Re: Improve elbowed-hairpin (issue 324800043 by address@hidden)

[thomasmorley65]

On 2017/04/24 14:42:18, david.nalesnik wrote:
> LGTM.

> I've pointed out two minor issues, but I don't believe they should
hold up the
> review process.

> By the way, applying

> \override Hairpin.style = #'dashed-line to the regtest
> input/regression/ferneyhough-hairpins.ly looks great!

> https://codereview.appspot.com/324800043/diff/1/scm/output-lib.scm
> File scm/output-lib.scm (right):


https://codereview.appspot.com/324800043/diff/1/scm/output-lib.scm#newcode1171
> scm/output-lib.scm:1171: (if (not (pair? (cdr pts)))
> Note that this will fail if pts is '().  Perhaps you should check the
length of
> pts in elbowed-hairpin -- there must be at least two pairs in the pts
alist for
> a viable line.

Yep.

I decided to let the condition for the inner 'connected-points' in
place, but to
emit a meaningful warning if it is called in 'make-connected-line' with
less than
two points. In this case an empty-stencil is returned.

Seems cleaner to do it early than later in 'elbowed-hairpin'.


https://codereview.appspot.com/324800043/diff/1/scm/output-lib.scm#newcode1172
> scm/output-lib.scm:1172: (apply ly:stencil-add empty-stencil ls)
> Why not use reduce here.

Done.

This warrants a new patchy-run because new functionality is added.



https://codereview.appspot.com/324800043/