March 2014 - Boost - lists.preview.boost.org

[config] std::regex unusable on gcc?
by Andrey Semashev 23 Mar '14

23 Mar '14

Hi, I'm trying to use std::regex and it's failing with even the simplest patterns: #include <regex> int main(int, char* argv[]) { std::regex rex("\\d"); return std::regex_match("1", rex); } g++ -std=c++11 rex_test.cpp -o rex_test ./rex_test terminate called after throwing an instance of 'std::regex_error' what(): regex_error Aborted (core dumped) GCC version: gcc version 4.8.1 (Ubuntu/Linaro 4.8.1-10ubuntu9). GCC 4.7.3 exhibits the same behavior. Replacing "\\d" pattern with "[0-9]" results in the same failure. It works with Boost.Regex, Boost.Xpressive and with clang 3.4 & libc++. Am I doing something wrong? If my use of std::regex is correct then for me this basically means that std::regex is unusable. In this case, could we define BOOST_NO_CXX11_HDR_REGEX in Boost.Config for all versions of libstdc++?

3 3

[algorithm] How to contribute?
by Maciej Poleski 23 Mar '14

23 Mar '14

Hello I want to spend some time working on Boost. I saw this: https://svn.boost.org/trac/boost/wiki/SoC2014#Algorithms. I have some "algorithms and data structures" knowledge and I think about implementing something that would be valuable to Boost users and give me opportunity to dig a bit into Boost. I think about something simple to start with (like Radix sort). If everything go ok, I will try something more complex (parallel algorithms?). Let me know if this piece of work would make sense. Feel free to give another propositions if something different is more important. Maciej Poleski

2 1

[GSoC 2014] Boost.Fixed-Point Proposal
by Nikhar Agrawal 22 Mar '14

22 Mar '14

Hello everyone, Kindly find my proposal for Boost.Fixed-Point on the following link. http://www.google-melange.com/gsoc/proposal/review/student/google/gsoc2014/… I have been in touch with the potential mentors of this project for quite some time now. The proposal is a result of my discussions with them. Do let me know what you think of it. Any feedback would be invaluable. If there are any visibility issues, kindly let me know. Thank you. Nikhar

6 10

Boost.Binaries
by Flavius Condrea 21 Mar '14

21 Mar '14

Hi everyone, I am very excited to have the opportunity to contribute to Boost C++ Libraries. This is my proposal for GSoC 2014: https://docs.google.com/file/d/0B93eH9bi8ER0UHQtdDdFNFNoRGc/edit Best regards, Flavius Condrea

1 0

Teeks GCC test failures - multiprecision, spirit and others
by John Maddock 21 Mar '14

21 Mar '14

There are a lot of inscrutable failures from the Teeks machines running GCC on linux that I'm unable to reproduce (and aren't reproduced by other testers), for example: http://www.boost.org/development/tests/develop/developer/output/teeks99-03d… All the failures refer to is_convertible, and have an instantiation back trace that leads nowhere (a line that doesn't instantiate anything, and which refers to Boost.Range which multiprecision doesn't use). I see similar failures in the spirit tests as well. Is anyone able to reproduce these? Any ideas what may be going on? The failures are on the master branch as well if that helps. Many thanks, John.

5 5

[mpl] Merging 'develop' to 'master'
by Edward Diener 21 Mar '14

21 Mar '14

I would like to merge the mpl 'develop' changes to mpl 'master'. The develop changes have cycled long enough IMO for the merge to take place, without any problems showing up in the changes. There are two changes of my own and a number of Stephen Kelly eliminating support for very old compiler versions. The changes will clean up many obsolete workarounds for older versions of compilers that no one doing template metaprogramming should seriously use anymore. My own two changes fix some minor problems I ran into when testing clang with VC++ RTL on Windows. The changes make the MPL easier to understand. If there are no objections to the merge I will be glad to carefully do it myself, or Dave Abrahams/Alex Gurtovoy or others who have write access to MPL can do it. But I think it is important to get these changes into 'master', and tested in the 'master' regression tests, before the next Boost release.

3 6

Re: [boost] Boost.Binaries Project
by Antony Polukhin 21 Mar '14

21 Mar '14

Good! Now look, you need to write an officail proposal. here's a template: https://svn.boost.org/trac/boost/wiki/SoCSubmissionTemplate Send it to the boost(a)lists.boost.org. After that I'll tell the list that you have the qualification, finished test task and I'm willing to mentor you. Get hurry! March 21 is a deadline for student application. A few thoughts about the Binary project: Signing/shifrating a whole section is good idea, I like it. The very initial idea was to allow user specify code block or variable location to shifrate/checksum. That's what for the task with PUT_HELLO_THERE_IN_BINARY was meant for. Inserting markers into binary we'll be able to find code blocks later (find(binary, "Hello there!")). As you've mentioned compiler eliminates dead code, however there is a way to bypass it: if we put the jmp instruction directly in assembly block, compiler will not eliminate it (at least 3 years ago it was not eliminating). Binary library must consist of two parts: postprocessing: tools to find markers, modify marked variables, change sections and entry point... core: markers for code, markers for variables, functions to do something at code block enter/leave. 2014-03-20 20:43 GMT+04:00 Flavius Condrea <flavius.condrea(a)gmail.com>: > Hi Antony, > > > I solved the task you gave me. I attached the source code where you will > find a comment, too. > > I worked last summer at Bitdefender as a full-time Malware Researcher. I > studied a lot of reverse engineering, Windows API and basic Python. While > working there, I made researches on several malwares, developing detection > and cleaning algorithms. > > Also, using Pin library, developed by Intel, I created a C++ > information-revealing application when running a binary file, such as: > > - For each call it obtains information about the instruction > call address , the procedure address, the procedure parameters including > the parameters values. > > - It shows if the function writes in memory and in case it does > the application obtains these information from memory > > - It reveals if a new process or thread is created > > Then I developed a plugin (written in Python) for IDA Disassembler > Application. The plugin inserts strategically-placed comments with the > information mentioned in above. This way, it can be done an easier > binary-file analysis. I attached a few print screens for seeing clearly > what actually my tool does. > > In the same time, if a binary file has encrypted code (most of the > malwares have this), the developed application dumps the image from memory > (in the memory the code is decrypted) and restore the PE Header and > imported functions for an easier IDA analysis of the malware. > > > Best regards, > > Flavius Condrea > > > PS: About this task: should I include it in the application form? > > > 2014-03-20 9:13 GMT+02:00 Antony Polukhin <antoshkka(a)gmail.com>: > > 2014-03-19 21:17 GMT+04:00 Flavius Condrea <flavius.condrea(a)gmail.com>: >> >>> Hi Antony, >>> >> >> Hi and welcome! >> >> >>> I was thinking to apply for Boost.Binaries project. Therefore lately >>> I made researches about what does this involve. I thought of a way to >>> encrypt an executable file and decrypt itself at runtime. I have the >>> following solution for this: >>> >>> 1. Create an encryption key and a decryption key (public or >>> private) >>> >>> 2. Iterate each section and see if it contains executable code and >>> then encrypt that section with the key from point one. >>> >>> 3. Create a new section that will contain the Original Entry >>> Point(OEP) and decryption algorithm that will be executed at runtime. The >>> decryption algorithm will contain (at the end) a jmp to the Original Entry >>> Point. The decryption key will be delivered separately. >>> >>> 4. Change the executable header (PE Header) so that makes it valid >>> after the effectuated changes (this involves modifying the entry point to >>> point to the decryption algorithm and other things) >>> >> Sounds pretty good. Looks like you've done some investigations and know >> what you are talking about. >> >> >> >>> Besides these things I am willing to add other features according to >>> your wishes. >>> >>> I am waiting for your opinion. >>> >> >> You've got a good start, but there is a requirement in community to give >> some small tasks for the student. >> If you are willing to start Boost.Binary project, then you need to make >> a main.cpp file, that contains the following >> >> #include <iostream> >> >> int main() { >> std::cout << "Hello "; >> goto foo; >> PUT_HELLO_THERE_IN_BINARY >> foo: >> std::cout << "World"; >> } >> >> There you must define PUT_HELLO_THERE_IN_BINARY macro to some code, that >> will put the "Hello there!" string between two calls to operator << in >> binary file. In disassembly there must be something like this: >> >> ... >> call basic_ostream::operator<< >> jmp foo >> 48 65 6c 6c 6f 20 74 68 65 72 65 21 ; 'Hello there!' in raw view >> foo: >> ... >> call basic_ostream::operator<< >> >> If you'll make your solution portable between MSVC and GCC compilers - >> you're almost certanly in :) >> >> >> -- >> Best regards, >> Antony Polukhin >> > > -- Best regards, Antony Polukhin

1 0

[git] Jamfile in repository
by Karsten Ahnert 20 Mar '14

20 Mar '14

Hi, can we put a Jamfile and/or a Jamroot in git repository of a boost library? We develop odeint in a separate github repository and we have a Jamroot there to build the library without the boost super-project. To merge our changes into the official boost odeint repository we want to pull from the development repository. I am a bit afraid that this Jamfile/Jamroot might disturb the build process of boost.

2 1

[Range] Are pull request in github being considered?
by Gonzalo BG 20 Mar '14

20 Mar '14

There are 3 pull-request in github with zero comments, none closed, and the oldest is 3 months old (one is from me). Are they being considered or should one submit patches for Boost.Range over trac? The FAQ doesn't say anything about the preferred way to submit patches and the modular boost documentation mentions pull-request as valid options, but I would like to be sure that I sent my patch through the preferred channel. Bests, Gonzalo BG

3 3

Standard c++ XML parser API (Boost.XML)
by Saksham Maheshwari 20 Mar '14

20 Mar '14

Hey guys, As part of this year's GSOC, I am interested in developing a Standard c++ XML parser for Boost (Boost.XML project in Ideas Page). What I have encountered till now, there are many libraries which could be used for parsing XML file, but there is no Standard XML API in c++ to do the same, contrary to other languages like Java or python. I think this is an appropriate place to discuss about this API and then develop it. Since a past couple of days, I have been reading and understanding the existing Boost.xml (sandbox project) and Arabica project. What I think now is that there is no need to develop such an API from scratch, it would just be a wastage of our resources, else we could use the existing codes to develop such an API. We can use multiple backends supported in Arabica API and merge it with out existing Boost.XML to make such a standard API, so that it can adapted to a wide range of existing XML libraries. Also, the SAX implementation of Boost.XML has to be changed and it has to be implemented/modified (i.e. reader API), comparing it with Arabica::SAX. DOM API is working very fine. So, according to me, the possible tasks will be 1. To enhance or change the SAX implementation of Boost.XML 2. To provide multiple backends to this API. But for now, I want to know what you guys think about the idea. Thanks, Saksham Maheshwari PS: I want to develop an API, not a library :-)

5 10