Render oblique ligatures with curve shapes #3962

lpugin · 2025-02-22T07:57:55Z

The PR render oblique ligatures with curved shapes.

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="https://music-encoding.org/schema/5.1/mei-all.rng" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="https://music-encoding.org/schema/5.1/mei-all.rng" type="application/xml" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<mei xmlns="http://www.music-encoding.org/ns/mei" meiversion="5.1">
   <meiHead>
      <fileDesc>
         <titleStmt>
            <title>Ligatures (second to octave)</title>
            <respStmt>
               <persName role="encoder">Laurent Pugin</persName>
            </respStmt>
         </titleStmt>
         <pubStmt>
            <date isodate="2025-02-21"/>
         </pubStmt>
         <seriesStmt>
            <title>Verovio test suite</title>
         </seriesStmt>
         <notesStmt>
            <annot />
         </notesStmt>
      </fileDesc>
      <encodingDesc>
         <appInfo>
            <application version="5.1.0" label="2">
               <name>Verovio</name>
            </application>
         </appInfo>
      </encodingDesc>
      <extMeta><![CDATA[ { "evenNoteSpacing": true }]]></extMeta>
   </meiHead>
   <music>
      <body>
         <mdiv>
            <score>
               <scoreDef>
                  <staffGrp>
                     <staffDef n="1" notationtype="mensural.black" lines="5" clef.shape="C" clef.line="1" />
                  </staffGrp>
               </scoreDef>
               <section>
                  <staff n="1">
                     <layer n="1">
                        <ligature>
                           <note dur="brevis" oct="4" pname="e" />
                           <note dur="brevis" oct="4" pname="c" />
                        </ligature>
                        <ligature>
                           <note dur="brevis" oct="4" pname="f" />
                           <note dur="brevis" oct="4" pname="c" />
                        </ligature>
                        <ligature>
                           <note dur="brevis" oct="4" pname="g" />
                           <note dur="brevis" oct="4" pname="c" />
                        </ligature>
                        <ligature>
                           <note dur="brevis" oct="4" pname="a" />
                           <note dur="brevis" oct="4" pname="c" />
                        </ligature>
                        <ligature>
                           <note dur="brevis" oct="4" pname="b" />
                           <note dur="brevis" oct="4" pname="c" />
                        </ligature>
                        <ligature>
                           <note dur="brevis" oct="5" pname="c" />
                           <note dur="brevis" oct="4" pname="c" />
                        </ligature>
                        <!-- c.o.p. -->
                        <ligature form="obliqua">
                           <note dur="semibrevis" oct="4" pname="c" />
                           <note dur="semibrevis" oct="4" pname="d" />
                        </ligature>
                        <ligature form="obliqua">
                           <note dur="semibrevis" oct="4" pname="c" />
                           <note dur="semibrevis" oct="4" pname="e" />
                        </ligature>
                        <ligature form="obliqua">
                           <note dur="semibrevis" oct="4" pname="c" />
                           <note dur="semibrevis" oct="4" pname="f" />
                        </ligature>
                        <ligature form="obliqua">
                           <note dur="semibrevis" oct="4" pname="c" />
                           <note dur="semibrevis" oct="4" pname="g" />
                        </ligature>
                        <ligature form="obliqua">
                           <note dur="semibrevis" oct="4" pname="c" />
                           <note dur="semibrevis" oct="4" pname="b" />
                        </ligature>
                        <ligature form="obliqua">
                           <note dur="semibrevis" oct="4" pname="c" />
                           <note dur="semibrevis" oct="5" pname="c" />
                        </ligature>
                     </layer>
                  </staff>
               </section>
            </score>
         </mdiv>
      </body>
   </music>
</mei>

Straight polygons can still be obtained with --ligature-straight

craigsapp · 2025-02-22T08:44:46Z

Black notation oblique ligatures are usually curved:

https://www.loc.gov/resource/gdcwdl.wdl_14175/?sp=23&st=image&r=-0.16,0.55,1.02,0.466,0

But white notation ones are usually not:

https://digi.vatlib.it/view/MSS_Capp.Sist.160

[Note the dot above the starting note of the oblique ligature's first note near the middle of the second line.]

Will white notation oblique ligatures remain straight? If not, it would be good to have them straight as a default. And in that case the option would be

--ligature [auto|curved|straight]

Or

--ligature-oblique [auto|curved|straight]

where auto would be the default, which in black notation would be curved, and in white notation it would be straight.

It might be useful to use this option style in case you add curves to white notation oblique ligatures in the future in any case.

(These examples are two random samples I was looking at today.)

rettinghaus · 2025-02-22T09:51:04Z

I agree with @craigsapp, the white mensural ligatures do look a bit weird after this change.

lpugin · 2025-02-22T10:56:01Z

I'll change the option as suggested by @craigsapp

craigsapp · 2025-02-22T11:08:13Z

Looking in Thomas Morley's book A Plain and Easie Introduction to Practicall Musicke (1771, second edition), I do see some (printed) curved oblique ligatures (first oblique ligature on two successive pages in the examples below):

https://vmirror.imslp.org/files/imglnks/usimg/8/81/IMSLP284437-PMLP09691-plaineasyintrodu00morl_1771.pdf#page=31

First edition (1597) is the same, usually straight, sometimes curved:

https://s9.imslp.org/files/imglnks/usimg/d/d4/IMSLP278915-PMLP09691-a_plaine_and_easie_introduction.pdf#page=8

But they are mostly straight.

rettinghaus · 2025-02-22T11:12:55Z

Maybe something for the Mensural Interest Group to see if there is a desire to encode curved ligatures individually …

lpugin · 2025-02-22T12:34:09Z

I don't think this is really an encoding question. Black notation is essentially handwritten, whereas white is equally handwritten and printed, so we see more of a mixed between curved and straight. As I said, I am happy to change it and have white ligatures straight by default.

craigsapp · 2025-02-22T15:46:58Z

I was thinking about individual ligature control as well, but as @lpugin notes, handwritten music will be variable without always having intention or meaning. The Morley example is curious since the curved/straight difference must be intentional. Maybe the music typesetter was working from a manuscript that had an exaggerated curvature to that particular ligature and they duplicated it to be safe rather than normalize it.

That is why the curved ligatures in the Morley print are interesting.

The white notation example I gave is straight but handwritten 😜

With lots of illuminations:

Clearly a luxury edition... It would be harder to keep the lines parallel if they are curved, so the cursive style of black notation probably fell out of favor when switching to white notation.

…ht` and `curved`

ahankinson · 2025-02-24T09:01:22Z

Is there an example of how the two options look now?

lpugin · 2025-02-24T09:15:26Z

I added one here (with option --ligature-oblique straight)

ahankinson · 2025-02-24T09:30:11Z

Am I supposed to be able to tell the difference between these two examples?

craigsapp · 2025-02-24T10:39:51Z

Yes: these two examples are the same but the curved one should be different.

lpugin · 2025-02-24T10:43:24Z

No, because:

currently 049 is still rendered straight
currently 050 does not do anything with --ligature-oblique straight

This is what you can see in the GH action for 049 (before and after):

050 is supposed to remain as before, and since the action is not detecting any difference, it is looking that it does remain the same

lpugin · 2025-02-24T10:44:44Z

(I agree this is all very confusing, sorry)

ahankinson · 2025-02-24T11:33:05Z

Ah, right. That won't change until this is merged.

Will do so.

lpugin added 7 commits February 21, 2025 14:53

Add method to draw bent parallelograms

ec3865a

Refactor ligature drawing code (WIP)

5eeb077

Draw bent parallelograms (WIP)

6ea8739

Draw bent ligatures

58e9696

Add option for --ligature-straight

aad7328

cleanup code

6f9fbe5

Adjust values

0375777

Change --ligatureStraight to ligatureOblique with auto, `straig…

ee1cfff

…ht` and `curved`

rettinghaus approved these changes Feb 23, 2025

View reviewed changes

lpugin requested a review from ahankinson February 24, 2025 08:20

ahankinson merged commit f094ea4 into develop Feb 24, 2025
9 checks passed

lpugin deleted the develop-mensural branch February 24, 2025 12:00

lpugin restored the develop-mensural branch February 24, 2025 12:00

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Render oblique ligatures with curve shapes #3962

Render oblique ligatures with curve shapes #3962

lpugin commented Feb 22, 2025

craigsapp commented Feb 22, 2025 •

edited

Loading

rettinghaus commented Feb 22, 2025

lpugin commented Feb 22, 2025

craigsapp commented Feb 22, 2025

rettinghaus commented Feb 22, 2025

lpugin commented Feb 22, 2025

craigsapp commented Feb 22, 2025 •

edited

Loading

ahankinson commented Feb 24, 2025

lpugin commented Feb 24, 2025

ahankinson commented Feb 24, 2025

craigsapp commented Feb 24, 2025

lpugin commented Feb 24, 2025 •

edited

Loading

lpugin commented Feb 24, 2025

ahankinson commented Feb 24, 2025

Render oblique ligatures with curve shapes #3962

Render oblique ligatures with curve shapes #3962

Conversation

lpugin commented Feb 22, 2025

craigsapp commented Feb 22, 2025 • edited Loading

rettinghaus commented Feb 22, 2025

lpugin commented Feb 22, 2025

craigsapp commented Feb 22, 2025

rettinghaus commented Feb 22, 2025

lpugin commented Feb 22, 2025

craigsapp commented Feb 22, 2025 • edited Loading

ahankinson commented Feb 24, 2025

lpugin commented Feb 24, 2025

ahankinson commented Feb 24, 2025

craigsapp commented Feb 24, 2025

lpugin commented Feb 24, 2025 • edited Loading

lpugin commented Feb 24, 2025

ahankinson commented Feb 24, 2025

craigsapp commented Feb 22, 2025 •

edited

Loading

craigsapp commented Feb 22, 2025 •

edited

Loading

lpugin commented Feb 24, 2025 •

edited

Loading