Re: Issue 5788: New French Beamimg Approach (issue 557500043 by address@hidden)

[torsten . haemmerle]

Proposed changes applied to my local branch (most of them), see
comments.

Ta,
Torsten


https://codereview.appspot.com/557500043/diff/551490044/Documentation/changes.tely
File Documentation/changes.tely (right):

https://codereview.appspot.com/557500043/diff/551490044/Documentation/changes.tely#newcode67
Documentation/changes.tely:67: @emph{exactly} behaves like standard
(i.e. default) beaming in every respect
On 2020/02/24 05:24:25, lemzwerg wrote:
> s/i.e./i.e.,/

Ah, then I suppose the changes.tely language is American English...
In Europe, i.e. (sic!) British English, there's usually no comma after
i.e. or e.g.
But I'll update this and insert a comma.
Done.

https://codereview.appspot.com/557500043/diff/551490044/Documentation/changes.tely#newcode68
Documentation/changes.tely:68: (beam positioning and placement of any
articulation/fingering/etc.).  The
On 2020/02/24 05:24:25, lemzwerg wrote:
> ... articulation, fingering, etc.

Done.

https://codereview.appspot.com/557500043/diff/551490044/Documentation/changes.tely#newcode69
Documentation/changes.tely:69: only remaining difference being inner
stems not passing through beams.
On 2020/02/24 05:24:25, lemzwerg wrote:
> s/being/are/
I've heard this a lot of times, but, OK, "are" is probably simpler.
Done.

https://codereview.appspot.com/557500043/diff/551490044/input/regression/beam-french.ly
File input/regression/beam-french.ly (right):

https://codereview.appspot.com/557500043/diff/551490044/input/regression/beam-french.ly#newcode23
input/regression/beam-french.ly:23: { d32[^1 e^2 f^3 g^4 a^5] } s64
On 2020/02/24 05:24:25, lemzwerg wrote:
> Please add one or more test cases for your 'french-correction'
property.

My 'french-correction' is an intrinsic part of the new French Beaming
implementation.  So I think the two existing regression files

* beam-french.ly and
* tuplet-number-french-kneed-beams.ly

already handle French beaming quite well and I've even added a few
problematic cases to beam-French.ly.  In the reg diffs (inserted by
James, cf. Issue 5788),  the most important changes can be clearly seen
in both files, even wit the existing examples.

Apart from that, probably hardly anybody will ever use French beaming,
and the most important regression test by far is all the non-French
cases that must not change by my intervention.

What specific French beaming examples are you missing?

https://codereview.appspot.com/557500043/diff/551490044/lily/include/beam.hh
File lily/include/beam.hh (right):

https://codereview.appspot.com/557500043/diff/551490044/lily/include/beam.hh#newcode36
lily/include/beam.hh:36: struct Beam_stem_length
On 2020/02/24 06:44:39, hanwenn wrote:
> Beam_stem_end ?

Agreed. Done.

https://codereview.appspot.com/557500043/