Group: LibreDWG/SummerOfCode

From LibrePlanet
Jump to: navigation, search

Future GSOC Ideas

See the TODO file.

In particular

  • encode support for r2004-r2018,
  • reverse engineering of the ACIS version 2 format,
  • decode support for some missing objects/classes, see examples/unknown.c
  • a simple graphical UI for the internal structures and reverse engineering (branch: e.g work/dwgiview)
  • AutoLISP adsrx, Java bindings

Past GSOC 2014

Google Summer of Code 2014 dates are out. The GNU Project has been approved as a mentoring organization.

This is the ideas page for GNU LibreDWG. Ideas for other GNU packages can be found here.

Please feel free to edit this page at any time. Put your name here if you are interested in being a mentor:

(Rodrigo Rodrigues da Silva)

(Hardeep Singh Rai Email: hardeep.rai AT gmail DOT com)

Students should either choose one of the ideas listed here or propose their own in order to write their applications.

We strongly recommend that students willing to apply write to us on the development mailing list before submitting a proposal.

We expect that candidates do a further research on topics related to the proposals, and make it clear in their applications. Applications containing design and implementation proposals will be highly considered. Feel free do criticize LibreDWG's internals. There is still much to be done.

Check our homepage and our wiki for further information about the project. Mail us at libredwg at gnu dot org if you have any questions.

Please read (and follow) the GNU guidelines for SoC before writing your application.

Automated test suite

Build an automatet test suite for LibreDWG. The test suite should have alive tests and also unit tests (test read and write for each object), compare test outputs with expected values, ecc. The output should be as detailed as "DWG_OBJECT_ENTITY_CIRCLE read failed in address 0xfa15 of file foo.dwg (R2004)"

  • Level: MEDIUM
  • Priority: VERY HIGH
  • Desirable skills: good knowledge of computer graphics and CAD, C, shellscript or python programming

3D Solid decoding support

Currently LibreDWG is only able to decode 3D solids partially. The solids in DWG are encoded into the SAT and SAB formats, used by the ACIS 3D modeling kernel. There are not free implementations of this kernel, which means that the SAT and SAB streams that we are able to extract are useless if not parsed. Once parsed, the solids must be converted to other openly documented formats, and properly rendered and worked within free software tools. It is yet not clear if converting belongs to the scope of this idea, since there aren't any known {sat,sab}2something free software converters. Anyway, if your application somehow gives a light in addressing this issue, either implementing or not the converter, that would be a plus.

  • Level: MEDIUM/HARD
  • Priority: VERY HIGH
  • Desirable skills: good knowledge of 3D solid and surface modeling, medium to advanced C programming. Familiarity with the SAT/SAB formats is a plus.
After working with 3D formats for a few years now, maybe I think this is out of the scope of libredwg. --Pitanga 15
57, 27 February 2014 (EST)

DWG write support

LibreDWG currently supports DWG versions R13, R14, R2000 and R2004 (R2007 is on the way) but only for reading. Some write operations for entities and objects are already written, and there is a very basic write framework. Although, headers and the whole file structure are not written. Write support is almost evil, but still needed, since there is not a well stablished free CAD format, and we don't want people to leave free CAD applications because they can't send their work back to DWG-only-CAD users.

  • Level: HARD
  • Priority: LOW/MEDIUM
  • Desirable skills: good knowledge of C programming, familiarity with 3D modeling or CAD applications, goof knowledge of the inners of the DWG format as described by the OpenDWG specification.

Currently, there is only partial 2000 support written by a GSoC'10 student, which means we can accept new proposals for this idea (specially if they include both finishing 2000 and writing 2004 support).

Other ideas

  • Make python bindings more python-ish, rewrite to use the new LibreDWG API (to be merged)
  • Write more bindings. Maybe Java? There are many GIS (Geographic Information System) programs written in Java
  • Complete R2004 read support
  • Complete R2007 read support
  • R2010 read support

Past proposals


Currently, LibreDWG users (CAD programs) have to browse through the low level DWG struct blindly. There are some helper functions, but they aren't enough to make the library usable for programmers. The student taking this project will potentially implement some sort of dwg2foo converter in order to understand the API needs. That done, they will write a reasonably simple and easy to use C API. There is currently a python binding, although all methods are C-like methods. The student should make python (ie, object oriented)-like methods. A C++ object-oriented API would be desirable as well.

  • Level: MEDIUM
  • Priority: MEDIUM
  • Desirable skills: good knowledge of C, C++, Python and binding-generation tools (such as SWIG), some knowledge of vectorial computer graphics or CAD formats.

Decode refactoring

Refactor decode functions, split the decode.c file into decode_r2000, decode_r2004, decode_r2007. Finish r2007 decode support, write on-demand decoding routine: first header, then layers, paper_space, model_space.

  • Level: HIGH
  • Priority: MEDIUM/HIGH
  • Desirable skills: trough knowledge of LibreDWG internals