior and make necessary corrections." Here's my suggestion: "This is where logging can help. The application stores all the essential information about its execution to a log, and when something goes wrong this information can be used to analyze the program's behavior and make the necessary corrections." "out-of-box" should be "out-of-the-box" This phrase sounds rather clunky to me: "...along with public interfaces ready to be used to extend the library." This too: "A user should be able to extend functionality of the library with regard to collecting and storing information into logs." "The library should make as least performance impact..." should be "The library should have as little performance impact..." http://boost-log.sourceforge.net/libs/log/doc/html/log/how_to_read.html: ================================================================================ In the beginning of the second paragraph, I don't like the phrase "to have the first glance on" How about something like "...to get a feel for the library's capabilities..." "The tutorial gives the overview of the library features..." Suggestion: "The tutorial gives an overview of the library's features..." "The simple form typically describes the most common and easy way to do the task and it is being recommended to be read by newbies." Suggestion: "The simple form typically describes the most common and easy way to do the task and is recommended for new users." "The advanced form usually gives an expanded way to do the same thing but with ability to do some extra customization and in-depth explanation." I would like a definite article before "ability" and I would also like "and in-depth explanation" to be placed differently so that it doesn't sound like you're saying that the user has the ability to do in-depth explanation. "This form may come handy for more experienced users and should generally be read if your needs are not satisfied with the easy way." Suggestion: "This form may come in handy for more experienced users and should generally be read if the easy way does not satisfy your needs." "This part gives description of other tools..." should be "This part gives descriptions of other tools..." "This chapter is better to be read on per-case basis." Suggestion: "This chapter is best read on a case by case basis." "there is a reference which gives the formal description of library components. " The work "description" should be plural, "descriptions". "For simplicity in the code snippets in this documentation it shall be assumed that the following namespace aliases were defined" Suggestion: "To keep the code snippets in this documentation simple, the following namespace aliases are assumed to be defined" "Aside from these namespaces the library also provides boost::log::experimental namespace." Suggestion: "In addition to these namespaces the library also provides the boost::log::experimental namespace." "and thus are not advertized for productional usage" The word "productional" should be "production" http://boost-log.sourceforge.net/libs/log/doc/html/log/supported_compilers.…: ================================================================================ "The library should build and work with a reasonably well-compliant compiler" delete "well-" http://boost-log.sourceforge.net/libs/log/doc/html/log/installation.html ================================================================================ "The library has the compiled part which should be built as described in the Getting Started article." Suggestion: "The library has a separately compiled part which should be built as described in the Getting Started guide." For BOOST_LOG_NO_THREADS "Has effect on both the library compilation and user's code compilation." Suggestion: "Affects compilation of both the library and users' code." "The macro is automatically defined if no threading support detected." add "is" before "detected" For: BOOST_LOG_USE_CHAR Same as the first comment for BOOST_LOG_NO_THREADS. (This appear many times, I won't repeat it again) For: BOOST_LOG_USE_WCHAR_T "it is assumed the both character types support is enabled" Suggestion: "support for both character types will be enabled" "Defining only one of them allows to significantly reduce binary size of the library." Suggestion: "Defining only one of them significantly reduces the binary size of the library." For BOOST_LOG_NO_QUERY_PERFORMANCE_COUNTER: "If defined, disables support for QueryPerformanceCounter API" Suggestion: "If defined, disables support for the QueryPerformanceCounter API" For BOOST_LOG_USE_NATIVE_SYSLOG: "Has effect on the library compilation stage." Suggestion: "Affects only the compilation of the library." (repeated later as well) "If for some reason the support for native SysLog API is not detected, define this macro to forcibly enable support for native SysLog." Suggestion: "If for some reason support for the native SysLog API is not detected automatically, define this macro to forcibly enable it." For BOOST_LOG_NO_SETTINGS_PARSERS_SUPPORT: "If defined, all facilities related to settings parsing are not built." Suggestion: "If defined, none of the facilities related to the parsers for settings will be built." "You may define configuration macros..." should be "You can define configuration macros..." http://boost-log.sourceforge.net/libs/log/doc/html/log/defs.html ================================================================================ For Log record: I don't like the expression "A single pack of information," although I can't think of anything better off hand. "that is candidate to be put into log" Suggestion: "that is a candidate to be put into the log" For Attribute "functional objects" should be "function objects" "which return the actual attribute value upon invoking" Suggestion: "which return the actual attribute value when invoked" For Log sink: "collected from user's application" should be "collected from the user's application" "It is sink's nature that defines where..." Suggestion: "The sink defines where..." For Log source "An entry point for user's application..." should be "An entry point for the user's application..." "upon user's request" should be "upon the user's request" For Log formatter "functional object" should be "function object" "...that forms up the final shape of the textual output" is awkward. "Some sinks, like binary logging sink" should be "Some sinks, like the binary logging sink" For Logging core "The global entity that maintains connection between" should be "The global entity that maintains the connection between" "It is mainly used on the logging library initialization stage." replace "on" with "in." http://boost-log.sourceforge.net/libs/log/doc/html/log/design.html ================================================================================ I htink "scheme" with "figure" "The library provides configuration facilities to leave only one needed version of the library to be compiled." Suggestion: "The library provides configuration facilities to compile only the version of the library needed." "For example, if your application is in critical state" should be "For example, if your application is in a critical state" "so that user sees an error message..." should be "so that the user sees an error message..." "This allows to use the library" Suggestion: "This allows use of the library" "...require users to use loggers to write logs" "use" and "log" are used too many times here. "The more generic term "log source" is used to designate the entity" Suggestion: "The more generic term "log source" designates the entity" "You may see on the scheme that the" suggestion: "You can see on the scheme that the" "and allows to quickly wipe away unneeded log records" suggestion: "and allows quickly wiping away unneeded log records" anyway, don't split infinitives. "The sink-specific filtering allows to direct log records to particular sinks." Suggestion: "The sink-specific filtering allows log records to be directed to particular sinks." "Backends, on the opposite" suggestion: "Backends, on the other hand" "most probable places of the library extension" suggestion "most likely places for extending the library" "Along with the described above primary facilities" should be "Along with the primary facilities described above" ================================================================================ Issues: Instead of compiling a narrow or wide library, why don't you use libboost_log and libboost_wlog? I agree with other comments that I saw about Attribute being a poor choice for what it does. I would prefer that the term "Attribute" should be used for what you call an "Attribute Value," and that there should be a different term, like "Attribute Extractor" for what you call an "Attribute." In Christ, Steven Watanabe

3 10

[ASIO] Create SSL stream from native socket?
by Marcel Fransen 24 Mar '10

24 Mar '10

Hi, I would like to create an ssl stream for an already accepted native socket, just like is possible with basic_stream_socket< tcp >. How should this be done? Currently there is only one ssl::stream constructor which takes one parameter that is passed to the basic_stream_socket: template <typename Arg, typename Context_Service> explicit stream(Arg& arg, basic_context<Context_Service>& context) : next_layer_(arg), service_(boost::asio::use_service<Service>(next_layer_.get_io_service())), impl_(service_.null()) { service_.create(impl_, next_layer_, context); } To be able to create an ssl stream from an existing socket we would need a constructor which passes three parameters to the basic_stream_socket like this: template <typename Arg1, typename Arg2, typename Arg3, typename Context_Service> explicit stream(Arg1& arg1, Arg2& arg2, Arg3& arg3, basic_context<Context_Service>& context) : next_layer_(arg1, arg2, arg3), service_(boost::asio::use_service<Service>(next_layer_.get_io_service())), impl_(service_.null()) { service_.create(impl_, next_layer_, context); } Would it be a problem to add constructors that passes two and three parameters to the basic_stream_socket or can this be done in another way? Kind regards, M. Fransen _________________________________________________________________ New Windows 7: Simplify what you do everyday. Find the right PC for you. http://windows.microsoft.com/shop

2 1

Re: [boost] GSOC 2010 Boost.Graph
by Chinmaya Poswalia 24 Mar '10

24 Mar '10

Topology Generators : 1) along with path, cycle, star how about also adding bus ,mesh and hybrid topologies. 2) we can also include dual, complement , line and power of a graph.

1 0

Re: [boost] GSOC 2010 Boost.Graph
by Chinmaya Poswalia 24 Mar '10

24 Mar '10

Bindings to Python : there are tools as SWIG as language bindings. is it good employ them in making the bindings. Could anyone please suggest some tutorial for binding creation. I was not not able to find on google

1 0

Willing to Join GSOC 2010
by Xin Shuai 24 Mar '10

24 Mar '10

Hi: I'm interested in your Boost.Graph project. I think my current graph theory related research as well as C++ and python programming experiences makes me fit for the project. I'm wondering who's the mentor for the Boost.Graph project and I hope he/she can give me useful suggestions about the project. -- Xin(David) Shuai PhD of Complex System in School of Informatics & Computing Indiana University Bloomington 812-606-8019 The way to success is to do as much as important things, and as less as unimportant things, as you can...

2 1

[GSoC]I am interested in Boost.Graph
by doraemonok 24 Mar '10

24 Mar '10

My name is Rui Diao. I am a student of Graduate University of Chinese Academy of Sciences. My major is Computational Mathematics. I have never contributed to any open software project, but only spread the open source spirit to the people around me and join some open source groups. The only contribution of my source file is developing a Client of IP Gateway of My University by Qt open source version. Contributing to Boost C++ Libraries is an interesting and challenging job, and also, of course, fantastic. I am interested in one project of Boost.Graph : Graph Connectives. I like Graph Theory very much. I have experiences on coding (and teaching the friends in open software association of my university) some algorithms about Graph Theory: Dijkstra, Floyd, Bellman Ford, Ford Fulkerson, SAP, Kuhn-Munkres, etc. I have learned a lot in "How to finish an excellent project" during the internship of Microsoft Research Asia. Moreover, I am familiar with Mathematica which has similar function with Boost.Graph. I am very familiar with STL, but seldom use Boost. I just have experiences on using Boost.Regex before. I have checked out the source code by Subversion and began to read the part of Boost.Graph recently. The Boost.Graph library is missing connectives which contains Graph Join, Graph Union, Graph Product, Graph Intersection, Graph Difference, Graph Sum, Graph Product, Line Graph, etc. I think the algorithms may be not too difficult to me, but how to code with "Boost Style" (of course not only the code style) is a problem. I will try my best to finish the challenging work. I think may be there is some mentor can give me some advices or suggestions. Thank you all. Rui Diao

2 1

[log] Boost.Log formal review
by barend 24 Mar '10

24 Mar '10

Hi, This is my review of Boost.Log submitted by Andrey Semashev. > Please explicitly state in your review whether > the library should be accepted. I've been doubting about this for a few days. When I started my review, by trying to use the library, I discovered that the library is not header-only. I had thought that a logging library should be lightweight and had therefore expected only headerfiles. I then tried the library from John Torjo from 2007 but it is not header- only neither, so no reason to favour that one (I didn't dive into it further). Because of some messages on the list, a.o.: > I think that any reasonable log library must be compiled > to it's own lib due to the intermodule singleton requirement I realized that it is indeed useful to have a non-header-only option as well. The answers of Andrey were satisfactory, a.o.: > If we have a list of places in Boost where logging may be applied, > perhaps we could compile a set of requirements for such a lightweight > wrapper. When we have it, it would be easier to tell, whether Boost.Log > fits the role or not, or whether the wrapper should be part of it or a > separate submission. I'd be interested to hear from the library authors > on this topic. And I found in the Boost.Log documentation, TODO-list: > Think over a header-only configuration. Perhaps, with a reduced > functionality. So, good. Then I started the real review and I started to like the library, it is definitely useful, is configurable, trivial logging is quite simple. It also contains very useful non-trivial features such as writing to the Windows Event Log and alarm critical states as balloon tips. So I finally decided to give my YES vote, contingent on the condition that a lightweight header-only configuration, e.g. single-threaded, single-module, will be available for library writers. Some more about this below. --------------------------------------------------------------------- > What is your evaluation of the design? I like the options, the trivial logging, multiple back-ends, modularity and extensibility. I didn't have a detailed look to the design but in general it looks good to me. > What is your evaluation of the implementation? I didn't have a detailed look either. I glanced through the sources and the implementation is looking neat. I wonder why the class "basic_slim_string" is necessary, to me it sounds very strange why a task as logging needs an own string implementation. However, I didn't have the time to dive into it and didn't ask questions on this on the list. > What is your evaluation of the documentation? The documentation is good. Pages like design overview, tutorial, installation are easy to follow. The tutorial pages however contain some omissions and you have to refer to the examples to see what is missing. I mean here the chapter "Trivial logging with filters" which seems to give a complete example, but misses the namespace aliases, two of them would have been enough. > What is your evaluation of the potential usefulness of the library? Very useful, definitely, for applications and for libraries. > Did you try to use the library? With what compiler? Did you have any > problems? Yes, I used it using Visual C++ 2008 Express Edition. Because Boost.Log is not header-only, and I don't have the Boost libs compiled on my machine (don't ask me why, I never have them), I had several issues which other reviewers probably didn't run into. The defines _CRT_SECURE_NO_WARNINGS and _SCL_SECURE_NO_WARNINGS were necessary to surpress M$ warnings. The define BOOST_ALL_NO_LIB is of course necessary to build it like I did, and it works for Boost.Log. The compiler complained about a missing simple_event_log.h by event_log_backend.cpp, so I drafted it by hand, but I later saw that it is documented and you should neatly use the message compiler, so OK. After adding some Boost source files from thread, filesystem, regex and system it worked, datetime was not necessary. A complete build (including all mentioned sources) costs about 2 minutes. The tutorial did work (see message above on filtering, adding namespace aliasses was necessary) and I experimented with some things, many options seems very useful to me such as the log format, automatic date time, log file rotation, log file max size, etc. etc. I like the streaming feature (so for me printf is not necessary). I have some reservations about the compiler support: GCC 3.4.5, the default on Linux and MinGW, is not supported and for a task as logging this seems to me as a disadvantage. > How much effort did you put into your evaluation? A glance? A quick > reading? In-depth study? About five to six hours spread over some days. > Are you knowledgeable about the problem domain? Yes, I always need logging. In our library at Geodan we built logging as a singleton in 1995. We rewrote it to a logging-over-DLL's in 2000, but these two were not 10% as sophisticated and complete as this logging library is. So yes, these things were not header-only neither. --------------------------------------------------------------------- So back on the logging for libraries. If Boost libraries which are header-only will start to use this library, without protection, thousands and thousands of project files and makefiles worldwide will be broken. This answer: > If an existing header-only library starts using the Boost.Log library > we'll need to update the makefiles. This is a reasonable cost (not from Andrey)is therefore not satisfactory to me. This answer: > Which probably means that the decision is with the author(s) of each > individual library. (from the review manager) surprised me a bit, I think that in a library under review such an important thing may certainly be discussed, per library. However, as said, it is not a reason to vote no, because the rest of the library is quite good. A small macro (sorry) wrapper would already do most of the trick, and avoid breaking project files: #if defined(BOOST_ALL_LOG) || defined(BOOST_GEOMETRY_LOG) #define BOOST_LOG_GEOMETRY(o, s) BOOST_LOG_TRIVIAL(o) << s #else #define BOOST_LOG_GEOMETRY(o, s) #endif Where it can be called in libraries as: BOOST_LOG_GEOMETRY(info, "Log line with a value " << 3); This is inspired on the general define "BOOST_ALL_NO_LIB" and library- specific define "BOOST_LOG_NO_LIB". This small wrapper would really be useful, at least for us, and replace the #ifdef BOOST_GEOMETRY_DEBUG lines which are currently in our Boost.Geometry library. This is zero cost when one of the macro's BOOST_ALL_LOG or BOOST_GEOMETRY_LOG is not defined (by default it is not), and contains much functionality of Boost.Log as filtering, rotation, etc, when one of these macros is defined. Of course it can be enhanced or there are better solutions. Anyway, a lightweight solution will not be difficult. --------------------------------------------------------------------- So thanks Andrey for submitting this library, thanks Volodya for managing the review, and I hope it will be accepted with a lightweight header-only configuration included. Regards, Barend

2 1

interprocess
by Jan Stetka 24 Mar '10

24 Mar '10

Has anyone else noticed that a named_mutex gets destroyed after a windows xp computer goes into hibernate?

4 3

Problem regarding downloading Boost 1.42.0
by Wasik Mursalin Rushafi 24 Mar '10

24 Mar '10

Hi, I am trying to download the latest boost library from here<http://sourceforge.net/projects/boost/files/boost/1.42.0/>but failing. is there any problem? isn't it available or I am the only one facing this problem? -- Rushafi

2 1

[log] cygwin gcc-4.3.2 compilation/run-time errors
by vicente.botet 23 Mar '10

23 Mar '10

Hi Andrey, I have found thre kind of errors while compiling Bost.Log with cygwin gcc-4.3.2 * WHAR related * SYSLOG related Do you know if we can install syslog on cygwin? * SETTING related (reading from a configuration file) I have not yet analyzed the SETTING errors. I will see later. I have prefixed any modified line with the BUG WCHAR/SYSLOG/SETTING on the attached patches. After all this I have reached to compile the whole library with tests and examples. Only 2 test passed and There are yet some link warnings that could be suspect and a lot of runtime errors (see below) It would be great if you can apply some of the modification on the attached patch (of course not the Jamfiles) Best, _____________________ Vicente Juan Botet Escribá gcc.link ../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.exe Info: resolving vtable for std::exception by linking to __imp___ZTVSt9exception (auto-import) Info: resolving vtable for std::bad_cast by linking to __imp___ZTVSt8bad_cast (auto-import) Info: resolving std::cout by linking to __imp___ZSt4cout (auto-import) Info: resolving std::cerr by linking to __imp___ZSt4cerr (auto-import) Info: resolving typeinfo for std::runtime_error by linking to __imp___ZTISt13runtime_error (auto-import) Info: resolving std::ios_base::in by linking to __imp___ZNSt8ios_base2inE (auto-import) Info: resolving std::ios_base::out by linking to __imp___ZNSt8ios_base3outE (auto-import) Info: resolving typeinfo for std::bad_alloc by linking to __imp___ZTISt9bad_alloc (auto-import) Info: resolving typeinfo for std::bad_cast by linking to __imp___ZTISt8bad_cast (auto-import) Info: resolving typeinfo for std::bad_typeid by linking to __imp___ZTISt10bad_typeid (auto-import) Info: resolving typeinfo for std::bad_exception by linking to __imp___ZTISt13bad_exception (auto-import) Info: resolving typeinfo for std::logic_error by linking to __imp___ZTISt11logic_error (auto-import) Info: resolving typeinfo for std::exception by linking to __imp___ZTISt9exception (auto-import) Info: resolving typeinfo for std::basic_ostringstream<char, std::char_traits<char>, std::allocator<char> > by linking to __imp___ZTISt19basic_ostringstreamIcSt11char_traitsIcESaIcEE (auto-import) Info: resolving typeinfo for std::basic_ostream<char, std::char_traits<char> > by linking to __imp___ZTISo (auto-import/usr/bin/ld: warning: auto-importing has been activated without --enable-auto-import specified on the command line. This should work unless it involves constant data structures referencing symbols from auto-imported DLLs.) testing.capture-output ../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.run ====== BEGIN OUTPUT ====== EXIT STATUS: 53 ====== END OUTPUT ====== PATH="//bin://lib://lib32://lib64:/boost_1_41_0/bin.v2/libs/date_time/build/gcc-4.3.2/debug/threading-multi:/boost_1_41_0/bin.v2/libs/filesystem/build/gcc-4.3.2/debug/threading-multi:/boost_1_41_0/bin.v2/libs/log/build/gcc-4.3.2/debug/threading-multi:/boost_1_41_0/bin.v2/libs/regex/build/gcc-4.3.2/debug/threading-multi:/boost_1_41_0/bin.v2/libs/system/build/gcc-4.3.2/debug/threading-multi:/boost_1_41_0/bin.v2/libs/test/build/gcc-4.3.2/debug/threading-multi:/boost_1_41_0/bin.v2/libs/thread/build/gcc-4.3.2/debug/threading-multi:$PATH" export PATH "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.exe" > "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.output" 2>&1 status=$? echo >> "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.output" echo EXIT STATUS: $status >> "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.output" if test $status -eq 0 ; then cp "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.output" "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.run" fi verbose=0 if test $status -ne 0 ; then verbose=1 fi if test $verbose -eq 1 ; then echo ====== BEGIN OUTPUT ====== cat "../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.output" echo ====== END OUTPUT ====== fi exit $status ...failed testing.capture-output ../../../bin.v2/libs/log/test/attr_attribute_values_view.test/gcc-4.3.2/debug/threading-multi/attr_attribute_values_view.run...

1 0

Re: [boost] 5 Observations - My experience with the boost libraries
by Brian Wood 23 Mar '10

23 Mar '10

Tom Brinkman writes: >> Exceptions are a fact of C++ life. You can't avoid them if for no other >> reason than std::bad_alloc, unless you don't allocate from the free store. > > Not true, there are no-throw versions of std::new. > > In any event, my point is that c++ exception handeling should be optional. > Boost libraries need to be updated to reflect this. I used to support both return codes and exceptions. Eventually I gave up supporting return codes, because attempting to support both was quite difficult. Dropping support for return codes was kind of a difficult decision for me -- but after not being able to think of another project that supports both, I decided to only support exceptions. Since making that decision, I feel I'm able to make faster progress in other areas. Do you know of any libraries that support both return codes and exceptions? -- Brian Wood Ebenezer Enterprises http://webEbenezer.net (651) 251-9384

1 0

Re: [boost] GSOC 2010 Boost.Graph
by Chinmaya Poswalia 23 Mar '10

23 Mar '10

Please comment on the following topics Hierarchical Layout Algorithms : 1) what is being implemented in spring based layout or what are the files(headers files) in the boost. Computer Vision Algorithm : 1) suggest me some examples regarding the "Implementing a specialized graph type for grids would also be needed in order to avoid storing the graph topology explicitly". I am not getting the point correctly.

2 1

[Review] ITL review extended until March 7th
by Hartmut Kaiser 23 Mar '10

23 Mar '10

Hey all, After getting the approval from our review wizards we decided to extend the review period of Joachim Faulhabers ITL library until March 7th, 2010. That should give everybody some more time to finish their reviews. Regards Hartmut --------------- Meet me at BoostCon www.boostcon.com

4 4

[Boost.Locale] Questions (was New Release)]
by Alexander Lamaison 23 Mar '10

23 Mar '10

Artyom wrote: >> I sent this to the list via Gmane but >> its stuck in the moderation queue so >> I'm sending it to you personally: >> > You probably should register to boost mailing list directly? > I did that too but it said because it was my first post to the list (I usually hang out in boost.user not .devel) it had to be moderated. It's been days now :( >> I compile my code with MSVC8: >> > > I hadn't tested this library with MSVC8 (2005), I would be glad if > you can build it and run unit-test (if you using nmake then > run "nmake test") If you are working with solution there should be > an option. > I did this and everything passed. Though when I run examples on the command line the output includes much that isn't readable. Is this because the terminal doesn't handle the codepages correctly? > So I would know that MSVC 2005 is fine as well. Once you run the > test, can you please send me: your os version and ICU Version so I can > update the tested platforms list? > I'm using Vista 32-bit with ICU 4.2.1 I've just started internationalising my project (Swish http://swish.sourceforge.net) using Boost.Locale so I may have many questions for you as I go along. I hope this will helps Boost.Locale as much as it helps me. Firstly, I'm trying to decide which format() to use: boost::format or boost::locale::format. If I don't use currencies or dates and always use the global locale, are there any advantages to boost::locale::format? The tutorial says of boost::format "It knows nothing about new Boost.Locale manipulators". I don't understand what this means. Which new manipulators? I notice that boost::locale::format has a constructor that takes a message as an argument and then does something internally that depends on the value of _translate but I don't understand why. boost::format can be used like this: boost::format(translate("Id string %1%")) % arg so what does the extra boost::locale::format constructor allow that this doesn't? Also I make gettext enforce correct translation of format strings by passing --boost to xgettext. This reports an error if the format tokens, %1% etc., don't match in the original and translated strings. Changing these to {1} etc. in boost::locale::format means this no longer works. What was the tecnical motivation for this change? My second question relates to the directory structure for the .mo files and applies to gettext as much as to boost::locale (I'm new to gettext too). Why do they expect to find the files at lang_ctry/LC_MESSAGES/proj.mo? This makes sense when translations for many projects are stored in a common location such as |/usr/share/locale/ but on Windows this wouldn't be likely. I plan to store my translation in my program's installation directory and wish I could just put them in a flat directory: translations/ar.mo translations/en_GB.mo etc. Is it possible to make the lookup fallback to this scheme or even make the scheme customisable? Many thanks for this excellent library. Alex |

2 1

Re: [boost] GSOC - Boost Libraries - BIGING
by swagat konchada 23 Mar '10

23 Mar '10

Hello everyone! On Tue, Mar 23, 2010 at 2:20 AM, Andrew Sutton <andrew.n.sutton(a)gmail.com>wrote: > > Hi Swagat, > > We are glad that you are interested in submitting a proposal for Boost in > this year's GSoC program. However, it's hard to comment on your proposal > unless you provide details about what you intend to do over the course of > the summer. GSoC is a very competitive program, and Boost is particularly > selective about awards. > > Please see [1] for ideas on how to write a good proposal. > > Thanks to Andrew Sutton for his kind reply and the relevant link. They have been of great use to me. I had a look around in the Boost Sandbox and I found a BigInt implementation done by zeux during GSoC 07. I had a look at the code and I wanted to know its deficiencies. I'm interested in proposing a BigDecimal implementation that is building on this code. There are 2 questions that I wanted to ask the list before moving further on this track: 1. Is there sufficient interest for a BigDecimal proposal on the lines of the Java implementation of BigDecimal (or has there been some prior work that can be used as a foundation?) 2. What are the deficiencies of BigInt that have prevented its inclusion into the Boost distribution so as to address the same as part of my GSoC proposal. Thanking you, Yours sincerely, -- Problem is a phenomenon indicating lack of thought. Swagat Konchada

1 0

[log] Boost.Log formal review closing down
by Vladimir Prus 23 Mar '10

23 Mar '10

Boosters, just a a friendly reminder that the formal review of the Boost.Log library, written by Andrey Semashev, is nearly the end, with about 24 hours remaining. If you planned to submit a review but have not done so, it's just about the time! The documentation for the current version is available at: http://boost-log.sourceforge.net The downloads are at: http://sourceforge.net/projects/boost-log Here are the important points: 1. All comments are appreciated. Even if you don't have the time for in-depth study of everything, comments on parts of the library are welcome. In particular, if you are already using logging in your applications and have specialized requirements, you might want to directly check how the proposed library handles them. 2. The reviews and all comments should be submitted to the developers list, boost(a)lists.boost.org and the email should have "[log]" prefix in the subject to make sure it's not missed. The deadline for reviews is 23:59, March 17, PST = 2:59, March 18, EST = 7:59 March 18, GMT = 10:59, March 18 MSK 3. Please explicitly state in your review whether the library should be accepted. 4. The general review checklist is provided below: - 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? Thanks, Volodya

7 48

GSoC 2010 idea
by Viktar Sialiuk 23 Mar '10

23 Mar '10

Dear the Boost developers, I'd like to participate in GSoC 2010 event. Could you please review my idea. I very much appreciate your opinion about this. We have ability to generate OOP wrappers by the XSD schema in the popular languages( such as Java or C#) but have not in the C++. With such a tool we will be able to generate C++ classes or structures with functionality of serialization/deserialization from/to XML. Implementation of SOAP webservice client will be easy task( or writing webservice itself). We can create tool (for example xsd2cpp) under boost.xml library and include it to the boost.build. Thanks, Viktar Sialiuk

6 10

[GSoC] Introduction and questions
by Felipe Tanus 23 Mar '10

23 Mar '10

Hi, My name is Felipe Tanus and I would like to participate (for the first time) of GSoC in this project, specifically in the boost.process library. While my proposal is not ready, you can check a little about me in my website[1]. I already downloaded and had a look in the code, and read the documentation provided. This turned me even more excited about the project. To be able to do a proper schedule and define milestones, I need to study more about of POSIX and Windows process architectures, and what the APIs let me to do. For this, I have been reading some man pages (unistd, signal...) and the MSDN documentation. My first question about this project is exactly about that: Do you have any suggestions about where can I find out about this matter? I would like it to be from a reliable source. You can reach me in IRC, on freenode's boost channel as "fotanus", and by this e-mail. Thanks in advance, -- Felipe de Oliveira Tanus E-mail: fotanus(a)gmail.com Blog: http://www.itlife.com.br [1]: Site: http://www.inf.ufrgs.br/~fotanus/ ----- "All we have to decide is what to do with the time that is given us." - Gandalf

3 5

[Boost][Gsoc 2010] Boost.Graph project
by Jakub Tlałka 23 Mar '10

23 Mar '10

Hello, I would like to engage in the Boost.Graph project. I have an idea of a graph architecture for simulating various processes in networks. I think it would be useful in the Network Workbench. It would consist of network graph generator. It is something like topology generator, but apart from inducing topologies, It should also induce given features characteristic to a certain network (ie density, or distribution of some vertices parameters over a graph). In addition there would be some iterate function, that would simulate the processes given by some dependencies between vertices or edges. That's the simple draft of the idea. I don't have much experience in this kind of projects so I'm not quite sure about the possibility/necessity of the idea. I'm interested in graph theory, I have algorithmic experience connected with Informatics Olympiad and I'm ready to spend my time expanding the boost library. I'm also ready to engage in the other listed problems regarding Boost.Graph. I would be very grateful for every comment, this is my second open source project and I'm willing to gain my experience in this field, and contribute not necessarily in connection with gsoc. Best wishes, Jakub Tlałka

2 1

Thread-safe singleton pattern
by Andrew Chinkoff 22 Mar '10

22 Mar '10

I propose thread safe singleton pattern implementation (see http://www.research.ibm.com/designpatterns/pubs/ph-jun96.txt) I have attached "singleton.hpp" to this letter. Below is the test program: #include <stdio.h> #include <singleton.hpp> class A : public boost::Singleton<A> { // // Only Singleton class can create and delete us! // SINGLETON_IS_MY_FRIEND; public: void foo(){ printf("foo!\n"); } A(){ printf("A()\n"); } ~A(){ printf("~A()\n"); } }; void main() { // It is not compiled: //A a = A::Instance(); // That is exactly as it should be used: A::Instance().foo(); } Test program outputs: A() foo! ~A() My question: Does Boost community think that boost::Singleton is convenient (or successful) solution?

15 20

Re: [boost] GSOC - Boost Libraries
by Andrew Sutton 22 Mar '10

22 Mar '10

> I am Swagat and C++ is my passion. > Hi Swagat, We are glad that you are interested in submitting a proposal for Boost in this year's GSoC program. However, it's hard to comment on your proposal unless you provide details about what you intend to do over the course of the summer. GSoC is a very competitive program, and Boost is particularly selective about awards. Please see [1] for ideas on how to write a good proposal. Andrew Sutton andrew.n.sutton(a)gmail.com [1] https://svn.boost.org/trac/boost/wiki/SoCHints

1 0