Add script to update instruments.xml based on online spreadsheet

.gitignore

@@ -29,6 +29,7 @@ MuseScorePortable
 *.qm
 share/manual/en
 share/manual/de
+/share/instruments/tsv
 Doc
 DocLib
 build.qt5

share/instruments/generate_tsv.py

@@ -0,0 +1,264 @@
+#!/usr/bin/env python3
+
+# If you get Unicode errors on Windows, try setting the environment variable
+# PYTHONIOENCODING=utf-8. More info at https://stackoverflow.com/a/12834315
+
+import csv
+import os
+import sys
+import xml.etree.ElementTree as ET
+
+null='[null]' # value to use in TSV when data is not specified in XML
+
+def first_leaf_node(indexes):
+    first_value = next(iter(indexes.values()))
+    if isinstance(first_value, dict):
+        return first_leaf_node(first_value)
+    return indexes
+
+def write_indexes(writer, indexes):
+    assert isinstance(indexes, dict)
+    first_value = next(iter(indexes.values()))
+    if isinstance(first_value, dict):
+        # assume all values are of type dict
+        for subindexes in indexes.values():
+            write_indexes(writer, subindexes)
+    else:
+        # assume all values are of type str
+        writer.writerow(indexes.values())
+
+def create_tsv(fname, indexes):
+    with open(fname, 'w', newline='\n', encoding='utf-8') as f:
+        writer = csv.writer(f, delimiter='\t') # TSV
+        writer.writerow(first_leaf_node(indexes).keys()) # header row
+        write_indexes(writer, indexes)
+
+def load_translations(file):
+    translations = {}
+    for message in ET.parse(file).getroot().findall('context/message'):
+        translations[message.find('source').text] = message.find('translation').text
+    return translations
+
+def find_text(element, tagName, default='[null]'):
+    tag = element.find(tagName)
+    return default if tag is None else tag.text
+
+os.chdir(sys.path[0]) # make all paths relative to this script's directory
+os.makedirs('tsv', exist_ok=True)
+
+# InstrumentsXML
+root = ET.parse('instruments.xml').getroot()
+
+# Roman's InstrumentsXML
+roman = {}
+if os.path.exists('tsv/instruments_roman.xml'):
+    for instrument in ET.parse('tsv/instruments_roman.xml').getroot().findall('InstrumentGroup/Instrument'):
+        instrument_id = instrument.attrib['id']
+        roman[instrument_id] = {
+            'id':           instrument_id,
+            'trackName':    find_text(instrument, 'trackName'),
+            'longName':     find_text(instrument, 'longName'),
+            'shortName':    find_text(instrument, 'shortName'),
+    'transpositionName':    find_text(instrument, 'transpositionName'),
+            'description':  find_text(instrument, 'description'),
+        }
+
+# Translations
+italian = load_translations('../locale/instruments_it.ts')
+german = load_translations('../locale/instruments_de.ts')
+
+genres = {}
+for genre in root.findall('Genre'):
+    id = genre.attrib['id']
+    genres[id] = {
+        'id': id,
+        'name': genre.find('name').text,
+    }
+create_tsv('tsv/genres.tsv', genres)
+
+families = {}
+for family in root.findall('Family'):
+    id = family.attrib['id']
+    families[id] = {
+        'id': id,
+        'name': family.find('name').text,
+    }
+create_tsv('tsv/families.tsv', families)
+
+articulation_names = {}
+for articulation in root.findall('Articulation'):
+    name = articulation.get('name', default='[null]')
+    articulation_names[name] = {
+        'name':     name,
+        'velocity': articulation.find('velocity').text,
+        'gateTime': articulation.find('gateTime').text,
+    }
+create_tsv('tsv/articulation_names.tsv', articulation_names)
+
+groups = {}
+instruments = {}
+channels = {}
+articulations = {}
+drumsets = {}
+for group in root.findall('InstrumentGroup'):
+    group_id = group.attrib['id']
+    groups[group_id] = {
+        'id':   group_id,
+        'name': group.find('name').text,
+    }
+    for instrument in group.findall('Instrument'):
+        instrument_id = instrument.attrib['id']
+
+        trackName = find_text(instrument, 'trackName')
+        longName = find_text(instrument, 'longName')
+        shortName = find_text(instrument, 'shortName')
+
+        if instrument_id in roman:
+            instrument_roman = roman.pop(instrument_id)
+            trackName_roman = instrument_roman['trackName']
+            longName_roman = instrument_roman['longName']
+            shortName_roman = instrument_roman['shortName']
+            transpositionName_roman = instrument_roman['transpositionName']
+            description_roman = instrument_roman['description']
+        else:
+            longName_roman = null
+            shortName_roman = null
+            transpositionName_roman = null
+            description_roman = null
+
+        stringdata = instrument.find('StringData')
+        strings_open = []
+        strings_fretted = []
+        if stringdata is None:
+            frets = null
+        else:
+            frets = stringdata.find('frets').text
+            for string in stringdata.findall('string'):
+                if string.get('open') == '1':
+                    strings_open.append(string.text)
+                else:
+                    strings_fretted.append(string.text)
+
+        stafftype = instrument.find('stafftype')
+        if stafftype is None:
+            stafftype_name = null
+            staffTypePreset = null
+        else:
+            stafftype_name = stafftype.text
+            staffTypePreset = stafftype.get('staffTypePreset', default=null)
+
+        try: minA, maxA = instrument.find('aPitchRange').text.split('-')
+        except AttributeError: minA, maxA = (null, null)
+
+        try: minP, maxP = instrument.find('pPitchRange').text.split('-')
+        except AttributeError: minP, maxP = (null, null)
+
+        chans = {}
+        for channel in instrument.findall('Channel'):
+            channel_name = channel.get('name', default=null)
+            try: bank_msb = int(channel.find('controller[@ctrl="0"]').get('value'))
+            except AttributeError: bank_msb = 0
+            try: bank_lsb = int(channel.find('controller[@ctrl="32"]').get('value'))
+            except AttributeError: bank_lsb = 0
+            bank = (128 * bank_msb) + bank_lsb
+            chans[channel_name] = {
+                'instrument':   instrument_id,
+                'group':        group_id,
+                'family':       find_text(instrument, 'family'),
+                'description':  find_text(instrument, 'description'),
+                'channel':      channel_name,
+                'Bank':         (128 * bank_msb) + bank_lsb,
+                'Prog':         channel.find('program').get('value'),
+            }
+        if chans:
+            channels[instrument_id] = chans
+
+        instruments[instrument_id] = {
+            'id':           instrument_id,
+            'group':        group_id,
+            'family':       find_text(instrument, 'family'),
+        'longName_roman':   longName_roman,
+            'longName':     longName,
+            'longName_it':  italian[longName] if longName in italian else null,
+            'longName_de':  german[longName] if longName in german else null,
+        'shortName_roman':  shortName_roman,
+            'shortName':    shortName,
+            'shortName_it': italian[shortName] if shortName in italian else null,
+            'shortName_de': german[shortName] if shortName in german else null,
+        'trackName_roman':  trackName_roman,
+            'trackName':    trackName,
+            'trackName_it': italian[trackName] if trackName in italian else null,
+            'trackName_de': german[trackName] if trackName in german else null,
+    'transpositionName':    transpositionName_roman,
+            'init':         find_text(instrument, 'init'),
+    'description_roman':    description_roman,
+            'description':  find_text(instrument, 'description'),
+            'musicXMLid':   find_text(instrument, 'musicXMLid'),
+            'frets':        frets,
+        'stringsOpen':      ';'.join(strings_open),
+        'stringsFretted':   ';'.join(strings_fretted),
+            'staves':       find_text(instrument, 'staves'),
+            'clefs':        ';'.join(clef.text for clef in instrument.findall('clef')),
+            'clefsT':       ';'.join(clef.text for clef in instrument.findall('transposingClef')),
+            'clefsC':       ';'.join(clef.text for clef in instrument.findall('concertClef')),
+            'stafftype':    stafftype_name,
+        'staffTypePreset':  staffTypePreset,
+            'bracket':      find_text(instrument, 'bracket'),
+            'bracketSpan':  find_text(instrument, 'bracketSpan'),
+            'barlineSpans': ';'.join(span.text for span in instrument.findall('barlineSpan')),
+            'drumset':      find_text(instrument, 'drumset'),
+            'minA':         minA,
+            'maxA':         maxA,
+            'minP':         minP,
+            'maxP':         maxP,
+            'transpDia':    find_text(instrument, 'transposeDiatonic'),
+            'transpChr':    find_text(instrument, 'transposeChromatic'),
+            'SND':          find_text(instrument, 'singleNoteDynamics'),
+        'articulations':    ';'.join(articulation.get('name', default=null) for articulation in instrument.findall('Articulation')),
+            'velocities':   ';'.join(find_text(articulation, 'velocity') for articulation in instrument.findall('Articulation')),
+            'gateTimes':    ';'.join(find_text(articulation, 'gateTime') for articulation in instrument.findall('Articulation')),
+            'genres':       ';'.join(genre.text for genre in instrument.findall('genre')),
+        }
+
+        artics = {}
+        for articulation in instrument.findall('Articulation'):
+            name = articulation.get('name', default=null)
+            artics[name] = {
+                'instrument':   instrument_id,
+                'group':        group_id,
+                'family':       find_text(instrument, 'family'),
+                'description':  find_text(instrument, 'description'),
+                'articulation': name,
+                'velocity':     find_text(articulation, 'velocity'),
+                'gateTime':     find_text(articulation, 'gateTime'),
+            }
+        if artics:
+            articulations[instrument_id] = artics
+
+        drums = {}
+        for drum in instrument.findall('Drum'):
+            pitch = drum.get('pitch', default=null)
+            drums[pitch] = {
+                'instrument':   instrument_id,
+                'group':        group_id,
+                'family':       find_text(instrument, 'family'),
+                'description':  find_text(instrument, 'description'),
+                'pitch':        pitch,
+                'drum':         find_text(drum, 'name'),
+                'head':         find_text(drum, 'head'),
+                'line':         find_text(drum, 'line'),
+                'voice':        find_text(drum, 'voice'),
+                'stem':         find_text(drum, 'stem'),
+                'shortcut':     find_text(drum, 'shortcut'),
+            }
+        if drums:
+            drumsets[instrument_id] = drums
+
+create_tsv('tsv/groups.tsv', groups)
+create_tsv('tsv/instruments.tsv', instruments)
+create_tsv('tsv/articulations.tsv', articulations)
+create_tsv('tsv/channels.tsv', channels)
+create_tsv('tsv/drumsets.tsv', drumsets)
+
+for inst in roman.values():
+    print(inst) # these had IDs changed