Re: [Lilypond-auto] Issue 4022 in lilypond: Patch: Allow specifying different alignment for grob and its parent

[lilypond]

Comment #34 on issue 4022 by address@hidden: Patch: Allow specifying different  
alignment for grob and its parent
http://code.google.com/p/lilypond/issues/detail?id=4022

Ok, at the danger of looking like a complete idiot, after sleeping over it  
I think I'd rather backpaddle on the use of a callback for  
parent-alignment-X/Y.

Here is what made me waver:  The offset-X callback will in this manner  
refer to self-alignment-X and parent-alignment-X, and the  
parent-alignment-X callback will again look up self-alignment-X.  This  
sounds like a few too many back-and-forth lookups.

So what else would I propose?  #'() is basically the "not set at all" value  
for properties.  Since parent-alignment-X is not useful without referring  
to self-alignment-X, the #'() setting for "just take the self-alignment-X  
value instead" is sort of natural, and indeed ly:grob-property has an  
optional default value argument that is taken when #'() would otherwise be  
the result.

So the changed proposal for parent-alignment-X/Y would be:
#'() => take default from self-alignment-X
##f => take coordinate 0 for aligning
Number => -1 means left/lower edge, 1 right/upper, anything else  
interpolates

It would be pretty much the same for self-alignment-X/Y except that there  
#'() and ##f probably would be interpreted identically since we don't have  
anything else to fall back to.

So with this kind of definition, we'd probably have  
parent-self-alignment-X/Y have a default of ##f in order to get the old  
behavior, and set it to #'() for lyrics (if I understood correctly), again  
getting sort of the old behavior.

And then we have the toolset to change behavior individually without having  
to meddle with several different callbacks.

I think that is somewhat nicer and easier to understand and use than the  
current behavior of the patch using inherit-from-property or what I called  
it.

Again sorry for causing unnecessary work (well, assuming that people agree  
that the rechanged proposal is indeed better than my first proposal using a  
callback).

--
You received this message because this project is configured to send all  
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings