Hacktoberfest 2026：维护者为十月标记出来的 issue，仍然开放、适合新手。 浏览 Hacktoberfest issue

MusicXML import ignores jumps written as `<sound>` attributes

未关闭
  • 哪里不对？
#2,079 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
28/100
Issue 类型
缺陷
描述清晰度
基本清楚
活跃度
活跃
技术栈
python
领域
data

调研方向

Start with SoundTagMixin.setSound in musicxml/xmlSoundParser.py, which handles only tempo and has TODOs for dacapo, dalsegno, tocoda, fine, segno and coda, then check how the <words> repeat expressions are created in the MusicXML importer. Run the reproduction script in the issue and confirm the expanded note list includes the second D E pass. Done means a <sound dalsegno> jump is honoured without matching text, and the design question of how multiple <sound> tags map to one jump expression is settled first.

由索引模型根据 Issue 内容生成。

描述

music21 version

10.5.0

Problem summary

The MusicXML importer only creates repeat expressions (Segno, Coda, DalSegnoAlCoda, ...) from the text of <words> directions. It ignores the dacapo, dalsegno, tocoda, fine, segno and coda attributes of <sound>, which are how MusicXML encodes jumps for playback. So when the <words> text isn't one of the recognized phrases, or there's no text, expandRepeats() skips the jump even though the <sound> says exactly what it is.

Steps to reproduce

from music21 import converter

def measure(number, step, before="", after=""):
    return (f'<measure number="{number}">{before}<note><pitch><step>{step}</step>'
            f'<octave>4</octave></pitch><duration>4</duration><type>whole</type></note>{after}</measure>')

xml = f'''<score-partwise version="4.0">
  <part-list><score-part id="P1"><part-name>Music</part-name></score-part></part-list>
  <part id="P1">
    {measure(1, "C", "<attributes><divisions>1</divisions></attributes>")}
    {measure(2, "D", '<direction><direction-type><segno/></direction-type><sound segno="segno"/></direction>')}
    {measure(3, "E", after='<direction><direction-type><words>To Coda</words></direction-type><sound tocoda="coda"/></direction>')}
    {measure(4, "F", after='<direction><direction-type><words>Repeat from sign, then go to the coda</words></direction-type><sound dalsegno="segno"/></direction>')}
    {measure(5, "G", '<direction><direction-type><coda/></direction-type><sound coda="coda"/></direction>')}
  </part>
</score-partwise>'''

expanded = converter.parse(xml, format='musicxml').expandRepeats()
print([n.name for n in expanded.flatten().notes])

Expected vs. actual behavior

Expected ['C', 'D', 'E', 'F', 'D', 'E', 'G'], from the dalsegno attribute. Actual:

['C', 'D', 'E', 'F', 'G']

With <words>D.S. al Coda</words> the jump is taken, so this only affects jumps whose text isn't a recognized phrase.

More information

SoundTagMixin.setSound in musicxml/xmlSoundParser.py only handles tempo, and has TODOs for coda, dacapo, dalsegno, fine, segno and tocoda.

A suggestion for how the two sources could combine: if only one of <words> and <sound> gives a jump, use it; if both do and they disagree, prefer <sound>, since it's the playback encoding, and maybe warn.

One complication: music21 represents a jump as a single expression, like DalSegnoAlCoda, but MusicXML spreads it over several <sound>s: dalsegno at the jump, tocoda at the "To Coda", and fine at the Fine. So a <sound dalsegno> on its own only says "D.S.", and telling whether it's "al Coda" or "al Fine" means looking for tocoda or fine elsewhere in the part. For the same reason, <words>D.S. al Coda</words> with <sound dalsegno="segno"/> isn't a disagreement, just a more specific description.

主要语言
Python
星标
2.6k
派生
459
平均合并
1 天 7 小时
30 天内合并 PR
40

环境准备

在 Codespaces 中打开

在浏览器里用你自己的 GitHub 账号启动这个项目的开发容器。

从这里开始

  1. 先读完整个 Issue，再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库，在一个分支上完成修改。
  4. 提交 Pull Request，并在描述里引用这个 Issue 编号。

cuthbertLab/music21 的其他 Issue

查看 cuthbertLab/music21 的全部 Issue

相似的 Issue

更多 Python Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。