Skip to Content.
Sympa Menu

cc-devel - [cc-devel] More review of liblicense changes since 0.4

cc-devel AT lists.ibiblio.org

Subject: Developer discussion for Creative Commons technology and tools

List archive

Chronological Thread  
  • From: Asheesh Laroia <asheesh AT creativecommons.org>
  • To: cc-devel AT lists.sourceforge.net
  • Subject: [cc-devel] More review of liblicense changes since 0.4
  • Date: Wed, 9 Jan 2008 12:33:53 -0800 (PST)
  • Resent-date: Wed, 9 Jan 2008 12:55:19 -0800 (PST)
  • Resent-from: Asheesh Laroia <paulproteus AT acm.jhu.edu>
  • Resent-to: cc-devel AT lists.ibiblio.org

I'm reading through all the patches and documenting what happened to this code between 0.4 and 0.5 in a ChangeLog. (I have no intention of keeping a full GNU-style ChangeLog since we have svn logs for that.) It's been mostly a fun, rewarding, endearing, love-for-humanity-growing experience. That's true for all the committers - yay! Let me be clear that I single out pmiller not because his work has been bad.

CODE STYLE

pmiller has committed an .indent.pro file to the source tree that switches us to GNU style indenting, which is fine. Curiously, he sets --tab-size0 rather than giving tabs a width > 0. Why? If there's no good reason, I'll just remove this line and stick with the GNU style default for tab width.

Further, I'll reindent the code base at some point to fully GNU style.

Further, pmiller has suggested we stick to C89 constructs. I'm feeling generous today, so I'll consider that reasonable. If we do want to stick to it, we should modify our compile process so that with GCC we used -std=c89 or -ansi so that the codebase does not compile with non-C89 constructs.

API CHANGES

pmiller added:

char ** ll_license_get_publisher
char ** ll_license_get_creator

I suggest we remove these as per my earlier suggestion. It's *crucial* that we document the named constants you can pass into ll_get_attribute() to make sure people don't forget about all the features we have, however. ll_license_get(LL_PUBLISHER, license_uri) would be the new way to do a call like that.

I still suggest shortening the named constants as much as they can bear, and then removing all the functions that are just trivial wrappers to the named constants.

That's all for now. This is short; if you care, reply.

-- Asheesh.

--
Learning French is trivial: the word for horse is cheval, and everything else
follows in the same way.
-- Alan J. Perlis





Archive powered by MHonArc 2.6.24.

Top of Page