Re: moving slur attachment points globally

[David Kastrup]

Urs Liska <address@hidden> writes:

> Am 17.02.2016 um 13:32 schrieb David Kastrup:
>> Kieren MacMillan <address@hidden> writes:
>>
>>> Hi David,
>>>
>>> Thanks for the response.
>>>
>>>> it is hard to guess which cases you are trying to cover or not.
>>> Essentially all of them (this is for a stylesheet, which intends to
>>> mimic an existing publishing house style). Which is why including any
>>> arbitrary example(s) in my original post made — and still makes — no
>>> sense to me.
>>>
>>> But here goes:
>>>
>>>     \score { { c'4( e' g'2) } }
>>>
>>> As I said in the original post, I would like the slur to start at the
>>> “4:30” position of the c’, and end at the “7:30” position of the
>>> g’.
>> Why?  Without answering that question it is not clear whether any future
>> changes in the slur code might thwart the intention or not.
>
> Well, Kieren *did* explicitly state that he wants LilyPond to engrave
> according to the engraving rules of a given publishing house. Or, put
> more generally, for a given context he needs to change LilyPond's
> default appearance with regard to slur attachment.
>
>>
>>> Any help would be appreciated.
>> Shrug.  I don't see that my answer would not have applied here.
>>
>>     \score { { <c'(>4 e' <g')>2 } }
>>
>> looks like described.  Not optimally so, admittedly.  This code is
>> obviously slated for improvement, but unless we get to know the
>> _rationale_ for your desire, it's not possible to tell whether such
>> improvements will run counter to it.
>>
>
> I think the rationale is very clear, and also the desired approach:
> determine some properties that can be overridden so the slur decides to
> be attached differently than LilyPond's default decision.

LilyPond's slur attachment typesetting is a fixed algorithm.  It can be
tweaked by overriding the settings for Slur.details and
PhrasingSlur.details predefined in scm/layout-slur.scm.  I don't see
obvious contenders there for manipulating the default attachment angle.

-- 
David Kastrup