[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Lilypond-auto] Issue 2507 in lilypond: Stop the entanglement of con
From: |
lilypond |
Subject: |
Re: [Lilypond-auto] Issue 2507 in lilypond: Stop the entanglement of context properties and grob property internals |
Date: |
Fri, 04 May 2012 16:33:36 +0000 |
Comment #2 on issue 2507 by address@hidden: Stop the entanglement of
context properties and grob property internals
http://code.google.com/p/lilypond/issues/detail?id=2507
I'm not familiar with GOOPS, but a quick look seems to support your
conclusion -- GOOPS would be a good choice for properties.
The invasive nature of your work on this is, IMO, due not to problems
associated with the change, but with a history of bad programming practices
in LilyPond. We have not properly used data abstractions, but instead have
littered our code with calls that take advantage of our knowledge of
property alists.
Proper practice would have been to define property structures, and then do
all of our accessing of properties through defined API calls that work with
the structures. Had we done that, the change from alist to GOOPS would
have been transparent and seamless.
I know you've observed this multiple places in the LilyPond code base, so
this is nothing new for you. I just wanted to get it out in the open to
make sure we had a common understanding.
Given the pervasive nature of this change, it would seem that this should
not be applied to 2.15, but instead to 2.17 (as we currently have a
release candidate again).
I think this is great work, and I'm very glad to see it happening.