Fix #22671: Add support for nested tuplets in MusicXML

[pacebes]

Resolves: #22671

• This PR is able to deal with nested tuplets when importing nested tuplets in MusicXML
• It has been tested with the following Score: musescore-22671-test.zip

• I signed the CLA
• The title of the PR describes the problem it addresses
• Each commit's message describes its purpose and effects, and references the issue it resolves
• If changes are extensive, there is a sequence of easily reviewable commits
• The code in the PR follows the coding rules
• There are no unnecessary changes
• The code compiles and runs on my machine, preferably after each commit individually
• I created a unit test or vtest to verify the changes I made (if applicable)

[pacebes]

Hi @cbjeukendrup

I'm a bit stuck with a scenario and I would appreciate some help from someone with more experience in these matters.
I have a version (the last PR) that works with several nested tuplets scenarios included in previously xml file and most of the regression tests.

I have a problem with the second measure of testTuplets5.xml file (red and blue notes):

XML:

    <measure number="2">
      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>1</duration>
        <voice>1</voice>
        <type>quarter</type>
        <dot/>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        <notations>
          <tuplet type="start" bracket="no"/>
          </notations>
        </note>
      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>1</duration>
        <voice>1</voice>
        <type>quarter</type>
        <dot/>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        <notations>
          <tuplet type="start" bracket="no"/>
          </notations>
        </note>

As I see it the first note (red one) creates a new tuplet :

      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>1</duration>
        <voice>1</voice>
        <type>quarter</type>
        <dot/>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        <notations>
          <tuplet type="start" bracket="no"/>
          </notations>
        </note>

... but then the second (blue one) note begins

      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>1</duration>
        <voice>1</voice>
        <type>quarter</type>
        <dot/>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        <notations>
          <tuplet type="start" bracket="no"/>
          </notations>
        </note>

The second note (blue one) declares that a new tuplet must be created. This is the scenario I have problems with:

• Prior to this PR (as no nested tuplets are supported) the solution is to stop the previous tuplet and create a new one
• Now (with this PR) we try to support nested tuplets. Therefore the code see a new "tuplet start" and decides to create a new nested tuplet because there is no way (at least I haven't been able to find one) to determine that the previous tuplet (red one) is full now.

I would appreciate if some one could provide some support because I don't know how to deal with it.

[pacebes]

Sorry to bother @cbjeukendrup

testTuplets5.xml is only test failing now. I don't' know if you are yet to analyze the prior comment or you think I am not right about the comment and there is a way to deal with this regression test

[cbjeukendrup]

Sorry, I will attempt to look into this within the next days. Or maybe @miiizen can help

[miiizen]

I'm fairly sure we have enough information to tell that this tuplet is full. The note has a duration of 1. Because divisions is also 1, this means the note's length is 1 quarter note. However, the type and dot tags tell us to display a dotted quarter, giving us a ratio of 3:2.

We can then compare the ratio between these two values to the time-modification tag. This is also 3:2, suggesting we have a full tuplet.

Let me know if that helps!

[pacebes]

I'm fairly sure we have enough information to tell that this tuplet is full. The note has a duration of 1. Because divisions is also 1, this means the note's length is 1 quarter note. However, the type and dot tags tell us to display a dotted quarter, giving us a ratio of 3:2.

We can then compare the ratio between these two values to the time-modification tag. This is also 3:2, suggesting we have a full tuplet.

Let me know if that helps!

Thanks. I'll review it in the coming days

[pacebes]

I'm fairly sure we have enough information to tell that this tuplet is full. The note has a duration of 1. Because divisions is also 1, this means the note's length is 1 quarter note. However, the type and dot tags tell us to display a dotted quarter, giving us a ratio of 3:2.

We can then compare the ratio between these two values to the time-modification tag. This is also 3:2, suggesting we have a full tuplet.

Let me know if that helps!

Hi @miiizen. I would appreciate additional clarification to avoid further mistakes because I am missing something.
Previous example (where divisions = 1):

Tuplet (3/2):

• Duration-1 + Divisions-1 => Real duration= 1/4
• Type: Quarter (1/4) + dotted => 3/8
• Ratio: (3/8)/ (1/4) => (3:2)

So, ratio of 3:2 and time-modification tag 3:2 => complete. OK

So if we get some other examples such as testTuplets2.xml with

<divisions>3</divisions>

and a tuplet defined as:

      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>2</duration>
        <voice>1</voice>
        <type>quarter</type>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        </note>
      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>1</duration>
        <voice>1</voice>
        <type>eighth</type>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        </note>

First part of the tuplet (3/2):

• Duration-2 + Divisions-3 => Real duration= 1/6
• Type: Quarter (1/4)
• Ratio: (1/4)/ (1/6) => (3/2)

Second part of the tuplet (3/2):

• Duration-1 + Divisions-3 => Real duration= 1/12
• Type: Eighth (1/8)
• Ratio: (1/8/ (1/12) => (3/2)

I always the same ratio in the first part of the tuplet and it's not complete. I am missing something and I would appreciate additional help, please.

[pacebes]

I'm fairly sure we have enough information to tell that this tuplet is full. The note has a duration of 1. Because divisions is also 1, this means the note's length is 1 quarter note. However, the type and dot tags tell us to display a dotted quarter, giving us a ratio of 3:2.
We can then compare the ratio between these two values to the time-modification tag. This is also 3:2, suggesting we have a full tuplet.
Let me know if that helps!

Hi @miiizen. I would appreciate additional clarification to avoid further mistakes because I am missing something. Previous example (where divisions = 1):

Tuplet (3/2):

* Duration-1 + Divisions-1 => Real duration= 1/4

* Type: Quarter (1/4) + dotted => 3/8

* Ratio: (3/8)/ (1/4) => (3:2)

So, ratio of 3:2 and time-modification tag 3:2 => complete. OK

So if we get some other examples such as testTuplets2.xml with

<divisions>3</divisions>

and a tuplet defined as:

      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>2</duration>
        <voice>1</voice>
        <type>quarter</type>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        </note>
      <note>
        <pitch>
          <step>G</step>
          <octave>4</octave>
          </pitch>
        <duration>1</duration>
        <voice>1</voice>
        <type>eighth</type>
        <time-modification>
          <actual-notes>3</actual-notes>
          <normal-notes>2</normal-notes>
          </time-modification>
        <stem>up</stem>
        </note>

First part of the tuplet (3/2):

* Duration-2 + Divisions-3 => Real duration= 1/6

* Type: Quarter (1/4)

* Ratio: (1/4)/ (1/6) => (3/2)

Second part of the tuplet (3/2):

* Duration-1 + Divisions-3 => Real duration= 1/12

* Type: Eighth (1/8)

* Ratio: (1/8/ (1/12) => (3/2)

I always the same ratio in the first part of the tuplet and it's not complete. I am missing something and I would appreciate additional help, please.

What I try to explain here is that, INMO, every note within a triplet has the same ratio (between type and real duration) as the tuplet. Therefore it can't be used to determine when a Triplet ends (At least I haven't been able to). I would appreciate any additional comments or explanations, please