[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [libredwg] Someone would like to join "again"

From: Felipe Castro
Subject: Re: [libredwg] Someone would like to join "again"
Date: Fri, 2 Aug 2013 09:04:36 -0300

Hi, "xará", :-)

I'm feeling some difficulties to translate all those macro code in my mind. I'm wondering if it would be possible to implement this idea with ordinary functions. But I promise to do my best to get in the mood to understand the new approach, as you had done to understand my esperanto-code before, ok? ;-) 

2013/8/1 Felipe Sanches <address@hidden>
Good to hear from you again, Felipe!

I'd like to clarify the reason for the macros and spec file. I've introduced these as a major refactoring to the codebase, which I think was my major contribution to LibreDWG.

The idea is that the file format is declared only once in the spec file, so that we avoid the duplicate-effort of implementing both the read and the write routines for the same datastructures. So, the overal file format is described only once, and read-specific or write-specific macros will convert the spec into C code for eighter reading or writing the file format.

happy hacking,
Felipe Sanches

On Thu, Aug 1, 2013 at 2:56 PM, Felipe Castro <address@hidden> wrote:

It's nice to see that people keep working on this stuff. I'm having a bit of time to dive into libredwg in these next few months, and I'd like to offer some help in order to reach a first release, 0.4 version for example.

When I started that libDWG project alone, I didn't hope to attract much interest with other collaborators, mainly because of the Esperanto stuff in there. For me it is really very pleasant to work with that language, but I know it does not have so much appeal for other programmers in general. And digging in the last git code of LibreDWG, I see some "esperantissues" lying around... So I may help to finish the code translation to English, as a starting point. :-)

What would be the main goal for version 0.4? Isn't it to get a working and useful reading library for the early versions (R13, R14, R2000, R2004)? If it is working, why not to release just like that, an alpha version?

Next step, try to work out the writing capability?

Next one, work on further versions (R2007, and so on)?

I'm trying to understand the changes in the "refactoring" branch. I think it's mainly a matter of splitting the code in some smaller files. More include files to worry about, maybe it's worthwhile, if it brings more organization. The doxygen stuff is cool also.

The API is a really huge work, and it could start with very simple things, because there are functions there that maybe would never be used in a simple CAD environment. But it doesn't hurt anyway to have access on every little peace of Autodesk creativity (handles, for example), that's the final goal if you want total control of the file format. Would it be useful to follow some DXF references? See:

I don't know what would be the best way to access the API, but I think including "api.h" will not be the final result for the library user, rigth? Will it be included from the "dwg.h" header?

And the r2007 stuff, I didn't touch it, too much for my brain now...

Another little suggestion, in general: what about turning back to the "near side of the moon..."? Some of those MACROS... hum, I don't know, just annoying, and those .spec files remember me RedHat... ;-)

Hoping not to hurt hearts with some of my considerations, sincerely yours,
Felipe Castro.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]