Incorrect Key Signature Parsing in MEI Files

[egorpol]

Description of the bug:
When parsing MEI files using music21.converter.parse(), key signatures are not being correctly interpreted or applied to the parsed pitches.

Steps to reproduce:

1. Load an MEI file using music21.converter.parse().
2. Extract notes and key signatures using the provided script.
3. Print or inspect the key signatures and note transpositions.

Expected behavior:
Key signatures should be correctly detected and applied, altering the pitch representation according to the key.

Actual behavior:
Key signatures seem to be ignored or incorrectly parsed, resulting in all notes being interpreted as if they were in the key of C, without any transposition applied. For example, pitches F and C are shown without the expected sharps.

Environment:

music21 version: 9.1.0
Python version: 3.12
Operating System: Windows 11

Additional context:
This issue appears to specifically affect MEI file parsing, as similar operations with MusicXML files do not exhibit this behavior.

Code example
Using an MEI file from official score samples: Bach-JS_Ein_feste_Burg.mei. For comparison, the MusicXML file from the Chorale-Corpus featuring the same chorale is also provided as a reference.

As demonstrated in the output DataFrame, the pitches F and C do not include the sharp that's defined in the key signature.

from music21 import converter, note, chord, pitch as pitch_module
import pandas as pd
import matplotlib.pyplot as plt
import requests
import tempfile
import os

def get_file_path(file_source):
    if file_source.startswith(('http://', 'https://')):
        try:
            with requests.get(file_source) as response:
                response.raise_for_status()
                _, file_extension = os.path.splitext(file_source)
                # Ensuring the temporary file is not deleted and is properly closed
                temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=file_extension, mode='wb+')
                temp_file.write(response.content)
                temp_file.flush()  # Ensure all data is written
                temp_file.close()  # Close the file to ensure it's accessible later
                return temp_file.name
        except requests.RequestException as e:
            raise ValueError(f"Error downloading the file: {e}")
    else:
        if os.path.exists(file_source):
            return file_source
        else:
            raise FileNotFoundError("Local file does not exist")

def extract_voice_data(score, measure_range):
    start_measure, end_measure = measure_range
    voice_data = []
    notes_and_chords = score.flatten().notesAndRests.stream()

    for element in notes_and_chords:
        if isinstance(element, (note.Note, chord.Chord)):
            measure_num = element.measureNumber
            if start_measure <= measure_num <= end_measure:
                position_within_measure = element.offset
                duration = element.duration.quarterLength
                pitches = [str(element.pitch)] if isinstance(element, note.Note) else [str(p) for p in element.pitches]
                voice_data.extend([(measure_num, position_within_measure, duration, pitch) for pitch in pitches])

    return voice_data

measure_range = (0, 1)

file_source = 'https://raw.githubusercontent.com/music-encoding/sample-encodings/main/MEI_5.0/Music/Complete_examples/Bach-JS_Ein_feste_Burg.mei'

#MusicXML version for reference
#file_source = 'https://github.com/MarkGotham/Chorale-Corpus/raw/main/Bach,_Johann_Sebastian/Chorales/020/short_score.mxl'

file_path = get_file_path(file_source)
score = converter.parse(file_path)
voice_data = extract_voice_data(score, measure_range)
df = pd.DataFrame(voice_data, columns=['Measure', 'Local Onset', 'Duration', 'Pitch'])
display(df)

Output DataFrame:

Measure	Local Onset	Duration	Pitch
0	0.0	1.0	D5
0	0.0	1.0	A4
0	0.0	1.0	F4
0	0.0	0.5	D4
0	0.5	0.5	C4
1	1.0	1.0	D5
1	1.0	1.0	D4
1	1.0	1.0	F4
1	1.0	1.0	B3
1	2.0	1.0	D5
1	2.0	0.5	D4
1	2.0	0.5	B3
1	2.0	0.5	A3
1	2.5	0.5	E4
1	2.5	0.5	C4
1	2.5	0.5	G3
1	3.0	0.5	A4
1	3.0	1.0	F4
1	3.0	1.0	D4
1	3.0	1.0	F3
1	3.5	0.5	B4
1	4.0	1.0	C5
1	4.0	1.0	G4
1	4.0	1.0	E4
1	4.0	1.0	E3