On 30-03-2011 11:47, Fernando Lucchesi wrote: > Hi libredwg developers, > > I'm Fernando Lucchesi, and I am at my third year studying Computer Engineering > at Unicamp, Brazil. I had the opportunity to talk to Rodrigo and Felipe at > FISL > last year =) > > (I've been lurking at the mailing list since then, but didn't introduce > myself) > > I'm also applying to Inkscape, but as they say it's not wise to put all eggs > in > one basket, here I am. > > The idea of this lib is very driving because it's about freedom of data, about > using the tool of your choice to edit your own files. I suffered a bit > fighting > to open some old cdr files on inkscape, so I understand where this comes from. > > Because of that, the idea that I would like to try is the DWG write support, > so > that the freedom goes both ways (and people are able to work with others while > using free tools, since there isn't a really open standard yet AFAIK).
You should be aware that Anderson Cardoso has written partial 2000 write support last year. For this year, we are hoping to see 2000 support completed (it is almost done) and 2004. Till Heuschmann has written great 2004 read code, and it would probably be easy to port. > I think that C wouldn't be a big problem, since I've been forced to learn it > at > the university, after 3 semesters of rather intense coding =) (the 4th was > with > other languages). I haven't made any program with binary output, though. > The biggest problem I see is to grok the code and the GNU coding style. Use emacs and it will ident in GNU style per default. Other IDE's and editors such as Eclipse have automatic GNU-style identing modes which are trivial to configure. > I have a little knowledge of Autocad, since I had a semester of classes (at my > former university). But it was back in 2007, so I wouldn't go as far as to say > that I remember a lot of it. So have us =]. Although, there are several people on this list that have good understanding of Autocad and could help you. > The problem is that I know nothing about the internals of DWG(or OpenDWG), so > I > would need to study it from the very beggining. You'll probably never know anything about OpenDWG internals as it is proprietary and closed source - unless you pay a US$150k founder-membership fee! For LibreDWG, nothing that a 2 days reading can't help. In the end, it is all C code that speaks for itself. Feel free to ask anything about some weird code (there are lots of them). > Is it worth a shot? Sure, specially because we have no project proposals so far =]. But keep in mind that time is short. You'd better have a project proposal in Melange (and posted to this list as well) before Monday. I am available for mentoring one project this year. If I may suggest you a different project proposal, I think that the automated test suite[0] would bring a greater benefit to the project at this moment, and would probably be easier to write (if you have a good idea) than write support. Guruprasad Rane has written some code, but it hasn't been commited yet. You could start by improving it. [0] http://libreplanet.org/wiki/LibreDWG/SummerOfCode#Automated_test_suite -- Rodrigo Rodrigues da Silva http://polignu.org http://garoa.net.br http://gnu.org
0xE67BD84E.asc
Description: application/pgp-keys
signature.asc
Description: OpenPGP digital signature
