- Boost - lists.boost.org

JSON Parser GSoC 2013 (part 2)
by Stephan Bourgeois 12 Apr '13

12 Apr '13

Hi everybody, thank you for all your replies. I will try to summarize my impressions so far: 1. Although there is a solution in Boost.PropertyTree, a majority of users consider it to be incomplete, and quite a few developers are using external libraries at present. 2. As Bjorn Reese pointed out, a solution should offer several different interfaces. Some being already implemented in the protoc library. Michael Marcin suggested that something similar to JsonFX in C# be added to Bjorn's list. 3. There is already a Lexer implementation as part of the tokenizer API in protoc. This is how I imagine I could contribute to the project: 1. Collect requirements for functionality and interfaces. 2. review existing Java, C++, C#, Android library solutions. 3. Write a Parser with error recovery using the Lexer from the protoc library. Or adapt an existing parser if there is a good reason for doing so. 4. review JSON specific issues like type casting and how to deserialize arrays with multiple types. (see posts by Arindam Mukherjee and Michael Marcin on deserialization) 5. With the help of my mentor, implement the different interfaces, preferably in a modular way. (I know an example of this in compiler design using the visitor pattern for code generation) Please don't be offended as I only had the time to have a quick look at a single library. I understand that I need to look at a wider code base from other libraries and familiarize myself with general issues related to JSON and serialization. Bjorn Reese offered to mentor this project, please let us know if someone more "officially involved" with Boost needs to approve this. Please let me know what you think. Stephan

1 0

[gsoc-2013] Physics Library Abstraction Layer
by Preston Hamlin 12 Apr '13

12 Apr '13

Hello, my name is Preston Hamlin and I am a current CS undergraduate at the Florida State University. I was reading over the list of ideas provided and then referenced the source code libraries to get a feel for what was being asked for. I found the prospect of writing an abstraction layer for PhysX or related libraries to be interesting. I am somewhat familiar with THREE.js which is an abstraction of WebGL. In THREE.js there is a heavy amount of abstraction to the degree of providing geometric primitives, particles effect presets, simple materials and basic data structures (matrices, vectors, etc...). Given the platform that THREE.js operates on, such a degree is understandable. How much of the physics library is to be abstracted away in generic structures and functions? I would think that having a few simple primitives (cube, low-poly sphere, etc...) would be required since they are commonly used. However more advanced meshes like a torus might be better produced through using modifiers on said primitives rather than a built-in preset. I think the degree of abstraction is most important for modifiers, and I desire to get a feel for what is being looked for. I assume the end goal is to be able to produce Blender/Unity/3dsMax usable code as well as standalone utilities. Many thanks, -Preston

4 7

Re: [boost] JSON Parser GSoC 2013
by John Venarchick 12 Apr '13

12 Apr '13

> 2013/4/12 Michael Marcin <mike.marcin(a)gmail.com> > >> Arindam Mukherjee wrote: >> >>> In JSON we typically deal with maps and arrays. The arrays themselves >>> could >>> have arbitrary types (string, object, array, numeric, boolean, null) as >>> elements. The key types in the maps are always strings and the value types >>> in the maps could be anything that can appear in an array, including >>> another map or array. >>> >>> Due to this, I'd imagine being able to use Boost.Variant or Boost.Any in a >>> list and as a value_type in a map would help. >>> >>> I have been doing some work with JSON and boost by modifying the current 1.53 ptree. I created a type for use as the data_type of the ptree that represents the various options for JSON data (no value, null value, bool, integer, float, string). It uses boost::variant and visitors. I have it working for my purposes, but would want to make it a little nicer before submitting a patch. I also implemented property_tree::translator_ between to make it work with the spirit classic parser in json_parser_read and json_parser_write. I can post more if anyone is interested. I'm also working a generic way to load arbitrary structures from the created ptree. -John V

1 0

Re: [boost] [gsoc-2013] Physics Library Abstraction Layer
by Preston Hamlin 12 Apr '13

12 Apr '13

Considering that I have received positive feedback on this project, I shall likely make a proposal for it when the time comes. Before then, I would like to pin down the scope and breadth of this library. I am in possession of the PhysX SDK and have been reading it over to come up with some questions to ask so that (should my proposal be accepted) I would not have to be overly concerned later. On the first page of the SDK documentation, it is denoted that several distributable (PhysX) DLLs are required for applications on the Windows platform. Since Boost is a library project, I am unsure what to do with DLL files other than unceremoniously stick them in a Resources folder until the time comes, or have a script built into one of the library files. Perhaps contact Nvidia? I have worked a little with WinAPI, but not in the case of using an SDK. Since this would be a beginning point for the library, might I assume that complex mesh deformations and simulations (e.g. fluid dynamics) are beyond the scope of what should be done over the summer? Certainly one could implement certain models with effort, but I mean that the library will lack explicit methods for something of the following degree: http://www.youtube.com/watch?v=7Who8EpbvCY The same question goes for Multithreading. I will continue to tinker with and read over the SDK(s) and brush up on my matrix arithmetic. For the proposal, should I have a couple concrete classes? Or would pseudocode/feature descriptors suffice? More questions on the way, -Preston

1 0

Fwd: How to start contributing for Boost
by Hasit Bhatt 12 Apr '13

12 Apr '13

I'm 3rd year B.Tech student in Computer Science with quite good proficiency in c,c++ and python. I'm applying for Google Summer Of Code. How should I start?

2 1

Can C++11 INVOKE be defined with C++98 compilers?
by Vicente J. Botet Escriba 12 Apr '13

12 Apr '13

Hi, I would like to implement C++11 INVOKE on C++98 compilers. I have a C++11 implementation on boost/thread/detail/invoke.hpp based on the one of libc++. Could this be done or there is something that can not be done inherently? Best, Vicente

2 4

GSoC 2013: ODE or B-trees?
by Артём Серов 11 Apr '13

11 Apr '13

Hello! My name is Artem Serov. I would like to participate GSoC in Boost and I have a dilemma. I don't have enough information to choose one of two projects in boost: B-trees and Ordinary Differential Equations and want to find out which project matches me better. Before composing a formal application (with all formal requirements) for a project and deeper investigating of the environment i hope to get a piece of advice about my doubts. A few facts: - the 3rd year student of Moscow Institute of Physics and Technology (State University), Russia. - ~1 year exprerience of C/C++ industrial programming in Intel as intern. - no experience in open souce project contribution. Both projects are very challenging and interesting for me. ODE: I had cources of Theory of ODE, Computional Mathematics and Parallel programming (OpenMPI) at the university, so i am familiar with the topic. B-trees: The topic "B-trees" is located in the additional list of ideas for GSoC, so there is no concrete information except the title. Could someone (maybe who have put this in the ideas list) will tell anything: maybe requirements, any thoughts, why is it mentioned in GSoC ideas list, why isn't this idea in a main list or any information like in projects in the main ideas list, please? What would you advise me: to apply for the concrete project from the main list or to develop the project from additional list? What is preferable for the community? Could you give a kind of very brief estimation and comparisson of difficulty of both projects? -- Thanks, Artem Serov

3 3

[Graph] Efficiently finding cycles in a (multi)graph
by Louis Dionne 11 Apr '13

11 Apr '13

Dear Boost, I have implemented the `hawick_circuits` algorithm described in [1] for finding all the elementary circuits of a directed (multi)graph. I implemented it in a generic fashion, planning to propose it for inclusion in Boost. Boost.Graph currently contains the `tiernan_all_cycles` algorithm for finding all the cycles in a graph. However, it is undocumented and it has a time bound (for which no precise analysis is provided) that makes it unsuitable even for moderately sized graphs. In fact, I implemented `hawick_circuits` because `tiernan_all_cycles` did not cut it for my application. My implementation is available at [2]. It is licensed under the Boost license and was tested using the test suite at [3]. I am posting this message for two reasons: 1. To know whether there is interest for such an algorithm in Boost.Graph. 2. To spread the word that this implementation exists for C++, to save people the hassle of reinventing the wheel before Boost.Graph officially provides it. I will create a Trac ticket if there is interest, but note that I won't submit a patch immediately because I am in deep juice with school right now. Regards, Louis Dionne P.S.: More information on the test suite can be found at [4]. [1] http://bit.ly/12LMwIr [2] http://bit.ly/10MSAhW [3] http://bit.ly/YdLHVn [4] http://bit.ly/16SvlW6

1 1

GSoC 2013
by İslam Yaşar 11 Apr '13

11 Apr '13

Hi, I'm a graduate student and work since 2 years as a C++ developer on Unix-like systems (FreeBSD and Linux), i want to implement a library for Boost. It is a thread-safe queue that is implemented on disk. You can use it as a IPC. Is anyone interested to mentor this project ? Thanks Islam Yasar

3 2

Boost.uBLAS for GSoC 2013
by Sathyam Vellal 11 Apr '13

11 Apr '13

Hello. I'm Sathyam Vellal, a 3rd year undergraduate student pursuing Computer Science. I'm interested in contributing to Boost.uBLAS in unifying the vector and matrix representations. I've looked at the docs ( http://www.boost.org/doc/libs/1_53_0/libs/numeric/ublas/doc/index.htm) to get an idea of the library. Does anyone have any suggestions on how I could proceed more and connected to this project? Thank you -- Sathyam M Vellal Computer Science and Engineering, P.E.S. Institute of Technology, Bangalore, India.

2 1